researchmap

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

Recent advancement in ability of a computing machine has developed an active trend for replacing an experiment with a simulation. In order to perform the analysis of a real structure pertained with extremely large fluctuations, both geometrically and materially, a large number of elements are required, and efficient analysis is indispensable. In contrast, it is often the case that an actual structure has a symmetrical shape in whole or in part, and its boundary value also have a symmetrical distribution. Therefore, there have been proposed a number of analytic techniques that take advantage of a symmetry of a structure. However, symmetry includes wide variations, such as plane symmetry, inverse symmetry, axial symmetry, helical symmetry, and short cake symmetry. Further, the generation method of a mirror image, the number of mirror images to be generated, and so on are different in each case. Therefore, a program corresponding to each case must be configured. Further, the presence of a problem of a mixture of different types of symmetry such as mixture of plane symmetry and axial symmetry increases the number of cases to be coped with, which causes serious problems in maintenance and expandability of a program. This paper has proposed a boundary element method processing boundary condition information and geometrical information individually as a method of efficiently analyzing structures that are symmetrical in geometry and boundary conditions, and has confirmed its validity through analyses of electrochemistry. Further, the proposed technique makes it possible to reduce analysis time significantly.

DOI： 10.1299/kikaia.75.21

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This paper presents Efficient Boundary Element Method for Analysing Target Region, which is developed by Yamagishi, et al. to efficiently compute unknown quantities only in a specific domain extended to 2D elastostatic problems. In this method, whole boundaries are devided into Neighborhood Boundaries which are near to Target Region and Distant Boundaries which are sufficienty far from it. We set up boundary integral equations whose source points are placed on all the elements of the Neighborhood Boundaries and express these integrals which contain the unknown quantities on the Distant Boundaries as low order multipole moments by using a multipole expansion of the fundamental solution. Then, we set up boundary integral equations whose source points are placed near to Distant Boundaries by the increment of those multipole moments and express these integrals which contain the unknown quantities on the Distant Boundaries approximately as the multipole moments by using a generalized inverse matrix. Thus the number of unknowns and boundary integral equations set up are decreased drastically by this method. When unknown quantities are required only in a specific domain, especially on large-scale boundary value problems, this method enable us to compute them efficiently. The capability of this method extended to 2D elastostatic problems is verified with some numerical experiments.

DOI： 10.1299/kikaia.75.993

CiNii Books

researchmap

Language：Japanese  

We developed the quantitative monitoring method for coating condition inside a ballast tank. We proposed the coating condition is evaluated with the surface resistance. We developed the identification method to obtain the whole surface resistance from the differential potential induced by the impressed current from an optional anode inside a tank. We introduced differential potential measurement and inverse analysis to obtain the surface resistance representing the coating condition. The potential measurement and quantitative evaluation were conducted in the actual ship. The verification was performed and there was the good agreement between the proposed method and the preliminary visual inspection.

DOI： 10.3323/jcorr.57.542

Scopus

researchmap

Language：Japanese  

This research proposes an effective inverse analysis technique for solving an inverse problem to detect corrosion in reinforced concrete non-destructively from potentials on a concrete surface. The technique can detect the number, the shape and the location of corrosion without particular assumption with GA (genetic algorithm), and increases accuracy of solution gradually if necessary. The following three strategies achieve the proposed technique: (1) Control of analysis precision, i.e. coarseness of BEM (boundary element method) meshes, depending on required precision of solution. (2) Use of the net elements for BEM for decreasing the number of elements for steels. (3) Analysis with higher precision only within detected corrosion area on lower precise analysis. We performed a numerical simulation to solve an inverse problem for detecting steel corrosion in concrete with the proposed technique. The result showed that the proposed method could detect corrosion with high accuracy. The calculation time of the proposed method was about 1/100 in comparison with that of the previously proposed method.

DOI： 10.3323/jcorr.57.282

Scopus

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

Copper electroplating is generally used for the fabrication of Large-Scale Integration (LSI) besides other methods such as sputtering because of its excellent via/trench filling ability, good adhesion and lower process temperature and cost. In the electroplating process, estimating a growth rate of electroplating in real-time is essential. It is possible to estimate a growth rate of electroplating from the current density on the LSI wafer surface because the rate mainly depends on the current density. In our previous research, applying inverse analysis methods, a new nondestructive method to monitor a growth rate of electroplating from the magnetic flux density outside the electroplating device measured by magnetic sensors was developed. In this paper, measurement experiments under several electroplating conditions are performed in order to verify the effectivity of this method. The growth rates of electroplating estimated by the method are compared with the thicknesses measured by the profilometer. Also the electric potentials in the electroplating device estimated by the method are compared with those measured by the electrometer. These results show that this method could be applicable to practical problems. In addition, we extend the method for an arbitrarily shaped electroplating device by applying FEM and BEM. We apply Tikhonov regularization to this method and develop a new determination procedure of the regularization parameter

DOI： 10.1299/kikaia.74.491

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00313532991?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

Copper electroplating is widely used for fabrications of Large-Scale Integration (LSI) besides other methods such as sputtering because of its excellent via/trench filling ability, good adhesion and lower process temperature and low cost. In the electroplating process, real-time estimating a growth rate of electroplating is essential. It is possible to estimate the growth rate of electroplating from the current density on the LSI wafer surface because the rate is mainly proportional to the current density. In this paper, a new monitoring method to estimate the LSI plating current density distribution from electric potentials in the plating bath is developed. The proposed method measures electric potentials at several points in the bath using capillaries, and estimates the current density on the LSI wafer surface by applying an inverse analysis method. In the inverse analysis, an observation equation is modeled by considering that the electric potentials in the bath obey the Laplace equation, which is solved by Bayesian estimation. In order to demonstrate the effectiveness of the proposed method, numerical simulations which involve measurement errors are performed for a sample polarization curve of Cu in CuSO₄. The results show that this method could be applied to practical problems. In addition, using a priori information for the possible shape of the polarization curve, the estimate accuracy of the proposed method is shown to be improved.

DOI： 10.1299/kikaia.74.1204

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00319511907?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

Copper electroplating is generally used for the fabrication of Large-Scale Integration (LSI) besides other methods such as sputtering because of its excellent via/trench filling ability, good adhesion and lower process temperature and cost. In the electroplating process, estimating a growth rate of electroplating in real-time is essential. It is possible to estimate a growth rate of electroplating from the current density on the LSI wafer surface because the rate mainly depends on the current density. In our previous research, applying inverse analysis methods, a new nondestructive method to monitor a growth rate of electroplating from the magnetic flux density outside the electroplating device measured by magnetic sensors was developed. In this paper, measurement experiments under several electroplating conditions are performed in order to verify the effectivity of this method. The growth rates of electroplating estimated by the method are compared with the thicknesses measured by the profilometer. Also the electric potentials in the electroplating device estimated by the method are compared with those measured by the electrometer. These results show that this method could be applicable to practical problems. In addition, we extend the method for an arbitrarily shaped electroplating device by applying FEM and BEM. We apply Tikhonov regularization to this method and develop a new determination procedure of the regularization parameter

DOI： 10.1299/kikaia.74.491

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00313532991?from=CiNii

DOI： 10.3323/jcorr.57.542

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

Copper electroplating is widely used for fabrications of Large-Scale Integration (LSI) besides other methods such as sputtering because of its excellent via/trench filling ability, good adhesion and lower process temperature and low cost. In the electroplating process, real-time estimating a growth rate of electroplating is essential. It is possible to estimate the growth rate of electroplating from the current density on the LSI wafer surface because the rate is mainly proportional to the current density. In this paper, a new monitoring method to estimate the LSI plating current density distribution from electric potentials in the plating bath is developed. The proposed method measures electric potentials at several points in the bath using capillaries, and estimates the current density on the LSI wafer surface by applying an inverse analysis method. In the inverse analysis, an observation equation is modeled by considering that the electric potentials in the bath obey the Laplace equation, which is solved by Bayesian estimation. In order to demonstrate the effectiveness of the proposed method, numerical simulations which involve measurement errors are performed for a sample polarization curve of Cu in CuSO₄. The results show that this method could be applied to practical problems. In addition, using a priori information for the possible shape of the polarization curve, the estimate accuracy of the proposed method is shown to be improved.

DOI： 10.1299/kikaia.74.1204

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00319511907?from=CiNii

DOI： 10.3323/jcorr.57.282

researchmap

Publisher：The Society of Instrument and Control Engineers  

This paper presents a nano-resolution measurement of magnetic fields near a magnetic probe using a thin-film magnetic sensor. The measurement method consists of a process, where the thin-film magnetic sensor is scanned over the sample, followed by a deconvolution scheme. In the scanning process, the thin-film magnetic sensor is scanned over the sample perpendicularly to the sensor. The sample is rotated horizontally under the sensor so that data measurements are obtained at different angles and positions. The deconvolution can be performed using existing methods, which are common in the field of computed tomography. An experiment was performed to verify the technique's capability of measuring a magnetic field using the thin-film magnetic sensor. Magnetic force microscopes are widely used for measuring magnetic fields. However, the magnetic field from the magnetic probe affects the magnetization of the sample when the sample's magnetization is small. Thin-film magnetic sensors do not have a magnetic field. Therefore the presented technique can measure small magnetic fields even from very sensitive magnetic samples. Currently, thin-film magnetic sensors are used as reading heads in hard drive disks (HDD). Types of heads used are thin-film inductive heads, anisotropic magnetoresistive (AMR) heads, and giant magnetoresistive (GMR) heads. Nowadays, the film thickness of the GMR head, which is most commonly used as the reading head in HDD, is less than 10nm. Previously, the resolution of the measurement was limited by the width of the sensor (ie. 100nm) which is always greater than the thickness. However, the new technique has the same resolution as the film thickness (ie. 10nm). The measurement techniques used in the HDD industry can be applied to our measurement method.

DOI： 10.9746/ve.sicetr1965.43.17

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new technique for boundary element analysis of potential problems using a multipole expansion to obtain the solution quickly only in a target region has been developed. The capability of the present technique was verified by numerical simulations. In this technique, the multipole expansion is applied to boundary integral equations on non-target boundaries, and the expansion is truncated at the term where the error bound of the higher terms is guaranteed. This technique decreases the calculation amount by introducing multipole moments as the unknown in place of the unknown in the non-target region. To demonstrate the effectiveness of the method, some example analyses were performed. When the solutions only in the target region are needed, especially in large size boundary value problems, this technique enables us to obtain them quickly and precisely.

DOI： 10.1299/kikaia.73.1225

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00303835316?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

The copper electroplating is tried to apply on the fabrication of Large-Scale Integration (LSI) besides other methods because of its excellent via/trench filling ability, good adhesion and lower process temperature and cost. Then, in the electroplating process, estimating the growth rate of the electroplating in real-time is essential. It is able to estimate the growth rate from the current density on the LSI wafer surface because the growth rate of the electroplating is proportional to the current density. However, measuring the current density on the LSI wafer surface directly is impossible. In this paper, a new monitoring method to estimate the growth rate of the electroplating by using magnetic sensors is developed. We focused on the magnetic flux density induced by the current density in the electroplating device. The present method is measuring the magnetic flux density outside the electroplating device by using magnetic sensors, and estimating the current density on the LSI wafer surface by applying the inverse analysis method. In order to demonstrate the effectivity of the present method, two cases of numerical simulations, which take account the measurement error, are performed. The results show that this method could be applicable to practical problems.

