A new short-term load forecast method based on neuro-evolutionary algorithm and chaotic feature selection

• A new hybrid neuro-evolutionary algorithm method is proposed for short-term load forecasting.
• Neural network’s weights and biases are best tuned by using Differential Evolutionary (DE) algorithm.
• A new chaotic feature selection is proposed for designing input vector.
• Phase space reconstruction under Taken’s embedding theorem is used in preparing candidate features.

In competitive environment of deregulated electricity market, short-term load forecasting (STLF) is a major discussion for efficient operation of power systems. Therefore, the area of electricity load forecasting is still essential need for more accurate and stable load forecast algorithm. However, the electricity load is a non-linear signal with high degree of volatility. In this paper, a new forecasted method based on neural network (NN) and chaotic intelligent feature selection is presented. The proposed feature selection method selects the best set of candidate input which is used as input data for the forecasted. The theory of phase space reconstruction under Taken’s embedding theorem is used to prepare candidate features. Then, candidate inputs relevance to target value are measured by using correlation analysis. Forecast engine is a multilayer perception layer (MLP) NN with hybrid Levenberg–Marquardt (LM) and Differential Evolutionary (DE) learning algorithm. The proposed STLF is tested on PJM and New England electricity markets and compared with some of recent STLF techniques.