EVOLUTIONARY PROGRAMMING BASED COMBINED ARTIFICIAL NEURAL NETWORK FOR SHORT TERM LOAD FORECASTING

摘要

This paper presents a new approach for short term load forecasting using Evolutionary Programming based Combined Artificial Neural Network (EPCANN) module. In this paper, a set of neural networks has been trained with different architecture and with different training parameters. The Artificial Neural Networks (ANNs) are trained and tested for the actual load data of Chennai city (India). A method of Optimal Linear Combination is used to combine selected networks to produce better results, rather than using a single best trained ANN. The obtained test results indicate that the proposed approach improves the accuracy of the load forecasting.