DOI： 10.1299/kikaia.73.709

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00296936731?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

The magnetostatic analysis is important on the corrosion detection problems and the nondestructive inspection of the structures. Boundary element method is often used to the magnetostatic analysis conventionally because boundary element method does not require the discretization of the magnetostatic field domain and has an advantage for open domain problems. However, in case of the current density should be considered as the source of the magnetostatic field, the conventional method requires not only numerical calculation of current density distribution, but requires numerical region integration of this current distribution which has been calculated at discrete sample points. These two steps of discrete numerical calculation of 3D space increases computational time and loses the accuracy. In this paper, a new boundary integral method for analyzing magnetostatic field without the calculation and the integration of current density distribution in the electrostatic field domain which obeys Laplace equation is developed. The region integration of current density distribution is reduced to the infinite region integration of the fundamental solution using the derivative of the boundary integral equation for electrical potential field. Since the analytical calculation of the infinite region integration could be achieved, the accuracy and computational time are improved. In order to demonstrate the accuracy and efficiency of the present method, some example problems are solved, and compared with the conventional method. Present method is also applied to the practical electro-plating problem.

DOI： 10.1299/kikaia.73.567

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00295222033?from=CiNii

Publisher：The Society of Instrument and Control Engineers  

This paper presents a nano-resolution measurement of magnetic fields near a magnetic probe using a thin-film magnetic sensor. The measurement method consists of a process, where the thin-film magnetic sensor is scanned over the sample, followed by a deconvolution scheme. In the scanning process, the thin-film magnetic sensor is scanned over the sample perpendicularly to the sensor. The sample is rotated horizontally under the sensor so that data measurements are obtained at different angles and positions. The deconvolution can be performed using existing methods, which are common in the field of computed tomography. An experiment was performed to verify the technique's capability of measuring a magnetic field using the thin-film magnetic sensor. Magnetic force microscopes are widely used for measuring magnetic fields. However, the magnetic field from the magnetic probe affects the magnetization of the sample when the sample's magnetization is small. Thin-film magnetic sensors do not have a magnetic field. Therefore the presented technique can measure small magnetic fields even from very sensitive magnetic samples. Currently, thin-film magnetic sensors are used as reading heads in hard drive disks (HDD). Types of heads used are thin-film inductive heads, anisotropic magnetoresistive (AMR) heads, and giant magnetoresistive (GMR) heads. Nowadays, the film thickness of the GMR head, which is most commonly used as the reading head in HDD, is less than 10nm. Previously, the resolution of the measurement was limited by the width of the sensor (ie. 100nm) which is always greater than the thickness. However, the new technique has the same resolution as the film thickness (ie. 10nm). The measurement techniques used in the HDD industry can be applied to our measurement method.

DOI： 10.9746/ve.sicetr1965.43.17

researchmap

Publisher：The Society of Instrument and Control Engineers  

This paper presents a nano-resolution measurement of magnetic fields near a magnetic probe using a thin-film magnetic sensor. The measurement method consists of a process, where the thin-film magnetic sensor is scanned over the sample, followed by a deconvolution scheme. In the scanning process, the thin-film magnetic sensor is scanned over the sample perpendicularly to the sensor. The sample is rotated horizontally under the sensor so that data measurements are obtained at different angles and positions. The deconvolution can be performed using existing methods, which are common in the field of computed tomography. An experiment was performed to verify the technique's capability of measuring a magnetic field using the thin-film magnetic sensor. Magnetic force microscopes are widely used for measuring magnetic fields. However, the magnetic field from the magnetic probe affects the magnetization of the sample when the sample's magnetization is small. Thin-film magnetic sensors do not have a magnetic field. Therefore the presented technique can measure small magnetic fields even from very sensitive magnetic samples. Currently, thin-film magnetic sensors are used as reading heads in hard drive disks (HDD). Types of heads used are thin-film inductive heads, anisotropic magnetoresistive (AMR) heads, and giant magnetoresistive (GMR) heads. Nowadays, the film thickness of the GMR head, which is most commonly used as the reading head in HDD, is less than 10nm. Previously, the resolution of the measurement was limited by the width of the sensor (ie. 100nm) which is always greater than the thickness. However, the new technique has the same resolution as the film thickness (ie. 10nm). The measurement techniques used in the HDD industry can be applied to our measurement method.

DOI： 10.9746/ve.sicetr1965.43.17

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new technique for boundary element analysis of potential problems using a multipole expansion to obtain the solution quickly only in a target region has been developed. The capability of the present technique was verified by numerical simulations. In this technique, the multipole expansion is applied to boundary integral equations on non-target boundaries, and the expansion is truncated at the term where the error bound of the higher terms is guaranteed. This technique decreases the calculation amount by introducing multipole moments as the unknown in place of the unknown in the non-target region. To demonstrate the effectiveness of the method, some example analyses were performed. When the solutions only in the target region are needed, especially in large size boundary value problems, this technique enables us to obtain them quickly and precisely.

DOI： 10.1299/kikaia.73.1225

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00303835316?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

The copper electroplating is tried to apply on the fabrication of Large-Scale Integration (LSI) besides other methods because of its excellent via/trench filling ability, good adhesion and lower process temperature and cost. Then, in the electroplating process, estimating the growth rate of the electroplating in real-time is essential. It is able to estimate the growth rate from the current density on the LSI wafer surface because the growth rate of the electroplating is proportional to the current density. However, measuring the current density on the LSI wafer surface directly is impossible. In this paper, a new monitoring method to estimate the growth rate of the electroplating by using magnetic sensors is developed. We focused on the magnetic flux density induced by the current density in the electroplating device. The present method is measuring the magnetic flux density outside the electroplating device by using magnetic sensors, and estimating the current density on the LSI wafer surface by applying the inverse analysis method. In order to demonstrate the effectivity of the present method, two cases of numerical simulations, which take account the measurement error, are performed. The results show that this method could be applicable to practical problems.

DOI： 10.1299/kikaia.73.709

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00296936731?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

The magnetostatic analysis is important on the corrosion detection problems and the nondestructive inspection of the structures. Boundary element method is often used to the magnetostatic analysis conventionally because boundary element method does not require the discretization of the magnetostatic field domain and has an advantage for open domain problems. However, in case of the current density should be considered as the source of the magnetostatic field, the conventional method requires not only numerical calculation of current density distribution, but requires numerical region integration of this current distribution which has been calculated at discrete sample points. These two steps of discrete numerical calculation of 3D space increases computational time and loses the accuracy. In this paper, a new boundary integral method for analyzing magnetostatic field without the calculation and the integration of current density distribution in the electrostatic field domain which obeys Laplace equation is developed. The region integration of current density distribution is reduced to the infinite region integration of the fundamental solution using the derivative of the boundary integral equation for electrical potential field. Since the analytical calculation of the infinite region integration could be achieved, the accuracy and computational time are improved. In order to demonstrate the accuracy and efficiency of the present method, some example problems are solved, and compared with the conventional method. Present method is also applied to the practical electro-plating problem.

DOI： 10.1299/kikaia.73.567

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00295222033?from=CiNii

DOI： 10.1299/kikaic.72.1478

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, a new inverse analysis method were introduced for optical computed tomography (optical C.T.). In optical C.T., near-inferred light is exposed into human body and scattered light is measured to detect a oxygenated part. This inverse problem is reduced to estimate the absorption distribution from scattered light on a surface of human body. To relax the ill-posed of this inverse problem, we proposed this problem is reduced to 0-1 integer problem by considering absoption coefficient as two discrete values of tumor and normal tissue. And severe constraint on discrete value is relaxed by treating absorption coefficient as probability variable. Simple 3D example problem was solved to demostrate the usefulness of this method.

DOI： 10.1299/kikaia.72.1437

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00286912219?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

An efficient boundary element method is proposed for three-dimensional corrosion analysis of structures having net geometory, such as rebars in a reinforced concrete structure and a guard fence for a sea-side structure. The proposed method is based on the Fourier series expansion of boundary quantities, i.e., potential and current density, in the circumferential and axial directions of a net structure. It is possible for the new method to reduce largely an enormous number of elements which would be necessary in the conventional method to discretize the structures having net geometory. In order to demonstrate the usefulness of the proposed method, a couple of example problems of rebars in a concrete structure with a cathodic protection system are presented.

DOI： 10.1299/kikaia.72.830

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00281577673?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new efficient and accurate method which identifies phase distribution of coherent light from its modulated intensity images has been developed. This method is summarized as the following. 1) The observation equations of the conventional method is simplified. The unknown phase distribution is expanded to Zernike series and neglected its higher order terms. Then the identification problem reduced to the non linear optimization problem whose design valuables are the Zernike coefficients. These treatment regularized the ill condition of the problem. The non-linear optimization problem could be solved by the quasi Newton method. 2) Fast Fourier Transform was applied to the large scale matrix calculation, in order to decrease the computational time. 3) The matrix which represents the coherent light propagation is accurately modeled with boundary element method. 4) Some numerical example problems are solved to demonstrate the effectiveness of the present method. 5) Practical experiment was performed, and the phase distribution could be identified from real data within two seconds. It is noted that various types of optical setting can be employed to modulate the light.

DOI： 10.1299/kikaic.72.1486

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

The optical phase distribution measurement is an important technology for many industrial applications. However, it requires a high precision measurement instrument which usually is expensive, delicate and requires laborious maintenance care. In this study, a new method which reconstructs optical phase distribution from intensity images has been developed. This method enables simple phase distribution measurement without expensive optical equipments and installation space. Firstly, the light is differently modulated by some optical systems, and intensity images of modulated light are measured by CCD camera. An inverse problem which reconstructs the phase distribution from the proposed measurement system is set up. This inverse problem is reduced to an unconstrained optimization problem which can be solved with non-linear optimization method such as BFGS method. The ill-posedness of the problem was overcome with regularization using a priori information whose unknown phase distribution is smooth. The effectiveness of the proposed method was demonstrated through some numerical simulation.

DOI： 10.1299/kikaic.72.1478

CiNii Books

researchmap

researchmap

Publisher：The Society of Instrument and Control Engineers  

This paper introduces a magnetic field measurement technique with nano-resolution using a thin-film magnetic sensor. The method consists of a process, where the thin-film magnetic sensor is scanned over the sample, followed by a processing using a deconvolution scheme. In the scanning process, the thin-film magnetic sensor is scanned over the sample perpendicularly to the sensor. The sample is rotated horizontally under the sensor so that data measurements are obtained at different angles and positions. The deconvolution can be performed using existing methods, which are familiar in computed tomography technology. Numerical simulations were performed to demonstrate the technique's capability of achieving nano-resolution measurements of a magnetic field using the thin-film magnetic sensor. Currently, thin-film magnetic sensors are used as reading heads in hard drive disks (HDD), such as thin-film inductive heads, anisotropic magnetoresistive (AMR) heads, and giant magnetoresistive (GMR) heads. Recently, the film thickness of the GMR head, which is most commonly used as the reading head in HDD, is less than 10nm. Previously, the resolution of the measurement was limited by the width of the sensor (ie. 100nm), however, the new technique has the same resolution as the film thickness (ie. 10nm). The measurement techniques of the HDD industry field can be applied to our measurement method.

