Hybrid Modified Particle Swarm Optimization and Adaptive Neuro-Fuzzy Inference System for Electric Power Consumption Prediction

Abstract

Load forecasting is crucial in power system energy management, aiding planning and operations. The latest modified forecasting methods are important for best prediction of the electric power consumption of the load which nowadays are not practically implemented. As a result, hybrid modified particle swarm optimization (MPSO)–adaptive neuro-fuzzy inference system (ANFIS) is applied in this work to predict the electric power consumption of the case study area. This research also focuses on using artificial neural network (ANN), ANFIS, and MPSO-ANFIS to model Hossana City’s power consumption in Ethiopia, forecasting beyond available data for the next decade. Addressing the challenges of complex and widely distributed power systems, the study incorporates demographic factors. The performance evaluations show MPSO-ANFIS found as the superior model, achieving a mean absolute percentage error (MAPE) of 1.2% and root mean square error (RMSE) of 0.87, due to MPSO. This model also boosts a high R2 value of 0.97. The model demonstrates effective load forecasting based on previous data training, promising reliable future predictions. So, the hybrid MPSO-ANFIS is recommended for electric power consumption prediction.