Research of Nonlinear Combination Forecasting Model for Insulators ESDD Based on Wavelet Neural Network

Wavelet neural network (WNN) effectively overcomes the intrinsic shortcomings of artificial neural network, namely, slow learning speed, difficult determination of structure and existence of local minima, combination forecasting model fully integrates the information of each model, and nonlinear one effectually conquers the difficulties and drawbacks in combined modeling non-stationary time serial by using linear model, hence, nonlinear combination forecasting model based on WNN possesses more flexible structure, higher data fitting and forecasting accuracy. Simulation experiments show, compared with other forecasting models, the predicted results are closer to the actual ones which show the nonlinear combination forecasting model for insulators ESDD based on WNN can efficaciously improve the speed and accuracy of the forecasting. Therefore, the method presented provides a doable thought for the computerization of pollution area map of power network.