DOI： 10.9746/sicetr1965.42.597

researchmap

DOI： 10.1299/kikaia.72.1437

researchmap

Publisher：The Society of Instrument and Control Engineers  

This paper introduces a magnetic field measurement technique with nano-resolution using a thin-film magnetic sensor. The method consists of a process, where the thin-film magnetic sensor is scanned over the sample, followed by a processing using a deconvolution scheme. In the scanning process, the thin-film magnetic sensor is scanned over the sample perpendicularly to the sensor. The sample is rotated horizontally under the sensor so that data measurements are obtained at different angles and positions. The deconvolution can be performed using existing methods, which are familiar in computed tomography technology. Numerical simulations were performed to demonstrate the technique's capability of achieving nano-resolution measurements of a magnetic field using the thin-film magnetic sensor. Currently, thin-film magnetic sensors are used as reading heads in hard drive disks (HDD), such as thin-film inductive heads, anisotropic magnetoresistive (AMR) heads, and giant magnetoresistive (GMR) heads. Recently, the film thickness of the GMR head, which is most commonly used as the reading head in HDD, is less than 10nm. Previously, the resolution of the measurement was limited by the width of the sensor (ie. 100nm), however, the new technique has the same resolution as the film thickness (ie. 10nm). The measurement techniques of the HDD industry field can be applied to our measurement method.

DOI： 10.9746/sicetr1965.42.597

researchmap

DOI： 10.1299/kikaia.72.830

researchmap

DOI： 10.1299/kikaic.72.1486

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

The optical phase distribution measurement is an important technology for many industrial applications. However, it requires a high precision measurement instrument which usually is expensive, delicate and requires laborious maintenance care. In this study, a new method which reconstructs optical phase distribution from intensity images has been developed. This method enables simple phase distribution measurement without expensive optical equipments and installation space. Firstly, the light is differently modulated by some optical systems, and intensity images of modulated light are measured by CCD camera. An inverse problem which reconstructs the phase distribution from the proposed measurement system is set up. This inverse problem is reduced to an unconstrained optimization problem which can be solved with non-linear optimization method such as BFGS method. The ill-posedness of the problem was overcome with regularization using a priori information whose unknown phase distribution is smooth. The effectiveness of the proposed method was demonstrated through some numerical simulation.

DOI： 10.1299/kikaic.72.1478

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, a new inverse analysis method were introduced for optical computed tomography (optical C.T.). In optical C.T., near-inferred light is exposed into human body and scattered light is measured to detect a oxygenated part. This inverse problem is reduced to estimate the absorption distribution from scattered light on a surface of human body. To relax the ill-posed of this inverse problem, we proposed this problem is reduced to 0-1 integer problem by considering absoption coefficient as two discrete values of tumor and normal tissue. And severe constraint on discrete value is relaxed by treating absorption coefficient as probability variable. Simple 3D example problem was solved to demostrate the usefulness of this method.

DOI： 10.1299/kikaia.72.1437

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00286912219?from=CiNii

Publisher：The Society of Instrument and Control Engineers  

This paper introduces a magnetic field measurement technique with nano-resolution using a thin-film magnetic sensor. The method consists of a process, where the thin-film magnetic sensor is scanned over the sample, followed by a processing using a deconvolution scheme. In the scanning process, the thin-film magnetic sensor is scanned over the sample perpendicularly to the sensor. The sample is rotated horizontally under the sensor so that data measurements are obtained at different angles and positions. The deconvolution can be performed using existing methods, which are familiar in computed tomography technology. Numerical simulations were performed to demonstrate the technique's capability of achieving nano-resolution measurements of a magnetic field using the thin-film magnetic sensor. Currently, thin-film magnetic sensors are used as reading heads in hard drive disks (HDD), such as thin-film inductive heads, anisotropic magnetoresistive (AMR) heads, and giant magnetoresistive (GMR) heads. Recently, the film thickness of the GMR head, which is most commonly used as the reading head in HDD, is less than 10nm. Previously, the resolution of the measurement was limited by the width of the sensor (ie. 100nm), however, the new technique has the same resolution as the film thickness (ie. 10nm). The measurement techniques of the HDD industry field can be applied to our measurement method.

DOI： 10.9746/sicetr1965.42.597

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

An efficient boundary element method is proposed for three-dimensional corrosion analysis of structures having net geometory, such as rebars in a reinforced concrete structure and a guard fence for a sea-side structure. The proposed method is based on the Fourier series expansion of boundary quantities, i.e., potential and current density, in the circumferential and axial directions of a net structure. It is possible for the new method to reduce largely an enormous number of elements which would be necessary in the conventional method to discretize the structures having net geometory. In order to demonstrate the usefulness of the proposed method, a couple of example problems of rebars in a concrete structure with a cathodic protection system are presented.

DOI： 10.1299/kikaia.72.830

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00281577673?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new efficient and accurate method which identifies phase distribution of coherent light from its modulated intensity images has been developed. This method is summarized as the following. 1) The observation equations of the conventional method is simplified. The unknown phase distribution is expanded to Zernike series and neglected its higher order terms. Then the identification problem reduced to the non linear optimization problem whose design valuables are the Zernike coefficients. These treatment regularized the ill condition of the problem. The non-linear optimization problem could be solved by the quasi Newton method. 2) Fast Fourier Transform was applied to the large scale matrix calculation, in order to decrease the computational time. 3) The matrix which represents the coherent light propagation is accurately modeled with boundary element method. 4) Some numerical example problems are solved to demonstrate the effectiveness of the present method. 5) Practical experiment was performed, and the phase distribution could be identified from real data within two seconds. It is noted that various types of optical setting can be employed to modulate the light.

DOI： 10.1299/kikaic.72.1486

CiNii Books

researchmap

Language：English   Publisher：WILEY-BLACKWELL  

An efficient implicit dynamic finite element method (FEM) for elastic 3D objects with uniform cross-sections was developed. In this method, the finite element mesh is generated in such a way that the object to be analysed is at first sliced into layers with the same thickness along its generatrix and then each layer is discretized into finite elements of the same pattern. This way of discretization makes the mass, viscosity, and stiffness matrices into the repetitive block tridiagonal matrices. The repetitive block tridiagonal matrix has the characteristic, that the sequence of matrices which appears in the Gaussian elimination for the repetitive block tridiagonal matrix is a rapid convergent sequence. The process of the Gaussian elimination can be terminated when the sequence converges. The rest of the sequence is not necessary to be stored. The present method can save the computational time and memory by utilising this characteristic of the repetitive block tridiagonal matrix. A few examples of analyses including whole Hopkinson-bar analysis were performed to demonstrate the effectiveness of the present method. The present method is applicable not only to the elasto-dynamics but also to many other problems, such as thermal problems, electrical problems, and plastic problems without geometric non-linearity. Copyright (c) 2005 John Wiley & Sons, Ltd.

DOI： 10.1002/nme.1340

Web of Science

researchmap

Language：English   Publisher：NATL ASSOC CORROSION ENG  

A method to identify corrosion of steel in concrete is introduced by combining magnetic field measurement and a boundary element inverse analysis. A better sensitivity of corrosion identification can be achieved if the superconducting quantum interference device (SQUID) sensor could be applied to measure a very low magnetic field on the concrete structure as a result of the impressed current and corrosion of the reinforcing steel. In this method, the inverse problem is formulated by minimizing the cost function, which is a residual between the measured and calculated values of magnetic flux densities on the concrete surface. The magnetic flux density is calculated based on the law of Biot and Savart and takes into account the current that flows in the concrete domain and in the reinforcing steels. Laplace's equation is solved by the boundary element method to obtain the potential and current density in the concrete domain and in the whole surface of the domain. Although the ferromagnetic effect is neglected in this simulation, it is suggested from the numerical simulations result that the method can be used to accurately detect an early corrosion of the steel in concrete and the corrosion of the second-layer steel in the concrete structures.

Web of Science

researchmap

researchmap

researchmap

researchmap

researchmap

Language：English   Publisher：ELSEVIER SCI LTD  

A new technique for boundary element analyses of 2D potential problems to quickly obtain the solution only in a target region was developed. This technique takes advantage of diffusion effects in Laplace fields. In fields governed by diffusion equations, high-frequency disturbances of boundary conditions on non-target boundaries far from the target region give little influence to the target region. In this technique, boundary conditions on non-target boundaries are transformed into the Fourier series, and only their low-frequency terms are utilized by using special weight functions for boundary integral equations. When the solutions only in a target region are needed, especially in large size boundary value problems, this technique enables us to obtain them quickly and precisely. The present method can be extended not only to Laplace equations but also to any kind of diffusive systems such as elastostatics. (C) 2004 Elsevier Ltd. All rights reserved.

DOI： 10.1016/j.enganabound.2003.12.008

Web of Science

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

DOI： 10.1299/kikaia.70.31

researchmap

researchmap

researchmap

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new boundary element method (BEM) for analyzing the potential field including thin plates is developed. It is known that the inaccurate solutions are encountered with the standard BEM for this potential problem. In case each source point on foreside and backside of the thin plate is close, the boundary integral equations for these sources nearly depend each other. This is because the fundamental solution distributions for each source are almost the same. In order to overcome this problem, the special weighting function whose form is not symmetric instead of the standard fundamental solution in the BEM is proposed. The weighting function for 2 D potential problem is derived. A few example problems are solved with this method to demonstrate its applicability and usefulness.

DOI： 10.1299/kikaia.70.31

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00235716356?from=CiNii

researchmap

researchmap

researchmap

researchmap

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new boundary element method (BEM) for analyzing the potential field including thin plates is developed. It is known that the inaccurate solutions are encountered with the standard BEM for this potential problem. In case each source point on foreside and backside of the thin plate is close, the boundary integral equations for these sources nearly depend each other. This is because the fundamental solution distributions for each source are almost the same. In order to overcome this problem, the special weighting function whose form is not symmetric instead of the standard fundamental solution in the BEM is proposed. The weighting function for 2 D potential problem is derived. A few example problems are solved with this method to demonstrate its applicability and usefulness.

DOI： 10.1299/kikaia.70.31

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00235716356?from=CiNii

researchmap

researchmap

researchmap

researchmap

researchmap

DOI： 10.1299/kikaia.70.31

researchmap

researchmap

researchmap

researchmap

researchmap

Language：English   Publisher：ELSEVIER SCI LTD  

Effective boundary element methods to analyze corrosion problems for complicated structures were presented. The methods can be classified into two groups. The first one makes use of the characteristics of the structure to be analyzed, i.e. (1) thousands of same size tubes in an heat exchanger, for example, are treated by introducing a concept 'macroscopic polarization curve', (2) slender parts in a structure such as an off-shore structure and a bridge girder are analyzed by a kind of 'zooming method', (3) a non-axisymmetric potential distribution in an axisymmetric structure such as a pipeline and a seawater pump affected with outside structures is calculated by 'Fourier series method', and (4) target sub-region to be analyzed in detail, e.g. interior of guide casing of seawater pump, is analyzed by treating accompanying sub-regions as 'equivalent boundary conditions'. The second one is an application of the Fast Multipole boundary element method (FMBEM), and can be used for an arbitrary structure. The method is also useful for optimum design of cathodic protection. (C) 2003 Elsevier Science Ltd. All rights reserved.

