Research On Transient Voltage Stability Evaluation Method Of Receiving Power Grid Based On Trajectory Cluster And Mblda

Transient voltage stability assessment is one of the important components of the power system safety and stability analysis method,and it plays an important role in ensuring the stable and efficient operation of the power system.The transient voltage stability evaluation method has also transitioned from the traditional simulation analysis method to the pattern recognition method.This article focuses on the transient voltage stability evaluation method based on pattern recognition.The main research contents are as follows:(1)Build an AC and DC receiving end system model including power electronic equipment,and analyze the factors that affect the transient voltage stability of the receiving end power grid.Connect the DC system at the AC system tie line,and build a wind turbine and photovoltaic power station model to replace the traditional generator set to build an AC and DC receiving end system with a high proportion of power electronic equipment.There are three main aspects that affect the transient voltage stability of the AC and DC receiving end system: the proportion of new energy system output,DC system control characteristics,and load dynamic characteristics.The basic electrical quantity information is obtained through simulation analysis,and the receiving end system is summarized after disturbance.The transient voltage state provides basic information for the subsequent feature construction method based on pattern recognition;(2)Simulate the PMU measurement system to obtain the voltage information of the key nodes to construct the original feature set,and use the feature extraction algorithm to screen out the important feature quantities to represent the transient voltage stability state.Through the BPA simulation software to simulate the PMU measurement function,the voltage trajectory of the key node after the fault is obtained,and the original feature set is constructed through the definition of the voltage trajectory cluster,which can effectively characterize the four states of transient voltage stability(rapid recovery of voltage,delayed recovery of voltage,continuous low voltage,voltage oscillation),and use the Relief F algorithm to filter out the feature subsets that are strongly related to transient voltage stability,reduce the spatial dimension of the input feature set,and improve the evaluation efficiency of the system’s transient voltage state.(3)Linear discriminant analysis,a classic linear learning method in pattern recognition,is introduced as a classifier model to transient stability assessment,and a power system transient stability assessment model based on voltage trajectory clusters and LDA is established.In order to further improve the model training speed,based on the linear discriminant analysis LDA,the eigenvalue equations of the high-dimensional voltage feature space are transformed into underdetermined homogeneous equations,and multiple between-class linear discriminant analysis model to establish a power system transient voltage stability assessment model based on voltage trajectory clusters and MBLDA.The proposed evaluation model further improves the solution speed of the model to meet the computational efficiency requirements when evaluating large-scale power systems.The simulation analysis of the modified IEEE-39 node system and the modified IEEE-145 node system verifies the accuracy and effectiveness of the proposed method.