Data-driven Power System Voltage Sag Location Method Considering Monitoring Point Optimization

Among the many power quality problems,voltage sag is one of the most prominent power quality problems in the power system,which brings a threat to the use of many electrical equipment.Among the many branches of voltage sag problem,realizing the rapid and accurate location of voltage sag source can better define the responsibilities of the power supply side and the user side,and can provide a reliable basis for future power grid development planning,which has important theoretical research significance and practical application value.Firstly,the concept of voltage sag is introduced in this thesis.Then the causes of voltage sag are described.The possible harm of voltage sag is briefly analyzed,and the relevant measures to control voltage sag are described in detail from the power supply side and the user side.After that,the principle analysis of several classic positioning methods of voltage sag source is carried out respectively,and the simulation experiment is carried out in MATLAB,and the advantages and limitations of each method are briefly analyzed according to the simulation results.The simulation results show that,Several classical methods of voltage sag location are not suitable for all types of faults.Secondly,a voltage sag source location method based on GA-PNN(probabilistic neural network)is proposed.First of all,the electrical characteristic quantity of several classical voltage sag methods is extracted as the input quantity of neural network,and the probabilistic neural network model is established.Genetic algorithm is used to optimize the identification characteristics of probabilistic neural network,so as to improve the recognition characteristics of neural network,so as to accurately judge the location of voltage sag source.In order to verify the accuracy of the proposed method,a dual power system simulation model is built in MATLAB to verify the proposed method.The simulation results show that the voltage sag positioning method based on GA-PNN has higher accuracy than the traditional voltage sag positioning method.Finally,this thesis introduces the theory and method of optimizing the layout of monitoring points and combines the upstream and downstream positioning method proposed in this thesis with the optimal layout of monitoring points,so as to narrow the fault area that needs to be located,reduce the calculation required for positioning,and improve the speed of voltage sag positioning.After that,the approximate estimate method of fault resistance is introduced,and according to the theory of fault location,the least square method is used to analyze and calculate the nodes and lines in the possible fault area respectively,and finally the more accurate fault location is obtained.In this thesis,IEEE14 node model is set up in MATLAB to prove the feasibility and accuracy of this method.This Thesis has 47 figures,13 tables and 75 references.