Research On On-line Monitoring And Cause Identification Of Voltage Sag In Distribution Network

With the development of science and the progress of society,high-end industries such as vehicle manufacturing and chip manufacturing are gradually increasing,and the voltage sensitive load in the power system is also increasing.Voltage sag disturbance is one of the most serious power quality problems affecting and harming.It is very easy to make computers,refrigeration motors,programmable logic controllers and other equipment operate abnormally,and even cause factory shutdown,equipment shutdown and other problems,resulting in huge economic losses to enterprises.The monitoring and control of voltage sag is related to the common interests of power grid and power consumption enterprises.Based on this,this paper focuses on the detection of voltage sag events and the analysis of disturbance sources,and studies the voltage sag detection method and the voltage sag cause identification method.The main research contents are as follows:(1)A detection method of voltage sag event is proposed.Firstly,based on the traditional morphological filtering method,aiming at the problem of large amount of calculation of morphological filter in practical application,the corrosion operation and expansion operation are modified to improve the real-time performance of morphological filtering.Secondly,based on the data updating process of microprocessor buffer,the morphological filtering process is simplified,and an improved sliding window morphological filtering method is proposed αβ-Dq transform is combined with voltage sag detection algorithm to realize fast and accurate detection of voltage sag events.Finally,the accuracy and real-time performance of the method are verified by simulation analysis and measured data.(2)A method for identifying the cause of voltage sag based on bilstm is proposed.Firstly,the typical simulation models of voltage sag such as different short-circuit faults,inductive motor startup and transformer operation are established,and the sample data of different types of voltage sag disturbance are obtained.According to the characteristics of different types of voltage sag,the time-domain characteristics of voltage sag waveform and Stransform energy entropy characteristics are extracted,and the comprehensive feature vector of voltage sag cause identification is constructed.Secondly,the characteristic data of voltage sag is divided into training set and test set,and the bilstm network model is built and trained to identify the cause of voltage sag.Finally,the accuracy rate,accuracy rate,recall rate and F1 value are selected as the indexes of the evaluation model to measure the quality of the identification results of the causes of voltage sag;The effectiveness and accuracy of the proposed method are verified by comparing with deep confidence network,support vector machine and decision tree method.(3)The software and hardware functional architecture of intelligent voltage sag monitoring and analysis device is designed.Firstly,the architecture and functional modules of voltage sag intelligent monitoring and analysis device are described,the hardware structure and parameters of voltage sag intelligent monitoring and analysis device are described,and the main functional interfaces of voltage sag intelligent monitoring and analysis software are introduced;Then,a reasonable and effective test scheme is designed for each function of the device,and a large number of simulation tests are carried out based on the laboratory environment.Finally,the device is tested and applied in a substation.