DOI： 10.1016/S0955-7997(02)00158-3

Web of Science

researchmap

Language：English   Publisher：SPRINGER  

A numerical simulation is performed to identify the material parameters of an elastic/visco-plastic material under tensile impact loading. The identification is based on an inverse analysis using the Kalman filter technique and the finite element method. The input data to the inverse analysis are the time histories of the strains on specimens. The measurement errors can be treated reasonably by the Kalman filter technique, and the finite element method enables us to employ arbitrarily shaped specimens. Notched specimens are employed to obtain high strain rates at the notch root. It is found that an additional test under a relatively low strain rate is needed together with a test under a high strain rate to increase the accuracy of identification. The optimum strain rate in the additional test is determined by minimizing the largest eigen value of the covariance matrix of the estimated parameters.

Web of Science

researchmap

researchmap

researchmap

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This paper presents a new inverse analysis method based on wavelet transform theory. For solving inverse problems, it is important to use a priori information efficiently. This method enables to use the local frequency-characteristic of estimates as a priori information which has not be used in conventional methods. Firstly, the estimates are divided into some components by using wavelet transform algorithm. Each component has different frequency-characteristic/resolution. Secondly, according to a priori information, elements in the component are manipulated. Thirdly, those components are identified from low to high frequency-characteristic/resolution component sequentially. Finally, the estimates are reconstructed from those identified components by using inverse wavelet transform algorithm. A numerical simulation is performed to demonstrate the validity and usefulness of the proposed technique.

DOI： 10.1299/kikaia.69.225

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00164135327?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This paper presents a new inverse analysis method based on wavelet transform theory. For solving inverse problems, it is important to use a priori information efficiently. This method enables to use the local frequency-characteristic of estimates as a priori information which has not be used in conventional methods. Firstly, the estimates are divided into some components by using wavelet transform algorithm. Each component has different frequency-characteristic/resolution. Secondly, according to a priori information, elements in the component are manipulated. Thirdly, those components are identified from low to high frequency-characteristic/resolution component sequentially. Finally, the estimates are reconstructed from those identified components by using inverse wavelet transform algorithm. A numerical simulation is performed to demonstrate the validity and usefulness of the proposed technique.

DOI： 10.1299/kikaia.69.225

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00164135327?from=CiNii

researchmap

researchmap

researchmap

researchmap

DOI： 10.1016/B978-008044268-6/50027-2

researchmap

researchmap

researchmap

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this research, the material constants of Gurson model for porous elastic-plastic materials are identified. Measurement data, obtained by simple experiments (compressive test and tensile test) with carbon steel (S25C) specimen, are used. The identification method is based on an inverse analysis using Bayes Formula and the finite element method. Preliminarily, sensitivity analyses are performed for verification of parameter-independency, and efficient identification procedure is proposed.

DOI： 10.1299/kikaia.69.832

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00208632393?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this research, the material constants of Gurson model for porous elastic-plastic materials are identified. Measurement data, obtained by simple experiments (compressive test and tensile test) with carbon steel (S25C) specimen, are used. The identification method is based on an inverse analysis using Bayes Formula and the finite element method. Preliminarily, sensitivity analyses are performed for verification of parameter-independency, and efficient identification procedure is proposed.

DOI： 10.1299/kikaia.69.832

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00208632393?from=CiNii

researchmap

researchmap

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new technique of boundary element methods for 2D potential problems to obtain the solution only in a target region quickly was developed. This technique takes advantage of diffusion effects in Laplace fields. In fields governed by diffusion equations, high frequency disturbances of boundary conditions on non-target bounds far from the target region give little infection to the target region. In this technique, boundary conditions on non-target bound are transformed into Fourier series, and their only low frequency terms are utilized by using special weight functions for boundary integral equations. When the solutions only in a target region are needed especially in large size boundary value problems, this technique enable us to obtain them quickly and precisely with solving non-target solutions roughly.

DOI： 10.1299/jsmemecjo.2003.1.0_65

CiNii Books

researchmap

researchmap

DOI： 10.1299/kikaia.69.832

researchmap

DOI： 10.1299/kikaia.69.832

researchmap

DOI： 10.1299/kikaia.69.225

researchmap

DOI： 10.1299/kikaia.69.225

researchmap

researchmap

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this research, the material constants of Gurson model for porous elastic-plastic materials are identified. Measurement data, obtained by simple experiments (compressive test and tensile test) with carbon steel (S25C) specimen, are used. The identification method is based on an inverse analysis using Bayes Formula and the finite element method. Preliminarily, sensitivity analyses are performed for verification of parameter-independency, and efficient identification procedure is proposed.

DOI： 10.1299/kikaia.69.832

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00208632393?from=CiNii

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

DOI： 10.1016/B978-008044268-6/50027-2

researchmap

DOI： 10.1299/kikaia.68.313

researchmap

researchmap

researchmap

DOI： 10.1299/kikaia.68.924

researchmap

DOI： 10.1299/kikaia.68.313

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A numerical method to analyze a one-point-bend specimen with a fast propagating crack was developed using the Timoshenko beam theory. The displacement and the dynamic stress intensity factor can be obtained by this method if the time histories of the impact force and crack velocity are given. A numerical simulation of the one-point-bend impact test was carried out using the above method and the Hertz theory of contact. The time dependence of impact force (and hence the dynamic stress intensity factor) can be calculated if the impact velocity and the time history of crack velocity are given. The possibility of measuring the fast crack propagation toughness using a one-point-bend impact test machine was discussed.

DOI： 10.1299/kikaia.68.924

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00157653326?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new method, combination of a boundary element method and a finite element method, has been developed to simulate the electroplating of a silicon wafer with copper. In this method, the Laplace equation is solved by representing the complicated phenomena near the anode and cathode surface as polarization characteristics (PCs) and using the PCs as nonlinear boundary conditions. The electrical resistance distribution of thin copper and TaN films on a silicon wafer is then taken into account. To make a uniform copper film, the anode (copper plate) is divided suitably into several pieces. To design the size of each multiple anodes rationally, the ideal anode current density distribution which is obtained by solving inverse problem are used. In the inverse problem both of the potential and the current density on the cathode are given as the ideal values, i. e., the uniform current density and its corresponding potential. Optimization of current to be supplied to each piece of the anode is performed by employing the Simplex method.

DOI： 10.1299/kikaia.68.313

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00155463330?from=CiNii

researchmap

DOI： 10.1299/kikaia.68.313

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A numerical method to analyze a one-point-bend specimen with a fast propagating crack was developed using the Timoshenko beam theory. The displacement and the dynamic stress intensity factor can be obtained by this method if the time histories of the impact force and crack velocity are given. A numerical simulation of the one-point-bend impact test was carried out using the above method and the Hertz theory of contact. The time dependence of impact force (and hence the dynamic stress intensity factor) can be calculated if the impact velocity and the time history of crack velocity are given. The possibility of measuring the fast crack propagation toughness using a one-point-bend impact test machine was discussed.

DOI： 10.1299/kikaia.68.924

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00157653326?from=CiNii

researchmap

researchmap

researchmap

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A numerical method to analyze a one-point-bend specimen with a fast propagating crack was developed using the Timoshenko beam theory. The displacement and the dynamic stress intensity factor can be obtained by this method if the time histories of the impact force and crack velocity are given. A numerical simulation of the one-point-bend impact test was carried out using the above method and the Hertz theory of contact. The time dependence of impact force (and hence the dynamic stress intensity factor) can be calculated if the impact velocity and the time history of crack velocity are given. The possibility of measuring the fast crack propagation toughness using a one-point-bend impact test machine was discussed.

DOI： 10.1299/kikaia.68.924

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00157653326?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new method, combination of a boundary element method and a finite element method, has been developed to simulate the electroplating of a silicon wafer with copper. In this method, the Laplace equation is solved by representing the complicated phenomena near the anode and cathode surface as polarization characteristics (PCs) and using the PCs as nonlinear boundary conditions. The electrical resistance distribution of thin copper and TaN films on a silicon wafer is then taken into account. To make a uniform copper film, the anode (copper plate) is divided suitably into several pieces. To design the size of each multiple anodes rationally, the ideal anode current density distribution which is obtained by solving inverse problem are used. In the inverse problem both of the potential and the current density on the cathode are given as the ideal values, i. e., the uniform current density and its corresponding potential. Optimization of current to be supplied to each piece of the anode is performed by employing the Simplex method.

DOI： 10.1299/kikaia.68.313

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00155463330?from=CiNii

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

An efficient dynamic finite element method (FEM) for an elastic body with uniform cross-section was developed. In this method, the finite element mesh is generated in such a way that the body to be analyzed is at first sliced into layers with the same thickness and then each layer is discretized in the same way. The computational time and memory can be saved by utilizing the regularity in the mass, viscosity and stiffness matrices and making them repetitive block tridiagonal matrices, which results from the above mesh. A few example problems were solved with this method to demonstrate its effectiveness. This method is also applicable to many other problems, such as thermal and electrical problems.

DOI： 10.1299/kikaia.67.1583

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00151360923?from=CiNii

DOI： 10.1299/kikaia.67.1417

researchmap

DOI： 10.1299/kikaia.67.1263

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

An efficient dynamic finite element method (FEM) for an elastic body with uniform cross-section was developed. In this method, the finite element mesh is generated in such a way that the body to be analyzed is at first sliced into layers with the same thickness and then each layer is discretized in the same way. The computational time and memory can be saved by utilizing the regularity in the mass, viscosity and stiffness matrices and making them repetitive block tridiagonal matrices, which results from the above mesh. A few example problems were solved with this method to demonstrate its effectiveness. This method is also applicable to many other problems, such as thermal and electrical problems.

DOI： 10.1299/kikaia.67.1583

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00151360923?from=CiNii

DOI： 10.1299/kikaia.67.1417

researchmap

DOI： 10.1299/kikaia.67.1263

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

Language：Japanese   Publisher：Japan Society of Mechanical Engineers  

This paper presents a method for identifying the Gurson's material constants by inverse analysis. Firstly, in older to improve the accuracy, the measurement condition. such as geometries of specimens. locations of measurements and types of measurements. is optimized. This approach is based on the Kalman filter technique. The optimization is achieved by minimizing the maximum eigen value of a posteriori estimate error covariance matrix. Mathematically. this procedure is reduced to solve a combinational/discrete optimization problem. Secondly, the Gurson's material constants is estimated by using Kalman filter. A numerical identification is presented to demonstrate the validity and usefulness of proposed method.

