Distributed Reactive Power Optimization Control Research Of Active Distribution System

In recent years,the world’s strong support for the development of renewable energy causes the increasing penetration of distributed generation in traditional distribution network and the development of energy storage,micro-gas turbine,demand side of response further enhance the controllability of traditional distribution network,which has gradually evolved into active distribution system to promote the utilization of renewable energy and increase the flexibility,reliability and economy of the power supply.On the other hand,the integration of large-scale distributed generation makes the centralized optimization model difficult to meet requirements of control flexibility and power supply reliability.As for the problem of distributed reactive power optimization for active power distribution system,a distributed cooperative control architecture with features of centralized mode in areas and distributed mode among areas was established and the distributed reactive power optimization method for active distribution system respectively based on auxiliary problem principle algorithm and based on equal incremental rate of network loss was proposed to realize the distributed optimization of global network loss.The main contents are as follows:(1)Distributed reactive power optimization theory for active distribution systemFirst,reactive power output characteristics of the wind power generation system and the photovoltaic power generation system were analyzed and the centralized reactive power optimization model based on the second-order cone relaxation technology was established for active distribution system.Secondly,advantages and disadvantages of centralized optimization mode and distributed optimization mode were compared and analyzed.A distributed cooperative control architecture with features of centralized mode in areas and distributed mode among areas was established and two kinds of distributed reactive power optimization methods in this paper were compared.Finally,a partition method based on hierarchical clustering was presented for the foundation of active distribution system distributed optimization.(2)Distributed reactive power optimization method for active distribution system based on auxiliary problem principleA distributed reactive power optimization control method based on auxiliary problem principle algorithm for active distribution system was presented to achieve the distributed optimization of global network loss.Moreover,border variable standardized method was proposed to improve the convergence rate.Based on the proposed control method,global active power loss optimization can be realized in distributed manner and the global voltage quality was guaranteed,while the individual partition only needs gathering the power and voltage information and exchanges border variable information with neighborhood areas.Finally,the effectiveness of the proposed control method was verified via numerical simulation results based on the IEEE 69 nodes.(3)Distributed reactive power optimization method for active distribution system based on equal incremental rate of network lossIn order to further improve the convergence speed of distributed optimization method,based on equal incremental rate of network loss,a distributed reactive power optimization control method comprising of distributed reactive power optimization model without consideration of power and voltage constraints and incremental rate of network loss adjustment method for processing power and voltage constraints,was proposed to realize the global optimization of network loss and ensure the system voltage level and safe operation.The simulation results based on IEEE33 nodes and 69 nodes show that the proposed control method has well convergence speed and global optimization effect and can effectively deal with communication failure to ensure the reliability and economy of power supply,especially for active distribution system with high penetration of distributed generation and high number of partition.