A neural network-based optimization approach for induction motor design

This paper proposes a new approach, using artificial neural networks (ANNs), to optimize a set of design parameters of induction motors. The training patterns for the ANNs can be generated from a finite element method, an expert system or an experienced design engineer. The ANN will be trained to learn the relations governing the input and output of an electrical machine. Once the training process of the ANN is completed, the proposed ANN-based optimization approach can be utilized to provide a set of optimized design parameters for a given set of specifications and desired constraints. The results provided by this approach were presented and compared with a conventional optimization method. These results clearly demonstrated the effectiveness of the proposed approach as an optimization tool in electrical machine design.

[1]  J. Simkin,et al.  Optimization problems in electromagnetics , 1991 .

[2]  Doo-Jin Park,et al.  Optimum design of single-sided linear induction motor using the neural networks and finite element method , 1993, Proceedings of 1993 International Conference on Neural Networks (IJCNN-93-Nagoya, Japan).

[3]  M. K. Haldar,et al.  Optimization of electromagnetic devices: circuit models, neural networks and gradient methods in concert , 1995 .

[4]  Gordon R. Slemon,et al.  AN improved method of optimization for electrical machines , 1991 .

[5]  D. C. Park,et al.  Design optimization of electromagnetic devices using artificial neural networks , 1992, [1993] Proceedings of the Second International Forum on Applications of Neural Networks to Power Systems.

[6]  Robert A. Jacobs,et al.  Increased rates of convergence through learning rate adaptation , 1987, Neural Networks.

[7]  Leon O. Chua,et al.  Neural networks for nonlinear programming , 1988 .

[8]  A. Ray.  Magnetic Circuit Design of Saturated Electrical Machines By the Finite-Element Method , 1981, IEEE Transactions on Power Apparatus and Systems.

[9]  F. G. Uler,et al.  Design optimization of electromagnetic devices using artificial neural networks , 1992, 1992. Digests of Intermag. International Magnetics Conference.

[10]  Stefen Hui,et al.  On solving constrained optimization problems with neural networks: a penalty method approach , 1993, IEEE Trans. Neural Networks.

[11]  S.R.H. Hoole,et al.  Artificial neural networks in the solution of inverse electromagnetic field problems , 1993 .

[12]  Soo-Young Lee,et al.  A Self-Organizing Neural Network Approach for Automatic Mesh Generation , 1990 .

[13]  E. F. Fuchs,et al.  Comparison of Two Optimization Techniques as Applied to Three-Phase Induction Motor Design , 1989, IEEE Power Engineering Review.

[14]  Osama A. Mohammed,et al.  Utilizing Hopfield neural networks and an improved simulated annealing procedure for design optimization of electromagnetic devices , 1993 .

[15]  N.H. Fetih,et al.  Induction Motor Optimum Design, Including Active Power Loss Effect , 1986, IEEE Transactions on Energy Conversion.

[16]  B. G. Desai,et al.  A Comparative Study of Minimization Techniques for Optimization of Induction Motor Design , 1973 .

[17]  Austin H. Bonnett.  Understanding Efficiency in Squirrel-Cage Induction Motors , 1980, IEEE Transactions on Industry Applications.

[18]  J. Appelbaum,et al.  Optimization of Three-Phase Induction Motor Design. Part II: The Efficiency and Cost of an Optimal Design , 1987, IEEE Power Engineering Review.

[19]  Lee Soo-Young,et al.  A self-organizing neural network approach for automatic mesh generation , 1991 .