Intelligent Dual-Axis Solar Tracking System Using PLC-SCADA Integrated with an Adaptive Neuro-Fuzzy Predictive Control Algorithm for Maximized PV Energy Harvesting

Abstract

This research proposes an Intelligent Dual-Axis Solar Tracking System that integrates automation and supervision via SCADA and Programmable Logic Controllers (PLCs) with an Adaptive Neuro-Fuzzy Predictive Control (ANFPC) algorithm. The aim is to maximize the energy captured by solar panels by adjusting their real-time orientation based on environmental conditions. As irradiance varies, traditional LDR-based or astronomically driven tracking systems often exhibit a latent response, mechanical lag, and reduced efficiency. The proposed solution combines adaptive neuro-fuzzy predictive modeling with the predictive reliability of PLC control to eliminate these weaknesses. The ANFPC method combines the Solar Position method (SPA) with real-time irradiance, panel tilt feedback, variations in azimuth-elevation angles, and past-tracking history to dynamically predict the optimal panel orientation. The tracking error, actuator complexity, and smooth movement in dual-axis are reduced to a minimum when a predictive optimization layer is applied. Whereas SCADA provides continuous monitoring, remote parameter adjustment, system diagnostics, and predictive maintenance alerts, the PLC employs PID-tuned pulse control to implement corrective actions for stepper or servo motors. The smart system outperforms any other LDR-based tracker. It delivers a gain in daily energy production relative to a fixed setup, as experimentally verified on a 200 W dual-axis prototype. The tracker’s self-learning system adjusts to weather changes, temporary clouds, and the gradual degradation of the panel, ensuring its long-term longevity. The combination of PLC-SCADA and the ANFPC algorithm has proven highly reliable, adaptive, and efficient for next-generation solar tracking platforms, as shown by the results. The proposed PLC-SCADA method achieved superior performance with 0.52° tracking error, 92% power efficiency, 2.1 s response time, 15% actuator stress, and ΔW = 5.1 adaptation rate.