A regional hybrid GOA-SVM model based on similar day approach for short-term load forecasting in Assam, India

Abstract

In today's restructuring electricity market, short-term load forecasting (STLF) is an essential tool for the electricity utilities to predict future scenario and act towards a profitable policy. The electric load demand is highly influenced by the thermal inertia due to the climatic factors. These influential climatic factors are different in different regions. Therefore, it is necessary to have a region specific STLF model for load forecasting under regional climatic conditions. This paper proposes a regional hybrid STLF model utilizing SVM with a new technique, called grasshopper optimization algorithm (GOA), to evaluate its suitable parameters. This study is carried out in Assam, a state of India and proposed GOA-SVM model is targeted for forecasting the load under local climatic conditions. The proposed model uses the similar day approach (SDA) to satisfy the regional climatic requirements. The results of the proposed model show better accuracy comparing to results generated with classical STLF model of incorporating temperature universally as the only climatic factor. To further affirm the efficacy of the proposed model, same inputs are delivered in two alternative hybrid models, namely GA-SVM (GA with SVM) and PSO-SVM (PSO with SVM). The results indicate that the proposed model outperforms the other hybrid models.