Research On Fault Diagnosis Of Relay Protection System

With the continuous improvement of people's living standards and the increasing demand for electricity,the stability requirements for grid operation are getting higher and higher.However,the complexity of the power grid structure and the diversification of electrical equipment have caused the randomness of the operation of the power system to increase continuously.In addition,due to natural disasters,man-made damage,and operational errors of power grid workers,power system failures sometimes occur.After the fault occurs,it is necessary to confirm the fault cause and the position of the fault point as soon as possible,and restore the power supply as soon as possible to avoid fUrther expansion of the loss.For the grid fault,rapid information analysis and collation,fault type diagnosis and fault component location are the focus of this research.When the power system fails,a large amount of information is returned to the relay protection information system.The determination of the operating status and fault type of the electrical components is based on these data.The rapid diagnosis of grid equipment faults requires rapid processing of return information,screening out data feature quantities related to grid faults,and enabling rapid fault diagnosis and location based on feature quantities.After referring to the relevant research literature at home and abroad,this paper proposes an electrical equipment fault diagnosis model based on support vector machine algorithm based on the data of current value,voltage value,impedance value and protection switch position action information returned by relay protection information system.Aiming at the classification effect parameters of the support vector machine algorithm,a stochastic gradient descent method is proposed to optimize the parameters.And to establish an electrical equipment fault diagnosis module,design system related functions and implement based on the mainstream framework.The system can perform data processing and mining on the return information of the relay protection device,eliminate redundant information,extract fault diagnosis related data items from the multivariate data,and provide data unified interface support.The data processing is carried out by using the extracted actual operating data feature quantity,and the function verification of the electrical equipment fault diagnosis module is carried out,which indicates that the system can be well applied to power system fault diagnosis.