The digital twin (DT) technology implementation in photovoltaic (PV) systems provides an innovative approach to real-time performance monitoring and predictive maintenance. In this paper, an end-to-end DT framework for real-time performance analysis, fault detection, and optimization of a 250 W PV system is proposed. A physics-based equation and AI-based prediction hybrid DT model is developed through MATLAB/Simulink, trained from real data acquired by means of a testbed. The DT simulates the dynamic physical PV system behavior and adjusts itself using self-correcting algorithms to enhance precision in prediction and forecast power output at high fidelity. Results indicate that the DT gives the true response of the PV system with very small differences attributable to model approximations and sensor faults, 95% error minimization after compensation, and a root mean square error (RMSE) of 2.8 W, indicating its applicability for real-time monitoring and predictive main-maintenance. The work here focuses on the feasibility of applying DTs towards the autonomous optimization of distributed renewable energy systems.
Al-Qaisi et al. (Mon,) studied this question.