DOI： 10.1299/kikaia.67.1892

Scopus

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

An efficient dynamic finite element method (FEM) for an elastic body with uniform cross-section was developed. In this method, the finite element mesh is generated in such a way that the body to be analyzed is at first sliced into layers with the same thickness and then each layer is discretized in the same way. The computational time and memory can be saved by utilizing the regularity in the mass, viscosity and stiffness matrices and making them repetitive block tridiagonal matrices, which results from the above mesh. A few example problems were solved with this method to demonstrate its effectiveness. This method is also applicable to many other problems, such as thermal and electrical problems.

DOI： 10.1299/kikaia.67.1583

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00151360923?from=CiNii

Language：Japanese   Publisher：Japan Society of Mechanical Engineers  

A method for identifying the material constants of an elastic/visco-plastic meterial under tensile impact loading is proposed. In this method, the Kalman filter and the finite element method are used together with a few test data. To obtain data for high strain rates, notched specimens are employed, because arbitrarily shaped specimens are available in this method. To increase the accuracy of identification, a test under a relatively low strain rate is needed in addition to a test under a high strain rate. The optimum strain rate is determined in such a way that the largest eigen value of the covariance matrix of the estimated parameters becomes minimum. A numerical simulation in performed to demonstrate the validity and usefulness of this method.

DOI： 10.1299/kikaia.67.1

Scopus

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

DOI： 10.1299/kikaia.67.1583

researchmap

Language：Japanese   Publisher：Japan Society of Mechanical Engineers  

An effective optimization technique for cathodic protection design is developed. To cope with the large amount of memory and calculation, the fast multipole boundary element method (FMBEM) is applied. To achieve an effective optimization, the initial values of the design variables, i.e., the location and current of each electrode, are determined by solving an inverse problem. In the inverse problem both of the potential and the current density on the metal surfaces to be protected are given as the ideal values, i.e., the protection potential and its corresponding current density. The inverse problem is solved effectively by using the FMBEM. Using thus obtained initial values, the optimization problem is solved by the symplex method. A couple of example problems are solved to demonstrate the effectiveness and usefulness of the present method.

DOI： 10.1299/kikaia.67.1417

Scopus

researchmap

Language：Japanese   Publisher：Japan Society of Mechanical Engineers  

A new stochastic boundary element method for analyzing the corrosion of underground structure was developed. In case of analyzing the corrosion of underground structure buried in soil which electric conductivity and polarization characteristics are changing along time, it is necessary to consider about the stochastic character of these unstable condition (conductivity and polarization characteristics). The non-linear relationship between unknown variables and these unstable condition was linearized by Taylor series and, the mean and covariance of unknown variables are estimated. In order to demonstrate the effectiveness of the method, the simulation was performed and the result was compared with conventional Monte Carlo method and the field experimental result. A computer simulation of the identification is performed at first with a measured potential by using the Kalman filter which stochastic model was considered with the proposed method.

DOI： 10.1299/kikaia.67.1263

Scopus

researchmap

researchmap

Language：English   Publisher：Tokai University  

A new thermo resist enhanced optical lithography (TREOL) technique is proposed. Numerical simulation was performed on the TREOL. The TREOL offer the possibility of greatly enhanced resoluion and process window. The TREOL uses a thermo-resist instead of photoresist taking advantage of a nonlinear superposition effect. The nonlinear superposition utilizes the fact that thermo-resists do not follow the reciprocity law. This phenomena allows to separate a complex image into simple sub-sets which are imaged onto the same die. This concept will realize a fabrication of 1 features with existing steppers and without and proximity effects. Hopkin's four-dimensional integral is used for calculation of the diffraction intensity of projection image. Intensity of impressed heat flux is assumed to be proportional to the intensity of this diffraction image. Transient heat transfer in wafer and layer of thermo-resist is analyzed by non-linear 3 dimensional finite element method. The numerical result demonstrates an drastic improvement in resolution and process window.

DOI： 10.2494/photopolymer.13.441

Scopus

researchmap

DOI： 10.4139/sfj.51.425

researchmap

researchmap

DOI： 10.4139/sfj.51.425

researchmap

researchmap

researchmap

Language：Japanese   Publisher：The Surface Finishing Society of Japan  

A new boundary element method has been developed to simulate the electroplating of a silicon wafer with copper. In this method, the Laplace equation is solved with three-dimensional or axisymmetric elements by representing the complicated phenomena near the anode and cathode surfaces as polarization characteristics (PCs) and using the PCs as nonlinear boundary conditions. Prior to electroplating, sputtering is used to coat a silicon wafer with a Tantalum nitride (TaN) film. The electrical resistance of thin copper and TaN films on a silicon wafer is then taken into account. To make a uniform copper film, the anode (copper plate) is divided into several pieces, and optimization of the current supplied to each piece of the anode is performed through use of the Simplex method.

DOI： 10.4139/sfj.51.425

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00066318293?from=CiNii

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

DOI： 10.4139/sfj.51.425

researchmap

researchmap

DOI： 10.1299/kikaia.65.1493

researchmap

DOI： 10.1299/kikaic.65.4048

researchmap

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new method for non-linear state space estimation using genetic algorithm is developed. This method is named the Genetic algorithm filter (GA filter). State vactors are coded with genes and their survival rates are decided by a probability density functions. A crowd of genes represents the characteristic of a current state of a system. GA are general techniques for doing numerical optimization and are based on stochastic simulation. Their strength is that they can be applied to problems where an analytical approach is difficult or impossible. GA filter can handle non-linear system equations and measurement equations as well as non-Gaussian and non-additive noise. Implementing the GA requires less analytical work than implementing the extended Kalman filter, but the GA filters are more computer intensive. We will demonstrate effectiveness of the resulting new filters with simple examples.

DOI： 10.1299/kikaic.65.4329

CiNii Books

researchmap

researchmap

researchmap

researchmap

DOI： 10.1299/kikaia.65.1493

researchmap

DOI： 10.2472/jsms.48.1316

researchmap

DOI： 10.2464/jilm.49.452

researchmap

DOI： 10.2208/journalam.2.177

researchmap

researchmap

researchmap

Language：English   Publisher：Japan Soc of Corrosion Engineering  

A new method based on the boundary element inverse analysis was developed to improve the accuracy and efficiency of AC impedance method for monitoring corrosion of a rebar in a concrete structure. In this method, the potential in the concrete domain is described with the Laplace's equation, the validity of which was verified experimentally by using a concrete block specimen. The potential in the concrete domain is calculated by solving the Laplace's equation using the boundary element method, and an inverse analysis is carried out to estimate the location and the complex impedance of the corroded part on the surface of a rebar from a small number of potential data measured on the surface of a concrete structure. The cost function to be minimized in the inverse analysis is a function of the difference between the measured and calculated values of potential. An example using a prismatic concrete block with an embedded piece of rebar was presented to demonstrate the accuracy and efficiency of the proposed method.

DOI： 10.3323/jcorr1991.48.654

Scopus

researchmap

researchmap

Language：English   Publisher：SPIE-INT SOC OPTICAL ENGINEERING  

Numerical simulation was performed on a new high resolution micro-lithography using thermo-resist. This high resolution optical-thermal lithography uses a thermo-resist instead of photoresist taking advantage of a nonlinear superposition effect. The nonlinear superposition utilizes the fact that thermo-resists do not follow the reciprocity law. This phenomena allows tc, separate a complex image into simple sub-sets which are imaged onto the same die. This concept will realize a fabrication of 100nm features with existing steppers and without and proximity effects. Hopkin's four-dimensional integral is used for calculation of the diffraction intensity of projection image. Intensity of impressed heat flux is assumed to be proportional to the intensity of this diffraction image. Transient heat transfer in wafer and layer of thermo-resist is analyzed by non-linear 3 dimensional finite element method. The numerical result demonstrates all drastic improvement in resolution that can be obtained with conventional method using photo-resist.

Web of Science

researchmap

researchmap

Language：Japanese   Publisher：Japan Institute of Light Metals  

DOI： 10.2464/jilm.49.452

Scopus

researchmap

DOI： 10.2464/jilm.49.452

researchmap

Language：English  

A new method for non-linear state space estimation using genetic algorithm is developed. This method is named the Genetic algorithm filter (GA filter). State vactors are coded with genes and their survival rates are decided by a probability density functions. A crowd of genes represents the characteristic of a current state of a system. GA are general techniques for doing numerical optimization and are based on stochastic simulation. Their strength is that they can be applied to problems where an analytical approach is difficult or impossible. GA filter can handle non-linear system equations and measurement equations as well as non-Gaussian and non-additive noise. Implementing the GA requires less analytical work than implementing the extended Kalman filter, but the GA filters are more computer intensive. We will demonstrate effectiveness of the resulting new filters with simple examples. © 1999, The Japan Society of Mechanical Engineers. All rights reserved.

DOI： 10.1299/kikaic.65.4329

Scopus

researchmap

Language：Japanese   Publisher：Soc Mater Sci, Japan  

Application of the fast multipole boundary element method (FMBEM) to corrosion problems is studied. It is found that this method can be successfully applied to corrosion analysis by employing the previous research results on the potential problems together with the following procedures
(1) use of M00 of the root cell (which is obtained in the process of multipole expansion) for the infinite region problems, and (2) modification of the residual calculation formula in the Bi-CGSTAB iterative method to cope with the nonlinearity caused by polarization curves. A pipe element which is suitable to the FMBEM is also developed to increase the efficiency of calculation for pipelines. A few examples are shown to demonstrate the applicability of the FMBEM to corrosion problems.

DOI： 10.2472/jsms.48.1316

Scopus

researchmap

Language：Japanese   Publisher：Society of Materials Science Japan  

It was verified by two experiments that the potential in soil satisfied the Laplace equation, which is usually assumed in numerical analysis methods for the cathodic protection of buried steel structures. In the first experiment, where the current distribution was uniform, it was found that the relationship between the potential and the distance was linear. In the second experiment, where the current distribution was complicated, it was found that the experimental results agreed well with the calculated results based on the Laplace equation. The polarization curve for steel in soil was also measured and its characteristics were discussed from the viewpoint of use in the numerical analysis.

DOI： 10.2472/jsms.48.1244

Scopus

researchmap

Language：English  

An effective method for designing neural network to classify patterns in the multidimensional feature space is introduced. Voronoi diagram is constructed over points in feature space which represents input teachers in order to realize the desired classification. In order to reduce the size of neural network and to make the learning process efficiently, clustering procedure which enables to manage a number of teachers in a lump is implemented. Two examples demonstrate that the present method gives stable result and a calculation time is smaller than that of back propagation method. © 1999, The Japan Society of Mechanical Engineers. All rights reserved.

DOI： 10.1299/kikaic.65.4048

Scopus

researchmap

Language：Japanese   Publisher：Japan Society of Mechanical Engineers  

A direct application of conventional numerical procedures to practical problems requires huge number of boundary elements. To overcome this difficulty a new effective method is developed for predicting galvanic corrosion rate for complicated structures. In this method, the region to be analyzed is divided into a target region and accompanying regions. The effect of accompanying regions to the target region is reduced analytically to the boundary condition of the target region. An example for a sea water pump is presented to demonstrate the practical applicability of the proposed methods.

