Decentralized Model Predictive Voltage Control for Distributed Generation Clusters of Power Systems

With the extensive usage of distributed resources, the voltage fluctuations in the power system are frequent, and the distributed resources are scattered, which also increases the difficulty of unified management. A decentralized model predictive voltage control method for the power system based on the collaboration of clusters is proposed. The large-scale decentralized resources are aggregated into interrelated clusters, and the voltage control can be adjusted regionally. Firstly, because of the complex structure of the power system and the wide range of distributed power points, a power grid cluster division scheme based on the improved Louvain algorithm is designed. The modularity function is used and the voltage sensitivity of the cluster and the capacity of distributed power supply are taken into account. Secondly, because the power flow of the grid becomes more complex and variable after the access of distributed power supply, a decentralized model predictive control method is proposed based the division of clusters, which can cooperate with the regulation ability of each optical storage to suppress the voltage fluctuation of the power grid caused by the power fluctuation. Finally, a study case is given to verify the effectiveness of the proposed method.