Hybrid of Support Vector Regression, Genetic Algorithm, and Bat Optimization Algorithm Integrated with ANN for Short-Term Load Forecasting

Abstract

Accurate short-term load forecasting (STLF) is crucial for effective energy management and grid stability. This study introduces a hybrid approach that combines Support Vector Regression (SVR), Genetic Algorithm (GA), and Bat Optimization Algorithm (BA) with an Artificial Neural Network (ANN) to enhance the accuracy of STLF. The integration of machine learning techniques and optimisation algorithms in the proposed model improves forecasting precision and computational efficiency compared to traditional models. In this method, SVR acts as the initial predictor, with GA optimizing the SVR parameters and BA further refining the ANN's weights and biases. A comprehensive evaluation is conducted using real-world data – obtained from the National Control Centre of the Transmission Company of Nigeria, demonstrating the superior performance of the hybrid approach over conventional methods and existing hybrid models. The experimental results demonstrate that the proposed hybrid model significantly outperforms traditional forecasting methods in terms of both accuracy and computational efficiency. The results indicate that the proposed model can provide more accurate and reliable short-term load forecasts, contributing to better decision-making in power system operation and this makes it a promising tool for modern energy management systems.