DOI： 10.1299/kikaia.65.1493

Scopus

researchmap

DOI： 10.2208/journalam.2.177

researchmap

researchmap

researchmap

DOI： 10.2208/journalam.2.177

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

DOI： 10.2464/jilm.49.452

researchmap

researchmap

researchmap

DOI： 10.1299/kikaic.65.4329

researchmap

DOI： 10.2472/jsms.48.1316

researchmap

DOI： 10.1299/kikaic.65.4048

researchmap

DOI： 10.1299/kikaic.64.3088

researchmap

DOI： 10.3323/jcorr1991.47.454

researchmap

DOI： 10.1299/kikaic.64.3088

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, an effective boundary element method was developed for predicting galvanic corrosion rate for complicated structures. A direct application of conventional numerical procedures to this practical problem causes an increasing of boundary elements. Firstly, a rational criterion for an application of axis-symmetrical elements was proposed. Secondly, a method which analyzes only focused region of a structure was developed. By using these proposed methods, the number of boundary elements is decreased. A example for practical sea water pump is presented to demonstrate the practical applicability of the proposed methods.

DOI： 10.1299/kikaia.64.1761

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00054485365?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

The distribution of contact pressure on the surfaces of a crack in a ceramic specimen is estimated by a 3D finite element inverse analysis. The strains generated by the introduction of a crack are measured with strain gages on a side of a specimen, and this datum is used as the input to the inverse analysis. To reduce the number of parameters to be determined, the microscopic deformation characteristics related to the surface roughness are modeled by a non-linear spring with two parameters. After identifying these two parameters, the contact pressure distribution is calculated by a direct analysis. In order to save the computational time, the effective method which makes the system of equations small is developed. This method is applicable to a wide range of inverse problems.

DOI： 10.1299/kikaia.64.1176

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00053324467?from=CiNii

Language：Japanese   Publisher：Japan Society of Corrosion Engineering  

The authors have developed 2D, 3D and axisymmetric boundary-element-method (BEM) programs, and a BEM system based on these programs, in order to quantitatively estimate cathodic protection and macro-cell corrosion such as galvanic corrosion and differential-aeration-cell-corrosion. Because the knowledge of physical quantities (potential and current density) on the surface of corroding materials is of prime importance in corrosion problems, the authors have developed the BEM, which does not require discretization with internal elements. Whereas, conventional analysis techniques, e.g. finite-difference-method (FDM) and finite-element-method (FEM) require internal elements discretization which brings difficulties on the analysis of complicated 3D regions. In this paper, the mechanism of galvanic corrosion and cathodic protection, and the importance of the numerical analysis technique for the predictions of these corrosion problems are described firstly. Next, the history of development of the FDM, FEM and BEM are reviewed briefly. Lastly, the BEM and the BEM system are explained, and the usefulness of the BEM system is demonstrated by showing the application examples for corrosion protection design of seawater pumps.

DOI： 10.3323/jcorr1991.47.156

Scopus

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new simple method for identifying the parameters in the constitutive equation for an elastic/visco-plastic material is proposed. In the proposed method, the time-histories of the input and output stresses in the split Hopkinson's bar test are measured, and a finite element calculation is performed using the measured time history of the input stress. The identification employing the Kalman filter is carried out by comparing the calculated results with the measured time history of the output stress. To demonstrate the usefulness of the proposed method, a split Hopkinson's bar test is performed using copper specimens, and two parameters characterizing the visco-plasticity are identified.

DOI： 10.1299/kikaia.64.445

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00051636067?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A system which reconstructs 3D human movement from a stream of 2D input images is developed. The system finds the 3D shape and motion of the human body by fitting a simple skeleton model to the notable regions found in the 2D image. Surface markers which robust for obstruction were applied. The fitting is done using a nonlinear optimization approach. This optimization is essentially ill-conditioned because depth information can not be obtained from 2D image. In order to overcome this problem, Kalman filter using Fazzy reasoning which takes account of following a priori information is applied: 1 The versatile ekeleton model exhibiting a human body. 2 Postures at each keyframe for the type of human movement. We reconstructed human movement from a stream of Video images by this method in order to demonstrate its applicability.

DOI： 10.1299/kikaic.64.3088

CiNii Books

researchmap

DOI： 10.3323/jcorr1991.47.454

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A system which reconstructs 3D human movement from a stream of 2D input images is developed. The system finds the 3D shape and motion of the human body by fitting a simple skeleton model to the notable regions found in the 2D image. Surface markers which robust for obstruction were applied. The fitting is done using a nonlinear optimization approach. This optimization is essentially ill-conditioned because depth information can not be obtained from 2D image. In order to overcome this problem, Kalman filter using Fazzy reasoning which takes account of following a priori information is applied: 1 The versatile ekeleton model exhibiting a human body. 2 Postures at each keyframe for the type of human movement. We reconstructed human movement from a stream of Video images by this method in order to demonstrate its applicability.

DOI： 10.1299/kikaic.64.3088

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, an effective boundary element method was developed for predicting galvanic corrosion rate for complicated structures. A direct application of conventional numerical procedures to this practical problem causes an increasing of boundary elements. Firstly, a rational criterion for an application of axis-symmetrical elements was proposed. Secondly, a method which analyzes only focused region of a structure was developed. By using these proposed methods, the number of boundary elements is decreased. A example for practical sea water pump is presented to demonstrate the practical applicability of the proposed methods.

DOI： 10.1299/kikaia.64.1761

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00054485365?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

The distribution of contact pressure on the surfaces of a crack in a ceramic specimen is estimated by a 3D finite element inverse analysis. The strains generated by the introduction of a crack are measured with strain gages on a side of a specimen, and this datum is used as the input to the inverse analysis. To reduce the number of parameters to be determined, the microscopic deformation characteristics related to the surface roughness are modeled by a non-linear spring with two parameters. After identifying these two parameters, the contact pressure distribution is calculated by a direct analysis. In order to save the computational time, the effective method which makes the system of equations small is developed. This method is applicable to a wide range of inverse problems.

DOI： 10.1299/kikaia.64.1176

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00053324467?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new simple method for identifying the parameters in the constitutive equation for an elastic/visco-plastic material is proposed. In the proposed method, the time-histories of the input and output stresses in the split Hopkinson's bar test are measured, and a finite element calculation is performed using the measured time history of the input stress. The identification employing the Kalman filter is carried out by comparing the calculated results with the measured time history of the output stress. To demonstrate the usefulness of the proposed method, a split Hopkinson's bar test is performed using copper specimens, and two parameters characterizing the visco-plasticity are identified.

DOI： 10.1299/kikaia.64.445

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00051636067?from=CiNii

DOI： 10.1299/kikaic.63.1167

researchmap

DOI： 10.1299/kikaia.63.1280

researchmap

Language：Japanese   Publisher：Japan Society of Corrosion Engineering  

In order to make a practical application of boundary element method (BEM) to quantative predictions of cathodic protection and macro-cell corrosion, a BEM analysis system based on the programs for 2D, 3D and axi-symmetric regions was developed. The system consists of a polarization curve database and programs to perform element discretization, input file set up, BE analysis and graphic display of input and output data. Even an engineer with no knowledge of corrosion will be able to easily operate this system and perform effective analyses for optional 3 D regions. The usefulness of the system is demonstrated by an application example for the cathodic protection design of a seawater pump. In the system, the effect of flow-rate and time on the polarization curve can be taken into account for defining the boundary conditions.

DOI： 10.3323/jcorr1991.46.378

Scopus

researchmap

DOI： 10.1299/kikaia.63.589

researchmap

Language：Japanese   Publisher：Society of Materials Science Japan  

A method for improving accuracy of measuring polarization characteristics is proposed. In this method, a boundary element inverse analysis is used to eliminate the error due to the non-uniform distribution of current density on a specimen. In the inverse analysis an appropriate objective function, which is to be minimized, is proposed. A measurement of polarization characteristics of a steel on soil is made and the true characteristics is obtained by using this proposed method.

DOI： 10.2472/jsms.46.176

Scopus

researchmap

Language：Japanese   Publisher：Society of Materials Science Japan  

A new boundary element method for analyzing the galvanic corrosion of a long underground structure such as a pipeline and an oil well casing was developed. A long underground structure is often buried in soil with non-uniform electric conductivity. In case where the conductivity change is distinct, the conventional multi-region method is useful, while it gives an unrealistic solution with discontinuous jumps for gradual change of conductivity. To overcome the difficulty, the fundamental solution for a field with linear change of conductivity was introduced by using the Fourier transform technique, and the polarization characteristics were formulated to express its continuous change. A few example problems were solved with this method to demonstrate its applicability and usefulness.

DOI： 10.2472/jsms.46.170

Scopus

researchmap

DOI： 10.1299/kikaic.63.1167

researchmap

DOI： 10.1299/kikaic.63.1167

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

An effective boundary element method with high accuracy for an axisymmetrical electromagnetio field of a layered plate is developed. Quasi-static axisymmetrical Maxwell equations are represented by a vector potential which satisfies the coulomb gauge, and discretized using a standard BEM. Analytical relationships between the magnetic field and the vector potential of an axisymmetrical electromagnetic field are derived for two cases : (1) a layered plate, and (2) a plate whose near-surface conductivity varies as a hyperbolic tangent function of depth. For case (1), the relationship is represented by products of the transfer matrices. For case (2), the relationship is represented by a hypergeometric function. The discrete Hankel transform theory is applied to these analytical relationships, and the result is combined with the standard BEM. The distinct advantage of this method is that discretization is not necessary in a layered plate. It is found that the length of the element that combines the standard BEM with the analytical relationship has a lower limit in this method, because the analytical relationships can be considered as a low-pass filter which relates the magnetic field to the vector potential on the upper surface of a layered plate. The accuracy of this method is demonstrated by solving some example problems.

DOI： 10.1299/kikaia.63.1280

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00047121928?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, an effective boundary element method with high accuracy for an axisymmetrical electromagnetlc field of a layered tube is developed. Quasi-static axisymmetrical Maxwell equations are represented by a vector potential satisfying Coulomb gauge and discretized with a standard BEM. The analytieal relationship between a magnetic field and a vector potential of an axisymmetrical eiectromagnetic field of a layered tube is derived using modified Bessel functions I₁ (z) and K₁ (z) and represented by products of transfer matrices. The discrete Fourier transform theory is applied to the analytical relationship, and the result is combined with the standard BEM. The distinct advantage of this method is that the discretization is not necessary in a layered tube. It is found that the length of the element which combines the standard BEM with the analytical relationship has upper and lower bounds in this method, because the discrete Fourier transform theory is based on the Shannon sampling theoren and a low-pass filter relates a magnetic field and a vector potential on the inner surface of a layered tube. The accuracy of this method is demonstrated by solving some example problems.

DOI： 10.1299/kikaia.63.589

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00045970627?from=CiNii

DOI： 10.2472/jsms.46.170

researchmap

DOI： 10.1299/kikaic.63.1167

researchmap

DOI： 10.1299/kikaia.62.2344

researchmap

DOI： 10.1299/kikaia.62.1741

researchmap

DOI： 10.1299/kikaia.62.2344

researchmap

DOI： 10.1299/kikaia.62.1741

researchmap

Language：Japanese   Publisher：Society of Materials Science Japan  

The influence of parameters in the Gurson's constitutive equation for porous plastic materials on ductile fracture at an interface crack under mixed mode loading is discussed based on a comparison of computational and experimental results. A simple method to estimate the parameters using the deformation data of tensile specimens and the Kalman filter is proposed. A finite element simulation is performed to demonstrate the usefulness of the proposal method. It is suggested that two different specimens (e. g., one is a rectangular plate with a center hole and the other is a side notched plate) are necessary to guarantee a correct estimation.

