Sistem Monitoring Tingkat Pencahayaan Photovoltaic Berbasis Fuzzy Logic

This study designs a 100 Wp Solar Power Plant (PLTS) monitoring system based on Arduino Uno implementing the Mamdani Fuzzy Logic method for real-time performance classification. Hardware testing indicates DC voltage sensor error of 1.45% – 11.18%, ACS712 current sensor error of 2% – 4%, and PZEM-004T AC voltage error of 0.04% - 0.43%. Based on actual data testing (Light 964 lux and DC Voltage 12.4 V), fuzzification yields membership values for High light (0.18) and Very High (0.82), and Very Low voltage (0.04) and Low (0.96). Out of 25 Mamdani rule bases, MIN-MAX inference and aggregation yield a defuzzification numerator of 146,8 and denominator of 2,16, resulting in a crisp Output of 67,96% (74.71 on system reading). This classifies PLTS performance into the "Good" category, keeping the load relay active (ON) with a Fuzzy calculation error of 9,93%.