科研之友
微信
新浪微博
Facebook
分享链接
摘要
With the awareness of global environmental protection, the generation of photovoltaic (PV) power is increasing rapidly. Local voltage/Var control is widely studied to overcome the real-time change in PV power generation. The stability condition of local controls is strictly dependent on the system information, and the control parameter is chosen considering both the stability condition and dynamic performance, which limits its scalability and makes the parameter setting difficult. In this work, the coupling between node voltages is separated in the prediction time, and a local voltage predictive control strategy is proposed to reduce the voltage deviation with respect to the reference voltage. The proposed control is implemented by using historical local data and current local measurements. The control parameter is the data length, which has less correlation with the system information, and the control performance is less sensitive to the control parameter. The proposed control can be regarded as a conventional local control with dynamic compensation, which improves the dynamic performance and the steady-state error. The proposed local control is tested on a practical distribution system. The simulation result demonstrates the proposed control achieves better performance than the conventional local control under different application scenarios and weather conditions.
