<![CDATA[The continuous growth in electricity consumption demands a reliable prediction system to support sustainable energy planning. This study aims to forecast electricity consumption using the Support Vector Machine (SVM) method, in which the parameters are optimized through the Particle Swarm Optimization (PSO) algorithm. PSO is employed to determine the optimal parameters, namely weight and bias, by minimizing prediction errors measured using Mean Squared Error (MSE). The implementation was carried out using Visual Basic for Applications (VBA) in Excel, based on historical electricity usage data from January 2023 to July 2025 The prediction results indicate a high level of accuracy, as evidenced by the lowest Mean Absolute Percentage Error (MAPE) value of 0.006, observed in the Mbt District.. The model was then used to predict electricity consumption until December 2027, revealing a gradual increase in usage across all districts. These findings indicate that the integration of SVM and PSO is effective in producing accurate and reliable prediction models to support decision-making in electricity demand management.]]>
Copyrights © 2025
