Ricky Maulana, Syafii, Aulia
Indonesia's transition toward a renewable energy-dominated power grid is progressing to meet increasing energy demands while reducing dependence on fossil fuels. According to the National Energy General Plan, their goal is to have 23% of the energy mix come from renewables by 2025 and 31% by 2050. Accurate forecasting of photovoltaic (PV) power output is crucial to address the intermittent nature of solar energy and ensure grid stability. A seasonal autoregressive integrated moving average (SARIMA) model was developed to estimate day-ahead photovoltaic power output in Padang City, Indonesia. Using NASA solar irradiance data from March 1-31, 2024, the SARIMA(1,0,1)(4,0,3)24 model achieved high accuracy with an NRMSE of 4.19%. To evaluate its performance, a comparative evaluation was conducted between the SARIMA model and two machine learning methods, namely artificial neural network (ANN) and long short-term memory (LSTM), in which SARIMA achieved the lowest forecasting error. These findings indicate that SARIMA remains an effective and interpretable statistical method for short-term PV forecasting, supporting reliable energy planning and power grid operations towards Indonesia's renewable energy goals. © This is an open access article under the CC BY-SA license. https://creativecommons.org/licenses/by-sa/4.0/
Department of Electrical Engineering, Faculty of Engineering, Universitas Andalas, Padang, Indonesia; Department of Electrical Engineering, Faculty of Engineering, Universitas Negeri Padang, Padang, Indonesia
