Digital Twin-Based Framework for Dynamic Stability Analysis of Grid-Tied Photovoltaic System

Abstract

This paper presents a Digital Twin (DT) framework for real-time stability assessment of grid-tied photovoltaic (PV) systems under varying operating conditions. The proposed DT operates in parallel with the physical hardware, serving as a high-fidelity mathematical model that continuously synchronizes with sensor measurements from the actual system to support control and monitoring. A Rank-based Adaptive Artificial Electric Field Optimization (RAAEFO) algorithm is employed to update DT parameters, ensuring accurate alignment with hardware behavior. For system control, an Adaptive Robust Cascaded Second-Order Generalized Integrator (AR-CSOGI) based Unit Vector Integrated Chattering-Free Enhanced Sliding Mode Controller (UVI-CFESMC) is developed to achieve precise current tracking and robust performance under disturbances. The framework is implemented using an OPAL-RT (OP4512) platform for the DT and an NI sbRIO-9636 FPGA for the hardware prototype. Experimental evaluations under scenarios including voltage sag and solar irradiance variations demonstrate that the DT reliably monitors system performance and predicts stability margins. Stability assessment is performed using Lyapunov theory, incorporating both controller and system stability indicators. Results confirm that the proposed DT framework improves monitoring accuracy and ensures stable operation across dynamic conditions.