Application of artificial neural networks and adaptive neuro-fuzzy inference system to estimate the energy generation of a solar power plant in Ain Beni-Mathar (Morocco)

Currently, concentrated solar power (CSP) plant has gained large importance due to its good environmental aspect. However, the process of the CSP power production presents important variability, as it relied on meteorological conditions, which gives new challenges to power system operators. To handle these challenges, it is significant to be able to anticipate and observe production levels. The present paper presents a comparison between two artificial intelligence models artificial neural networks (ANN) and adaptive neurofuzzy inference system (ANFIS), to estimate the daily electric energy generation of a solar power plant in Ain Beni-Mathar (34°00'35”N, 2°01 `29”W) in Eastern Morocco. Six input variables (daily direct normal irradiation, day of the month, daily mean ambient temperature, mean wind velocity, relative humidity and previous daily electric production) served as the inputs of the ANN and ANFIS models. The results showed that both the ANN and ANFIS models performed similarly in terms of prediction accuracy.