DOI： 10.2472/jsms.45.305

Scopus

researchmap

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new method for estimating the nonuniform distribution of moisture content in solids is developed using inverse analysis. Since this inverse problem is ill-conditioned, direct application of the conventional method for solving inverse problems results in an oscillatory solution. To avoid this difficulty, the nature of the material, that the distribution of moisture content is governed by the diffusion equation, is taken into account, and the number of unknown parameters is reduced. The multiple-domain boundary element method is employed for analyzing the electric field within the solid. Some example problems are solved using this method in order to demonstrate the accuracy of the estimation.

DOI： 10.1299/kikaia.62.2344

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00042808909?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new method for optimizing the cathodic protection of a ship in operation is proposed. First, from electric potential data measured using several sensors located on the wall of the ship, the polarization curve of the painted ship wall is estimated by solving an inverse problem using the boundary element method. Then, using the polarization curve, the optimum current to be supplied to each electrode (several electrodes are located on the ship wall) is determined in such a way that the electric potential of every part of the ship wall becomes lower than a critical value using minimum the necessary electric power. In case when the electric potentials at the sensors change abruptly due to paint damage, the location of the damage is estimated from the potential change by solving another inverse problem, and then the cathodic protection is optimized. An example problem is solved to demonstrate the applicability of the proposed method.

DOI： 10.1299/kikaia.62.1741

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00041037270?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new method for estimating the nonuniform distribution of moisture content in solids is developed using inverse analysis. Since this inverse problem is ill-conditioned, direct application of the conventional method for solving inverse problems results in an oscillatory solution. To avoid this difficulty, the nature of the material, that the distribution of moisture content is governed by the diffusion equation, is taken into account, and the number of unknown parameters is reduced. The multiple-domain boundary element method is employed for analyzing the electric field within the solid. Some example problems are solved using this method in order to demonstrate the accuracy of the estimation.

DOI： 10.1299/kikaia.62.2344

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00042808909?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new method for optimizing the cathodic protection of a ship in operation is proposed. First, from electric potential data measured using several sensors located on the wall of the ship, the polarization curve of the painted ship wall is estimated by solving an inverse problem using the boundary element method. Then, using the polarization curve, the optimum current to be supplied to each electrode (several electrodes are located on the ship wall) is determined in such a way that the electric potential of every part of the ship wall becomes lower than a critical value using minimum the necessary electric power. In case when the electric potentials at the sensors change abruptly due to paint damage, the location of the damage is estimated from the potential change by solving another inverse problem, and then the cathodic protection is optimized. An example problem is solved to demonstrate the applicability of the proposed method.

DOI： 10.1299/kikaia.62.1741

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00041037270?from=CiNii

DOI： 10.2472/jsms.45.305

researchmap

researchmap

DOI： 10.2472/jsms.44.238

researchmap

DOI： 10.1299/kikaia.61.1651

researchmap

DOI： 10.1299/kikaia.61.1651

researchmap

researchmap

DOI： 10.1299/kikaia.61.1841

researchmap

researchmap

DOI： 10.1299/kikaia.61.1841

researchmap

DOI： 10.1299/kikaia.61.1651

researchmap

Language：English  

The Scanning Vibrating Electrode Technique is an effective and non-destructive technique to evaluate the amount of pitting or intergranular corrosion by measuring the electropotential field near the surface of metal sample submerged in an electrolyte. However, the accurate determination of correct electropotential field is difficult, when the sensor probe approaches the metal surface and disturbes the field by itself. In order to overcome this difficulty, an inverse problem, in which the true current density distribution along the metal surface is estimated from the data having errors due to the disturbance, was solved in this paper. The 3-D boundary element method was used to deal with the governing Laplace equation. The ill condition of the inverse problem was overcome by proposing the following procedures. (1) A form of the polarization curve is presumed and the number of unknown parameters is reduced. Then the current density distribution along the metal surface is determined. (2) In order to improve the accuracy, additional electropotential is applied to the sample by a potentiostat and various data are obtained for different electropotential fields. B using the above procedures, the numerical simulation of a sample specimen was carried out. The computed results showed a significant improvement in measuring accuracy by using the inverse analysis. © 1995, The Society of Materials Science, Japan. All rights reserved.

DOI： 10.2472/jsms.44.238

Scopus

researchmap

DOI： 10.1299/kikaia.61.1651

researchmap

DOI： 10.1299/kikaia.61.1651

researchmap

DOI： 10.2472/jsms.44.238

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

An electromagnetic field of an eddy current thickness sensor was analyzed for the purpose of improving measurement accuracy. Maxwell equations were discretized with the boundary element method. In order to facilitate the calculation, a quasi-stationary approximation and an axisymmetric one were employed. A numerical integration formula with high accuracy was applied to evaluate a singular integral. A special element was developed for effective analysis of a thin film. It is found from numerical results that the radial component of magnetic flux density produced by the eddy current is greater than the axial component in nickel-plated steel. Using this property, a new thickness sensor for a metal film on a metal substrate was realized.

DOI： 10.1299/kikaia.61.1841

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00034879870?from=CiNii

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

An electromagnetic field of an eddy current thickness sensor was analyzed for the purpose of improving measurement accuracy. Maxwell equations were discretized with the boundary element method. In order to facilitate the calculation, a quasi-stationary approximation and an axisymmetric one were employed. A numerical integration formula with high accuracy was applied to evaluate a singular integral. A special element was developed for effective analysis of a thin film. It is found from numerical results that the radial component of magnetic flux density produced by the eddy current is greater than the axial component in nickel-plated steel. Using this property, a new thickness sensor for a metal film on a metal substrate was realized.

DOI： 10.1299/kikaia.61.1841

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00034879870?from=CiNii

DOI： 10.1299/kikaia.61.1651

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

There are many optimization problems in which the objective function has local minima. If we apply gradient methods to such problems, we must repeat many calculations with different initial points. Furthermore, it is not guaranteed that the solution is a true global minimum, unless the initial points are numerous and widely scattered. Recently, genetic algorithm (GA) has been receiving strong interest as a tool for solving such problems. It is a population-and generation-based probabilistic search strategy. The global minimum can be found with higher probability by GA than by deterministic methods. However, it takes a long time for convergence. To overcome those difficulties, an effective method is developed by combining GA and the clustering technique. This method consists of the following procedures. (1) Perform GA up to an appropriate generation. There is no need for calculation until convergence is obtained. (2) Apply the clustering technique to the results of (1) and determine the cluster centroids. (3) Find the minima by certain gradient methods starting with the cluster centroids obtained in (2), and take the smallest result as the optimal one. Two sample problems are solved to demonstrate the applicability of the proposed method.

DOI： 10.1299/kikaia.60.2903

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00030504059?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

The acceleration and deceleration behaviors of a semi-infinite crack in an elastic plate with a finite width, which is subjected to time-dependent loading, are analyzed and the caustic method for measuring the fast crack propagation toughness K_ID is simulated by using the boundary element method. It is shown that the "measured" K_ID depends on the shape of the specimen when the dynamic stress intensity factor K_I ( t, v) and/or the crack speed v changes rapidly. The dependence is explained by the numerical results that the K_I(t, v) -controlled field does not develop enough to cover the location of the initial curve of caustics, i. e., the displacement field near the initial curve is influenced not only by K_I (t, v) but also by (1) the superposition of stress wave reflected at the boundary and (2) the change of the crack acceleration due to the reflected stress wave

DOI： 10.1299/kikaia.60.2532

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00029625451?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new method to quickly determine weights in a layered neural network is introduced. By making use of the relation between the input of one unit in the (k+1) th layer and the output of all the units in the K th layer, all the weights can be obtained easily provided that the number of units in at least one layer is greater than or equal to the number of teacher sets plus 1. From two examples, it is shown that calculation time required by the new method is much shorter than that of back propagation method.

DOI： 10.1299/kikaia.60.574

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00025114676?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new method to quickly determine weights in a layered neural network is introduced. By making use of the relation between the input of one unit in the (k+1) th layer and the output of all the units in the K th layer, all the weights can be obtained easily provided that the number of units in at least one layer is greater than or equal to the number of teacher sets plus 1. From two examples, it is shown that calculation time required by the new method is much shorter than that of back propagation method.

DOI： 10.1299/kikaia.60.574

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00025114676?from=CiNii

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

There are many optimization problems in which the objective function has local minima. If we apply gradient methods to such problems, we must repeat many calculations with different initial points. Furthermore, it is not guaranteed that the solution is a true global minimum, unless the initial points are numerous and widely scattered. Recently, genetic algorithm (GA) has been receiving strong interest as a tool for solving such problems. It is a population-and generation-based probabilistic search strategy. The global minimum can be found with higher probability by GA than by deterministic methods. However, it takes a long time for convergence. To overcome those difficulties, an effective method is developed by combining GA and the clustering technique. This method consists of the following procedures. (1) Perform GA up to an appropriate generation. There is no need for calculation until convergence is obtained. (2) Apply the clustering technique to the results of (1) and determine the cluster centroids. (3) Find the minima by certain gradient methods starting with the cluster centroids obtained in (2), and take the smallest result as the optimal one. Two sample problems are solved to demonstrate the applicability of the proposed method.

DOI： 10.1299/kikaia.60.2903

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00030504059?from=CiNii

Language：English  

A new effective three-dimensional boundary element method of analyzing electric field produced by localized corrosions has been developed to estimate the accuracy of SVET (Scanning Vibrating Electrode Technique). With decreasing height of the electrode probe from the surface of a sample in an attempt to improve the resolution of SVET, the electrode probe disturbs the electric field and the measuring error of the current distribution increases. It is, therefore, desirable that this error is corrected by numerical analysis considering the disturbance of the electrode probe. However the scale ratio of the whole vessel of the instrument to the measurement area is larger than a factor of 105, so that the conventional BEM has a difficulty in treating such a huge number of elements. In this study, at first the distribution of current density on the slender surface of a sample where current flows out was analyzed with the two-dimensional BEM and then a three-dimensional Boundary Element analysis was performed by considering the slender surface as a simple line-shape element which has the characteristics obtained by the first 2-D analysis. By using the proposed method, the number of elements and repetitive calculations due to scanning of electrode probe can be decreased. A few numerical examples are presented to demonstrate the practical applicability of the proposed method. © 1994, The Society of Materials Science, Japan. All rights reserved.

