Multi-Period Electrical Array Reconfiguration for Multi-PV Systems Considering Net Power Fluctuation Mitigation

The electrical array reconfiguration (EAR) approach has become a promising solution to enhance photovoltaic (PV) system performance under partial shading conditions. Existing studies focus on maximizing single-period PV generation but neglect the impact of power fluctuation on grid stability. To address this, we propose a multi-period EAR method for multi-PV systems that considers smoothing net power fluctuations. First, we design a multi-period EAR model to maximize total revenue by balancing electricity sales and net power fluctuation penalties, formulated as a stochastic mixed-integer quadratic programming problem. The model incorporates constraints on the average number of switching actions per unit time to ensure practical implementation. Then, to handle the unpredictability of partial shading conditions, we develop a Lyapunov optimization-based online algorithm to decouple the time-coupling constraints involving state transitions. Additionally, we propose a reduced set of EAR strategies to improve computational efficiency. Numerical studies demonstrate that the proposed method significantly reduces net power fluctuations in distribution networks with high PV penetration and enhances total revenue compared to conventional methods.