Artificial neural network optimisation methodology for the estimation of the critical flashover voltage on insulators

To describe an artificial neural network (ANN) methodology in order to estimate the critical flashover voltage on polluted insulators is the objective here. The methodology uses as input variables characteristics of the insulator such as diameter, height, creepage distance, form factor and equivalent salt deposit density, and it estimates the critical flashover voltage based on an ANN. For each ANN training algorithm, an optimisation process is conducted regarding the values of crucial parameters such as the number of neurons and so on using the training set. The success of each algorithm in estimating the critical flashover voltage is measured by the correlation index between the experimental and estimated values for the evaluation set, and finally the ANN with the correlation index closest to 1 is specified. For this ANN and the respective algorithm, the critical flashover voltage of the test set insulators is estimated and the respective confidence intervals are calculated through the re-sampling method.

[1]  P. Dixit,et al.  ANN based three stage classification of arc gradient of contaminated porcelain insulators , 2004, Proceedings of the 2004 IEEE International Conference on Solid Dielectrics, 2004. ICSD 2004..

[2]  C. Tourreil,et al.  Aging of composite insulators: simulation by electrical tests , 1990 .

[3]  Martin A. Riedmiller,et al.  A direct adaptive method for faster backpropagation learning: the RPROP algorithm , 1993, IEEE International Conference on Neural Networks.

[4]  Sivaji Chakravorti,et al.  Estimation of time-to-flashover characteristics of contaminated electrolytic surfaces using a neural network , 1995 .

[5]  Roberto Battiti,et al.  First- and Second-Order Methods for Learning: Between Steepest Descent and Newton's Method , 1992, Neural Computation.

[6]  M. Sanaye-Pasand,et al.  Prediction of leakage current of composite insulators in salt fog test using neural network , 2005, CEIDP '05. 2005 Annual Report Conference on Electrical Insulation and Dielectric Phenomena, 2005..

[7]  D. W. Auckland,et al.  Neural networks to analyze surface tracking on solid insulators , 1997 .

[8]  Frangiskos V. Topalis,et al.  Genetic algorithm approach to the modelling of polluted insulators , 2002 .

[9]  Z. Aydogmus,et al.  A new flashover dynamic model of polluted HV insulators , 2004, IEEE Transactions on Dielectrics and Electrical Insulation.

[10]  Martin Fodslette Møller,et al.  A scaled conjugate gradient algorithm for fast supervised learning , 1993, Neural Networks.

[11]  B. Engstrom,et al.  Pollution tests for coastal conditions on an 800 kV composite bushing , 2003 .

[12]  Kenneth Levenberg.  A METHOD FOR THE SOLUTION OF CERTAIN NON – LINEAR PROBLEMS IN LEAST SQUARES , 1944 .

[13]  George J. Tsekouras,et al.  Design of an artificial neural network for the estimation of the flashover voltage on insulators , 2007 .

[14]  Duan Li,et al.  On Restart Procedures for the Conjugate Gradient Method , 2004, Numerical Algorithms.

[15]  D. Marquardt.  An Algorithm for Least-Squares Estimation of Nonlinear Parameters , 1963 .

[16]  M. K. Soni,et al.  Artificial Neural Network-Based Peak Load Forecasting Using Conjugate Gradient Methods , 2002, IEEE Power Engineering Review.

[17]  Vincent Del Toro,et al.  Electric Power Systems , 1991 .

[18]  Laurent Krähenbühl,et al.  Computation of electric fields and potential on polluted insulators using a boundary element method , 1992 .

[19]  R. S. Gorur,et al.  Computer-aided design of porcelain insulators under polluted conditions , 1995 .

[20]  N. Chatterjee,et al.  Estimation of time-to-flashover characteristics of contaminated electrolytic surfaces using a neural netlwork , 1995, IEEE Transactions on Dielectrics and Electrical Insulation.

[21]  F. Topalis,et al.  Dielectric behaviour of polluted porcelain insulators , 2001 .

[22]  Gill G. Richards,et al.  Use of pollution monitors with a neural network to predict insulator flashover , 1997 .

[23]  S. Ghosh,et al.  Modeling PD inception voltage of epoxy resin post insulators using an adaptive neural network , 1999 .

[24]  C. M. Reeves,et al.  Function minimization by conjugate gradients , 1964, Comput. J..

[25]  Ruobing Zhang,et al.  Calculation of DC and AC flashover voltage of polluted insulators , 1990 .

[26]  L. S. Moulin,et al.  Confidence intervals for neural network based short-term load forecasting , 2000 .

[27]  S. Hyakin,et al.  Neural Networks: A Comprehensive Foundation , 1994 .

[28]  M. Sanaye-Pasand,et al.  A neural network based method for leakage current prediction of polymeric insulators , 2006, IEEE Transactions on Power Delivery.