Design of an integrated forecasting and scheduling model for power plants to balance solar and wind energy variability using real-time weather data

Abstract

The integration of variable renewable energy sources such as solar and wind creates challenges for power system stability and operational scheduling due to their intermittent characteristics. This study proposes an integrated forecasting and scheduling framework using real-time weather data for a hybrid renewable power system consisting of photovoltaic, wind, geothermal, and hydropower plants. Solar irradiance and wind speed data were collected using pyranometer and anemometer sensors and modeled using ARIMA for 24-hour-ahead forecasting. Based on AIC and BIC evaluation, ARIMA (2, 1, 2) and ARIMA (1, 1, 1) were selected for solar irradiance and wind speed forecasting, respectively. The forecasting results achieved MAPE values of 18.43% for solar irradiance and 14.12% for wind speed. The forecasted renewable outputs were integrated into a generation scheduling model, where geothermal power operated as a base-load unit and hydropower acted as a balancing source. The proposed scheduling strategy was evaluated through a 24-hour Newton-Raphson load flow simulation. Results showed that system power losses remained below 2% and bus voltage levels were maintained within acceptable limits, demonstrating reliable operation under fluctuating weather conditions.