Reverse Overload Mitigation for Rural Low-Voltage Distribution Areas Based on PV Power Forecasting and Energy Storage Operation Optimization

With the advancement of new energy technologies and a progressive increase in new energy penetration, the quantity and installed capacity of distributed photovoltaic (PV) power generation in rural low-voltage distribution areas are increasing. Excessive PV power generation readily leads to reverse overloading of the distribution transformers, affecting the power supply quality and transformer longevity. Energy storage systems can regulate energy’s spatial and temporal distribution and have become a significant means of addressing this issue. However, the energy storage scheduling strategy must consider economic factors while mitigating reverse overload. Consequently, this paper initially proposes the XGBoost-Informer-based forecasting method to enhance the forecasting accuracy of PV and loads, thereby obtaining PV forecasting data for the subsequent day. Subsequently, based on the adaptive particle swarm algorithm, an optimization model of the storage charging and discharging power is established to minimize the degrees of reverse overload and ensure the economic viability of the storage operation. Finally, the model was implemented and simulation experiments were conducted for verification. Simulation results show that the proposed method has high prediction accuracy and can effectively reduce the hazardous degree of reverse heavy overload while ensuring the economic benefits of energy storage operation.