<![CDATA[Purpose: Short-term electricity load forecasting (STLF) plays an important role in power system management because electricity demand is dynamic and influenced by factors such as community activities, weather conditions, and seasonal patterns. However, conventional forecasting methods often have limitations in modeling nonlinear and fluctuating electricity load data. Therefore, this study aims to develop a more accurate and stable forecasting model by integrating artificial intelligence methods within a Graphical User Interface (GUI)-based system. Methods: This research proposes a hybrid model combining Genetic Algorithm (GA), Adaptive Momentum Backpropagation (AMBP), and Support Vector Regression (SVR). GA is used to optimize SVR parameters, SVR performs nonlinear regression forecasting, and AMBP improves learning stability. The dataset consists of electricity load data from Gunung Sari District, Lombok, collected during 2015–2024 with 3,650 daily samples, divided into 80% training data and 20% testing data. Model performance was evaluated using MSE, RMSE, and MAPE. Result: The experimental results show that the proposed GA–SVR–AMBP hybrid model achieves better forecasting performance than single and partial hybrid models. In the testing phase, the model produced an MSE of 0.1556, RMSE of 0.3945, and MAPE of 1.13%, with an accuracy of 98.86%. Using the entire dataset, the model achieved an MSE of 0.4213, RMSE of 0.6491, and MAPE of 1.6953% with an accuracy of 98.30%, indicating good generalization capability and low prediction error. Novelty: The novelty of this study lies in the development of a hybrid GA–SVR–AMBP forecasting model integrated into a MATLAB-based GUI system that facilitates data analysis, model execution, and visualization of prediction results for short-term electricity load forecasting and decision support in power system management.]]>