DOI： 10.2472/jsms.43.1628

Scopus

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

The acceleration and deceleration behaviors of a semi-infinite crack in an elastic plate with a finite width, which is subjected to time-dependent loading, are analyzed and the caustic method for measuring the fast crack propagation toughness K_ID is simulated by using the boundary element method. It is shown that the "measured" K_ID depends on the shape of the specimen when the dynamic stress intensity factor K_I ( t, v) and/or the crack speed v changes rapidly. The dependence is explained by the numerical results that the K_I(t, v) -controlled field does not develop enough to cover the location of the initial curve of caustics, i. e., the displacement field near the initial curve is influenced not only by K_I (t, v) but also by (1) the superposition of stress wave reflected at the boundary and (2) the change of the crack acceleration due to the reflected stress wave

DOI： 10.1299/kikaia.60.2532

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00029625451?from=CiNii

DOI： 10.1299/kikaia.60.574

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A new method to quickly determine weights in a layered neural network is introduced. By making use of the relation between the input of one unit in the (k+1) th layer and the output of all the units in the K th layer, all the weights can be obtained easily provided that the number of units in at least one layer is greater than or equal to the number of teacher sets plus 1. From two examples, it is shown that calculation time required by the new method is much shorter than that of back propagation method.

DOI： 10.1299/kikaia.60.574

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00025114676?from=CiNii

researchmap

Language：English  

A new effective three-dimensional boundary element method is developed for predicting the galvanic corrosion rate of heat exchanger. Since a heat exchanger consists of a huge number of slender tubes and a brass tube plate direct application of conventional numerical procedures needs so many elements that computation becomes impossible. In the newly proposed method, the surface of the tube plate (which is connected to many tubes) is regarded as a surface of one virtual metal. At first, the polarization curve of the virtual metal is obtained by analyzing a tube unit which consists of one tube and a part of tube plate. Then the side water room is analyzed with a rather coarse mesh by employing the polarization curve of the virtual metal. Finally, the potential and current density thus calculated are used to determine the detailed distributions of potential and current density on the surfaces of an arbitrary tube and the tube plate. A real heat exchanger is analyzed by the proposed method to demonstrate its practical applicability. This method may be applicable not only to a corrosion problem of a heat exchanger, but also to heat conduction problems of similar structures. © 1993, The Society of Materials Science, Japan. All rights reserved.

DOI： 10.2472/jsms.42.849

Scopus

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A method for estimating the galvanic corrosion rate was developed. In the case where the electrochemical polarization curves of metal in the electrolyte considered are available, the galvanic corrosion rate can be estimated by solving us Laplace equation with boundary conditions based on the polarization curves. The polarization curves, however, are not always available. In this case, it is necessary to estimate the density of the current across the metal surfaces from the potential values measured at some points far from the metal surfaces. This inverse problem was formulated by employing the boundary element method. Since the system of linear equations obtained was ill-conditioned, direct application of conventional numerical procedures to the equation system results in an oscillatory solution. In this paper, at first the method which estimates the relationship between potential and current density was applied. In order to improve the accuracy, a new method using apriori information which as expressed as fuzzy membership functions was developed. A sample problem was solved by the new method in order to demonstrate the accuracy.

DOI： 10.1299/kikaia.59.1562

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00021442735?from=CiNii

DOI： 10.2472/jsms.42.849

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A method for estimating the galvanic corrosion rate was developed. In the case where the electrochemical polarization curves of metal in the electrolyte considered are available, the galvanic corrosion rate can be estimated by solving us Laplace equation with boundary conditions based on the polarization curves. The polarization curves, however, are not always available. In this case, it is necessary to estimate the density of the current across the metal surfaces from the potential values measured at some points far from the metal surfaces. This inverse problem was formulated by employing the boundary element method. Since the system of linear equations obtained was ill-conditioned, direct application of conventional numerical procedures to the equation system results in an oscillatory solution. In this paper, at first the method which estimates the relationship between potential and current density was applied. In order to improve the accuracy, a new method using apriori information which as expressed as fuzzy membership functions was developed. A sample problem was solved by the new method in order to demonstrate the accuracy.

DOI： 10.1299/kikaia.59.1562

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00021442735?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

An elastodynamic boundary element analysis of a fast crack propagation test is performed. A propagating crack subject to a time-dependent uniform pressure on its surfaces is considered. The crack speed is determined in such a way that the dynamic stress intensity factor K₁(t, v) is equal to the fast crack propagation toughness K_ID. Caustics in reflection are generated by making the assumption of plane stress. The time variation of the "measured" K_ID is obtained from the diameter of the generated caustics. The K_ID thus obtained is found to depend on crack propagation history due to the transient nature of the elastic stress field near a propagating crack tip. This may explain to some extent the physical meaning of the experimental data showing the dependence of K_ID on crack acceleration.

DOI： 10.1299/kikaia.58.1650

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00017071741?from=CiNii

DOI： 10.1299/kikaia.58.1508

researchmap

DOI： 10.1299/kikaia.58.1234

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

An elastodynamic boundary element analysis of a fast crack propagation test is performed. A propagating crack subject to a time-dependent uniform pressure on its surfaces is considered. The crack speed is determined in such a way that the dynamic stress intensity factor K₁(t, v) is equal to the fast crack propagation toughness K_ID. Caustics in reflection are generated by making the assumption of plane stress. The time variation of the "measured" K_ID is obtained from the diameter of the generated caustics. The K_ID thus obtained is found to depend on crack propagation history due to the transient nature of the elastic stress field near a propagating crack tip. This may explain to some extent the physical meaning of the experimental data showing the dependence of K_ID on crack acceleration.

DOI： 10.1299/kikaia.58.1650

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00017071741?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

An experiment to examine the possibility of identification of coating defect was performed on a model of pipe-line in sea water. The pipe-line model which is made of SS steel pipe is set in resin made vessel which was filled with 0.001% NaC1 solution. A small current was impressed to the model. In order to keep the temperature of the solution at 30±0.5°C, the vessel was placed in an isothermal chamber filled with water. The distribution of the potential on the surface of the solution was measured using an electrometer with the glass capillary tip. A platinum counter electrode was used with a reference electrode ( SCE : Saturated Calomel Electrode ). The Poisson's equation, which describes the potential in the electrolyte, was so1ved by the boundary element method. The detection was performed by minimizing the difference between actual and calculated values of potential. By using the data of the above experiment, locatons of coating defects were succesfully detected, and the corrosion rates were correctly estimated.

DOI： 10.1299/kikaia.58.1508

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00016548565?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

An effective three-dimensional boundary element method was developed for predicting electrogalvanic field response of a structure which consists of slender parts. First, the corrosion field which has slender surfaces is analyzed by the boundary element method with simple line-shape elements. By this method, the axial distributions of potential and current density of slender parts can be obtained, but not circumferential distributions. Then, the cross section of the slender part considered is discretized with two-dimensional elements, and the circumferential distributions of potential and current density are obtained, using the values which were obtained with the simple line-shape elements. By using the proposed method, the CPU time is decreased. A few numerical examples are presented to demonstrate the practical applicability of the proposed method.

DOI： 10.1299/kikaia.58.1234

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00016230668?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This paper presents a boundary element application to determine the optimum impressed current densities and optimum location of electrode in a cathodic protection system. The potential within the electrolyte is described by Poisson's equation with nonlinear boundary conditions which are enforced based on experimentally determined electrochemical polarization curves. The optimal impressed current densities are determined in order to minimize the power supply for protection under the protecting conditions that the electric potential of every part of the structure to be protected should be less than some critical value. The solution is obtained by using the conjugate gradient method in which the protecting conditions are taken into account by the penalty function method. The boundary element method is employed to discretize the governing equations. Several numerical examples are presented to demonstrate the practical applicability of the pro-posed method.

DOI： 10.1299/kikaia.57.1431

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00009884823?from=CiNii

Language：English  

A three-dimensional boundary element method was developed for optimizing the locations and impressed currents of electrodes in a cathodic protection system. The electrodes were regarded as the sources of current, and the potential in the electrolyte was described by the Poisson’s equation with the boundary condition, in which the polarization of metal to be protected was taken into account. The Poisson’s equation was solved by the boundary element method, and the optimization was performed by minimizing the power necessary to keep the potential on the metal surface below a critical value. An effective method is proposed for obtaining the derivatives of potential, which are needed in the optimizing procedure. In order to demonstrate the usefulness of the method, some example problems are presented. © 1991, The Society of Materials Science, Japan. All rights reserved.

DOI： 10.2472/jsms.40.1390

Scopus

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In many cathodic protection systems, the electrodes must be located directly on the wall of a structure for various reasons. An effective boundary element method for determining the optimum impressed current and optimum location of the electrode in such a cathodic protection system is proposed in this paper. This problem is reduced to minimize the power supply under the protecting condition that the electric potential of every part of a structure to be protected should be less than some critical value. In order to avoid the remeshing which is necessary in each iteration in the usual optimization procedures, the boundary integral equation is modified in such a way that the design variables become independent of the boundary element mesh. The solution is obtained by using the conjugate gradient method in which the protecting conditions are taken into account by the penalty function method. A few numerical examples are presented to demonstrate the practical applicability of the proposed method.

DOI： 10.1299/kikaia.57.1465

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00010348031?from=CiNii

DOI： 10.1299/kikaia.57.1431

researchmap

DOI： 10.2472/jsms.40.1390

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In many cathodic protection systems, the electrodes must be located directly on the wall of a structure for various reasons. An effective boundary element method for determining the optimum impressed current and optimum location of the electrode in such a cathodic protection system is proposed in this paper. This problem is reduced to minimize the power supply under the protecting condition that the electric potential of every part of a structure to be protected should be less than some critical value. In order to avoid the remeshing which is necessary in each iteration in the usual optimization procedures, the boundary integral equation is modified in such a way that the design variables become independent of the boundary element mesh. The solution is obtained by using the conjugate gradient method in which the protecting conditions are taken into account by the penalty function method. A few numerical examples are presented to demonstrate the practical applicability of the proposed method.

DOI： 10.1299/kikaia.57.1465

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00010348031?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This paper presents a boundary element application to determine the optimal impressed current densities in a cathodic protection system. In this system, enough current must be impressed to lower the potential distribution on the metal surface to the critical values. The potential within the electrolyte is described by the Laplace's equation with nonlinear boundary conditions which are enforced based on experimentally determined electrochemical polarization curves. The optimal impressed current densities are determined in order to minimize the power supply for protection. The solution is obtained by using the conjugate gradient method in which the governing equations and the protecting conditions are taken into account by the penalty function method. The boundary element method is employed to descretize the governing equations. Several numerical examples are presented to demonstrate the practical applicability of the proposed method.

DOI： 10.1299/kikaia.56.1282

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00002871936?from=CiNii

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This paper presents a boundary element application to determine the optimal impressed current densities in a cathodic protection system. In this system, enough current must be impressed to lower the potential distribution on the metal surface to the critical values. The potential within the electrolyte is described by the Laplace's equation with nonlinear boundary conditions which are enforced based on experimentally determined electrochemical polarization curves. The optimal impressed current densities are determined in order to minimize the power supply for protection. The solution is obtained by using the conjugate gradient method in which the governing equations and the protecting conditions are taken into account by the penalty function method. The boundary element method is employed to descretize the governing equations. Several numerical examples are presented to demonstrate the practical applicability of the proposed method.

DOI： 10.1299/kikaia.56.1282

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00002871936?from=CiNii

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap