Adaptive neuro fuzzy selection of the most important factors for photovoltaic pumping system performance prediction

Abstract

The main aim of the study was to perform a selection procedure to determine the most important factors for the photovoltaic pumping system performance prediction. As the input factors, photovoltaic modules loss, water control counter loss, pump loss and pipe system loss were used along. As the output factor performance ratio of the photovoltaic pumping system was used. The performance ratio of photovoltaic pumping system is an important parameter for quality measurement of the system. There is a need to analyze which factors have the most influence on the performance ratio in order to perform satiable calibration in order to reduce losses of the system. Adaptive neuro fuzzy inference system (ANFIS) was used to determine factors’ influence on the performance ration of the system. Root mean square error (RMSE) was used to assets the influences. Based on the RMSE values loss of photovoltaic modules has the strongest impact on the performance ratio of the system. The obtained results could be of practical usage for improving performance ration of the photovoltaic pumping system.