Research On Intelligent Algorithms For Integrated Reactive Power Optimization Of Active Distribution Network

With the construction and development of distribution automation,there are more and more active devices and controllable devices in distribution network.At the same time,the automation level of distribution network is also constantly improving.The emergence and development of distributed generation(DG)poses new challenges to reactive power optimization in distribution networks.As a supplement to reactive power supply,distributed generation needs to be considered comprehensively in reactive power optimization.The traditional reactive power optimization methods of distribution network can not meet the development of active distribution network,so it is of great significance to study the integrated reactive power optimization of active distribution network.In this paper,active planning,active management and active control of active distribution network are analyzed,and three power flow analysis methods are proposed.On this basis,a more suitable power flow calculation method for active distribution network is proposed,which lays a foundation for reactive power optimization of active distribution network.A comprehensive reactive power optimization model for active distribution network is constructed and quantum krill swarm algorithm is proposed to solve the model.Aiming at the non-convex and non-linear problem of the model,a convex reactive power optimization model for active distribution network is proposed,and the convex model is solved by using the improved grey wolf algorithm.The specific work is as follows:1.Aiming at the characteristics of active distribution network with various DG and variable network structure,three power flow analysis methods are proposed,which are more suitable for the access of active distribution network distributed generators than the traditional methods of power flow calculation and planning optimization,and can better analyze the power flow of active distribution network.On this basis,the steps of reactive power optimization for active distribution network are proposed.2.Aiming at the problem that traditional reactive power optimization algorithm can not adapt to large-scale distributed generation access,a comprehensive reactive power optimization method for active distribution network based on quantum krill swarm is proposed.Firstly,according to the characteristics of active distribution network,a reactive power optimization model of active distribution network based on proportional coefficient is proposed.Secondly,a quantum krill swarm algorithm is proposed to overcome the shortcomings of the krill swarm algorithm,which has many iterations and easily falls into local optimum.Finally,simulation experiments are carried out on the improved IEEE33 node and IEEE69 to verify the effectiveness of the model and algorithm.3.Considering that the reactive power optimization model of active distribution network is non-convex and non-linear,a convex reactive power optimization model of active distribution network is proposed.The convex reactive power optimization model of active distribution network is relaxed by second-order cone relaxation(SOCP)technology.An Improved Grey Wolf algorithm is proposed to optimize the model.The simulation experiments are carried out at IEEE33 and IEEE69 nodes,and compared with the quantum krill swarm optimization algorithm(QKH).The experimental results verify the effectiveness of the proposed algorithm in different situations.