Implementation of Metaheuristic-Optimized Algorithms in Tuning ANFIS for Photovoltaic Power Output Forecasting

Abstrak

As Indonesia accelerates its transition toward renewable energy utilization, the large-scale deployment of Photovoltaic (PV) systems presents challenges regarding grid stability due to the intermittent nature of solar irradiance. To address this issue, the current study proposes a robust forecasting framework that integrates the Adaptive Neuro-Fuzzy Inference System (ANFIS) with metaheuristic optimization algorithms to accurately predict PV power output. This study focuses on optimizing ANFIS parameters to address the nonlinear complexities inherent in local weather data. Comparative experimental analysis demonstrated that the hybrid approach significantly enhanced prediction performance. The ANFIS-Particle Swarm Optimization (ANFIS-PSO) model outperformed other tested configurations, achieving the highest accuracy with a Root Mean Square Error (RMSE) of 0.68. These findings suggest that the ANFIS-PSO architecture is a highly effective tool for reliable power forecasting, thereby supporting the efficient integration of solar energy into Indonesia's electrical grid.

Identifier

• DOI: https://doi.org/10.48084/etasr.18534
• OpenAlex: https://openalex.org/W7202243403