PROJECT TITLE :

A Novel Wavelet-Based Ensemble Method for Short-Term Load Forecasting with Hybrid Neural Networks and Feature Selection

ABSTRACT:

During this paper, a brand new ensemble forecasting model for short-term load forecasting (STLF) is proposed based mostly on extreme learning machine (ELM). Four necessary enhancements are used to support the ELM for increased forecasting performance. 1st, a novel wavelet-primarily based ensemble scheme is applied to generate the individual ELM-based forecasters. Second, a hybrid learning algorithm blending ELM and the Levenberg–Marquardt method is proposed to improve the training accuracy of neural networks. Third, a feature choice method primarily based on the conditional mutual data is developed to pick out a compact set of input variables for the forecasting model. Fourth, to realize an correct ensemble forecast, partial least squares regression is used as a combining approach to combination the individual forecasts. Numerical testing shows that proposed methodology will get better forecasting leads to comparison with other commonplace and state-of-the-art ways.