Midterm Demand Prediction of Electrical Power Systems Using a New Hybrid Forecast Technique

Prediction of daily peak load for next month is an important type of medium-term load forecast (MTLF) for electrical power systems, which provides useful information for maintenance scheduling, adequacy assessment, scheduling of fuel supplies and limited energy resources, etc. However, the exclusive characteristics of daily peak load signal, such as its nonstationary, nonlinear and volatile behavior, present a number of challenges for this task. In this paper, a new hybrid forecast engine is proposed for this purpose. The proposed engine has an iterative training mechanism composed of a novel stochastic search technique and Levenberg-Marquardt (LM) learning algorithm. The effectiveness of the proposed forecast strategy is extensively evaluated based on several benchmark datasets.