Research On The Link-like Agent Genetic Algorithm In Distribution Power Optimal Configuration

In recent years,Power generation technology with its flexible,efficient,clean and so on has been vigorously advocated and development.Distribution generation and distribution network power supply will be the future trend of the power industry.Introduced the multi-agent and multi-intelligent system.As genetic algorithm easy to early convergence and multi-agent can enhance the coordination of information between the advantages of cooperation,puts forward the idea of grid-based multi-agent genetic algorithm.Based on the testing of two-dimensional space,found the lack of this method.In addition,the algorithm is easy to increase the amount of calculation at the update and not conducive to the application in real-time practical applications.Proposed a link-like agent genetic algorithm which through the combination of link-like multi-agent algorithm and genetic algorithm.Through the comprehensive analysis function test results,the link-like agent genetic algorithm has obvious improvement in convergence accuracy,convergence speed and stability.Introduced the working principle of various distributed power supply and the node type of the grid interface.Introduced the processing methods of the various types of nodes in the power flow calculation of distribution network with distributed power supply.Combined with the distribution network node stratification method,and through the forward push back to find the power distribution network trend.Bulided objective function model that the minimum active loss of the system.Finally,On the IEEE3-node test system demonstrate that using this power flow calculation method to solve the distributed power supply is advantageous.Introduced the mathematic model of DG optimal configuration with active net loss as a single objective function,and the mathematical model of DG is optimized by using the system active power loss and voltage stability as the objective function.By considering the DG location and volume measurement in the case of isolated scenes.Considering the DG optimized in the 24-h load timing characteristics and climatic characteristics of the planning area.So,on the IEEE33 node test system of the first situation and on the PG & E69 node test system of the second situation.The access position,capacity and power factor of DG are in influence of the active loss and network voltage distribution.The Experiments demonstrate that the reasonableaddition of DG can effectively improve the voltage stability of the distribution system and reduce the active loss of the system.In the processing of DG containing distribution network,the link-like agent genetic algorithm has certain correctness and practicability.