역전파 신경망을 이용한 고전력 반도체 소자 모델링

Modeling High Power Semiconductor Device Using Backpropagation Neural Network

  • 발행 : 2003.05.01

초록

Using a backpropagation neural network (BPNN), a high power semiconductor device was empirically modeled. The device modeled is a n-LDMOSFET and its electrical characteristics were measured with a HP4156A and a Tektronix curve tracer 370A. The drain-source current $(I_{DS})$ was measured over the drain-source voltage $(V_{DS})$ ranging between 1 V to 200 V at each gate-source voltage $(V_{GS}).$ For each $V_{GS},$ the BPNN was trained with 100 training data, and the trained model was tested with another 100 test data not pertaining to the training data. The prediction accuracy of each $V_{GS}$ model was optimized as a function of training factors, including training tolerance, number of hidden neurons, initial weight distribution, and two gradients of activation functions. Predictions from optimized models were highly consistent with actual measurements.

키워드

참고문헌

  1. PSPICE 8.0 Reference Manual, ORCAD-Microsim Corporation, 1997
  2. I. Budiharjo and P. Lauritzen, 'Performance requirements for power MOSFET models', IEEE Transactions on Power Electronics, vol. 12, no. 1, pp. 36-45, 1997 https://doi.org/10.1109/63.554167
  3. H. Yee and P. Lauritzen , 'SPICE models for power MOSFET- an update', Proceedings of IEEE APEC Conf., pp. 281-289, 1988 https://doi.org/10.1109/APEC.1988.10575
  4. J. Sen and P. Palmar, 'A SPICE model for the accurate simulation of a powr MOSFET during switching', Proceedings of EPE'91 Conf., pp. 286-288, 1991
  5. S. Malounyans, 'SPICE computer model for HEXFET power MOSFETs', International Rectifiers, Application note AN 9751, 1991
  6. I. Budiharjo and P. Lauritzen, 'The lumped charge power MOSFET model, including parameters extraction', IEEE Transactions on Power Electronics, vol. 10, no. 3, pp. 379-387, 1995 https://doi.org/10.1109/63.388005
  7. C. Xu, D. Schroder, 'A unified model for the power mOSFET including the inverse diode and the parasitic bipolar transistor', Proceedings of EPE'89, pp. 139, 1989
  8. A. Maxim and G. Maxim, 'A high accuracy power MOSFET SPICE behavioral macromodel including the device self-heating and safe operating area simulation', Proceedings of 14th APEC, vol. 1, pp. 177-183, 1999 https://doi.org/10.1109/APEC.1999.749507
  9. P. M. Watson, M. Weatherspoon, L. Dunleavy, and G. L. Creech, 'Accurate and efficient small-signal modeling of active deivices using artificial neural networks', Proceedings of IEEE Gallium Arsenide Integrated Circuit Symposium, pp. 95-98, 1998 https://doi.org/10.1109/GAAS.1998.722636
  10. A. H. Zaabab, Q. J. Zhang, M. S. Nakhla, 'Device and circuit-level modeling using neural networks with faster training based on network sparsity', IEEE Transactions on Microwavetheory and Techniques, vol. 45, no. 10, pp. 1696-1704, 1997 https://doi.org/10.1109/22.641714
  11. S. Bila, Y. Harkouss, M. Ibrahim, J. Rousset, E. N'Goya, D. Baillargeat, S. Verdeyme, M. Aubourg, and P. Guillon, 'An accurate wavelet neural-network-based model for electromagnetic optimization of microwave circuits', Internat'I J. RF & Microwave computer-Aided Eng., vol. 9, no. 3, pp. 297, 1999 https://doi.org/10.1002/(SICI)1099-047X(199905)9:3<297::AID-MMCE13>3.0.CO;2-W
  12. D. E. Rummelhart and J. L. McClelland, Parallel Distributed Processing, Cambridfe, MIT Press, 1986
  13. B. Kim and S. Park, 'An optimal neural networks plamsa model: a case study', Chemometrics and Intelligent Laboratory Systems, vol. 56, no. 1, pp. 39-50, 2001 https://doi.org/10.1016/S0169-7439(01)00107-1
  14. T.M Roh, D.W. Lee, J. Kim, S.G. Kim, Q.S. Song, J.Y. Kang, J.G. Koo, K.S. Nam, and K.I. Cho, 'High-voltage SOI power IC technology with non-RESURF n-LDMOSFET and RESURE p-LDMOSFET for PDP scan-driver application', Journal of Korean Physical Societry, vol 37, No. 6, pp. 889-892, 2000 https://doi.org/10.3938/jkps.37.889