Research on PV Access Control Method Based on Risk Sharing and Co-Optimization

The current problems of voltage overruns and limited power transfer caused by a high percentage of photovoltaic (PV) access to the grid have revealed the inadequacy of existing control methods. For example, traditional generation limitation measures can maintain security but cause clean energy waste, energy storage scheduling is costly, probabilistic trend analysis can only assess uncertainty but lacks economic incentives, and robust optimization methods are often too conservative. In this paper, we propose a new framework for PV access control based on economic risk sharing and cooperative optimization to address the above deficiencies. The framework utilizes the Shapley value to equitably allocate the risk responsibilities of all parties, introduces market mechanisms such as overload option contracts and output insurance to hedge the grid overload risk, adopts the CVaR (Conditional Value-at-Risk) to measure the tail risk, and combines the LSTM deep learning model to improve the accuracy of the PV output prediction, and constructs the upper and lower layers of the co-optimization model. The proposed method quantifies the technical risk of PV access into economic cost, and maximizes PV power consumption under ensuring grid security through two-layer optimization. Simulation results show that the framework significantly reduces the extreme risk level caused by PV output uncertainty, improves voltage stability and energy utilization; compared with the traditional method, the annual <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$P V$</tex> abandonment rate is significantly reduced, the PV revenue improvement ratio is nearly 57%, and the total system cost is further reduced, which verifies the effectiveness and economy of the method.