Optimization of Artificial Neural Networks Based on Chaotic Time Series in Power Load Forecasting Model

Abstract

According to the chaotic and non-linear characters of power load data, the model of artificial neural networks ANN based on Lyapunov exponents was established. The time series matrix was established according to the theory of phase-space reconstruction, and then Lyapunov exponents was computed to determine time delay and embedding dimension. Then artificial neural networks algorithm was used to predict power load. In order to prove the rationality of chosen dimension, another two random dimensions and BP algorithm singly were selected to compare with the calculated dimension. The results show that the model which has been chosen is effective and highly accurate in the forecasting of short-term power load.