NDT IDENTIFICATION OF A CRACK USING ANNS WITH STOCHASTIC GRADIENT DESCENT

被引：6

作者：

ARKADAN, AA ^{[1
]}

CHEN, Y ^{[1
]}

SUBRAMANIAM, S ^{[1
]}

HOOLE, SRH ^{[1
]}

机构：

[1] HARVEY MUDD COLL,DEPT ENGN,CLAREMOUNT,CA 91711

来源：

IEEE TRANSACTIONS ON MAGNETICS | 1995年 / 31卷 / 03期

关键词：

D O I：

10.1109/20.376431

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 [电气工程]; 0809 [电子科学与技术];

摘要：

Nondestructive testing (NDT) is used to identify the anomalies and defects in inaccessible locations. Various techniques of optimization are used in NDT. In this work, the Artificial Neural Networks (ANNs) are applied with NDT to identify a crack in a conducting medium. In general, deterministic techniques are used with the back propagation algorithm (BP) to train the neural networks. The ANNs which are trained by a deterministic method have a tendency to get trapped in local minima. In this paper a stochastic version of the gradient descent is applied to train the ANNs and it overcomes the difficulties of local minima caused by the sinusoidal fields. The stochastic version used in this approach is based on the Metropolis algorithm which is frequently used in the simulated annealing.

引用

页码：1984 / 1987

页数：4

共 12 条

[1]

Dobbeni D., The Role of Eddy Currents and ACPD in Effective Inspection, Improving the Effectiveness and Reliability of Non-Destructive Testing, pp. 47-86, (1992)

[2]

Palanisamy R., Lord W., Finite Element Modeling of Electromagnetic NDT Phenomena, IEEE Trans. Magn., MAG-15, 6, pp. 1479-1481, (1979)

[3]

Hoole S.R.H., Subramaniam S., Saldanha R., Coulomb J.L., Sabonnadiere J.C., Inverse Problem Methodology and Finite Elements in the Identification of Cracks, Sources, Materials, and their Geometry in Inaccessiable Locations, IEEE Trans. on Magn., 27, 3, pp. 3433-3443, (1991)

[4]

Subramaniam S., Kanaganathan S., Hoole S.R.H., Two Requisite Tools in the Optimal Design of Electromagnetic Devices, IEEE Trans. Magn., 27, 5, pp. 4105-4109, (1991)

[5]

Hoole S.R.H., Weeber K., Subramaniam S., Fictitious Minima of Object Functions, Finite Element Meshes, and Edge Elements in Electromagnetic Device Synthesis, IEEE Trans. Magn., 27, 6, pp. 5214-5216, (1991)

[6]

Arkadan A.A., Sareen T., Subramaniam S., Genetic Algorithms for Nondestructive Testing in Crack Identification, Presented at MMM-INTERMAG'94, and accepted for publication in th IEEE Trans. Magn., (1994)

[7]

Koh C.S., Mohammed O.A., Hahn S., Detection of Magnetic body using Artificial Neural Network with Modified Simulated Annealing, IEEE Trans. Magn., 30, 5, pp. 3644-3647, (1994)

[8]

Metropolis N.A., Rosenbluth A., Rosenbluth M., Teller A., Teller E., Equation of State Calculations by Fast Computing Machines, Jour. Chem. Phys., 21, pp. 1087-1092, (1953)

[9]

Aarts E., Korst J., Simulated Annealing and Boltzmann Machines, (1989)

[10]

Chen Y., Artificial Neural Networks for the Inverse Electromagnetic Field Problem, (1994)

← 1 2 →