Design of Active Smart Sensor Based on an Adaptive Neuro-Fuzzy Inference System (ANFIS) for Detecting the Sun’s Position in Photovoltaics Based on Solar Tracker

The LDR (Light Dependent Resistor) sensor is an important electronic component used in active solar trackers. This study evaluated three LDR sensor placement schemes, and the optimization of LDR sensor placement based on the Adaptive Neuro-Fuzzy Inference System (ANFIS) was performed. The inputs for the ANFIS method include solar radiation, light intensity, panel temperature, voltage, and current generated by solar panels, and the output is the power generated by solar panels. Based on the optimization results obtained from ANFIS, a solar tracker with an active smart sensor was designed, featuring a switching scheme between the two best variations to optimize energy collection. The two best variations used in the switching scheme are Scheme 1, where sensors are placed on each side of the PV, and Scheme 3, where sensors are divided into two groups, each consisting of two sensors placed on opposite sides. Scheme 1 produced MSE, RMSE, and MAE values of 0.071, 0.266, and 0.087, respectively, while Scheme 3 produced 0.271, 0.521, and 0.126. The net energy produced by solar panels in schemes 1, 2, 3, the switching scheme, and the fixed scheme were 1165.29 Wh, 1233.74 Wh, 1222.63 Wh, 1313.78 Wh, and 996.43 Wh, respectively. Active smart sensor solar tracker showed an energy performance improvement of 12.742%, 6.487%, 7.455%, and 31.848% over Schemes 1, 2, 3, and the fixed scheme, respectively.