Study On Optimal Configuration Of Power Quality Monitoring Based On Topology Partition And NSGA-Ⅱ

With the expansion of the electric energy usage and the development of power electronics technology,a large number of impacts,non-linear loads and unbalanced loads are infiltrated into the distribution network.At the same time,for the development and popularization of new power generation methods and energy storage methods,high-voltage direct current transmission technology,and new types of control methods,a large number of non-linear devices are connected to the grid.These devices and loads bring harmonics to the grid,but also cause other power quality problems.With the rapid development of China’s industrial science and technology,the loss caused by power quality is becoming more and more serious.On the other hand,many fields,including finance,communications,data centers,semiconductors,precision processing,electronics and modern service companies,place high demands on power quality.Monitoring power quality is the premise for analyzing and solving power quality problems.It is also the basis for power companies and power users to assume responsibility for power quality problems according to the requirements of the agreement.Therefore,it is of great significance to strengthen the monitoring of power quality of power grids.The Power Quality Monitor(PQM)can effectively monitor the power quality of the grid.However,the current price of the PQM is generally high.When the network topology is known,the PQM can be properly configured to monitor power quality.Simultaneously reduce equipment construction and maintenance costs.This paper briefly describes the power quality evaluation indicators,as well as the causes and suppression measures of the corresponding power quality problems of each indicator,and introduces the main functions and development requirements of PQM.The multi-objective optimization problem and multi-objective optimization algorithm are summarized.The fast non-dominated sorting genetic algorithm(NSGA-Ⅱ algorithm)and its superiority are introduced.At the same time,NSGA-Ⅱ and the other two mainstream multi-objective evolutionary algorithms are optimized for multiple test functions.The algorithm is tested and analyzed,and the advantages of NSGA-Ⅱ in the convergence of the algorithm,the distribution of the solution set and the efficiency of the algorithm are verified.Focus on the PQM optimization configuration method,establish the mapping relationship between network map and tree map,establish a tree-based equilibrium deviation network topology partition model,and use the training and learning optimization algorithm(TLBO algorithm)to solve the model,the original network topology is decomposed into several subnets.Furthermore,for each sub-network,a multi-objective solution model with PQM configuration number and system average ambiguity is established,and a subnet and whole network PQM configuration model is constructed.The model is solved by NSGA-Ⅱ algorithm,and the multi-objective relationship diagram of each sub-network is obtained.Then comprehensive analysis and demonstration of the relationship between each sub-network and the whole network,comprehensively derived the original network multi-objective relationship map and the final PQM configuration.Finally,the proposed algorithm is applied to the IEEE37 node test network and a 10 k V distribution network in a certain city of China Southern Power Grid.The comparison and analysis verify the efficiency of the configuration algorithm under the premise of feasibility and economy.