Forecasting Industrial Electricity Consumption in Iran: A Novel Hybrid Approach for Sustainable Energy Management

Accurately predicting industrial electricity consumption is essential for optimizing energy efficiency, and reducing costs in industrial operations. This study presents a novel hybrid prediction model based on radial basis function neural network (RBFNN) and kernelized support vector regression (KSVR) methods (RBFNN-KSVR) for estimating industrial electricity consumption. Key input variables include population, electricity price in the industry sector, gross domestic product (GDP), and the number of electricity subscribers. The proposed hybrid model was implemented in a real-world case study to estimate industrial electricity consumption in Iran and compared against base methods (RBFNN, SVR, and KSVR). Extensive evaluation reveals the superior performance of the RBFNN-KSVR model in predicting industrial electricity consumption. This study provides a robust and reliable approach for industrial stakeholders to enhance energy planning, identify energy-saving opportunities, and ensure a stable power supply. The findings have significant implications for optimizing energy usage, improving efficiency, and reducing costs in industrial operations.