Golden Eagle Extreme Learning Machine for Hourly Solar Irradiance Forecasting
Sarunyoo Boriratrit, Rongrit Chatthaworn · 2022 International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME) · 2022
Nowadays, the photovoltaic system which is one of the renewable energy sources has an important role in generating electricity. To forecast photovoltaic system generation, the time-series information of solar irradiance should be considered for implementation with the computational model. Machine learning is a highly successful model that can forecast information accurately. However, the matter of machine learning model can cause the instability of forecasting; therefore, the forecasting result is unstable, which cannot apply in real forecasting electricity. To reduce the cause of instability, this paper proposes a novel machine learning model which applied the golden eagle optimization combined with the extreme learning machine model. The experiment results showed that the proposed model's minimum root mean square error was achieved at 0.0791, which was better than the comparative models.