Font Size: a A A

Power Tele-control System Fault Diagnosis Based On Bayesian Network

Posted on:2019-09-30Degree:MasterType:Thesis
Country:ChinaCandidate:Y ZangFull Text:PDF
GTID:2392330590492258Subject:Control engineering
Abstract/Summary:PDF Full Text Request
Power tele-control system is the basis of automation and management modernization of power system.The power tele-control transmission channel is the main channel for dispatching control information in power system and is an important power infrastructure to ensure the safe,stable and economical operation of power system.As more and more intelligent electronic devices are connected to the power tele-control system,the size of the power telecontrol system is getting larger and larger,and the problem of fault management in the power telecontrol system is becoming increasingly prominent.Therefore,fault diagnosis of tele-control systems is studied has a very significant practical significance.This paper will study the topic of fault diagnosis of power telecontrol system from two aspects-anomaly detection and fault localization,the following are the main contents:(1)In the view of the complex spatial distribution of tele-control transmission data in power system,it's difficult for the traditional Euclidean distance to meet the requirements.In this paper,a power system tele-control transmission anomaly detection algorithm based on density clustering with distance metric learning is proposed.The result of simulation experiments show that the proposed method can significantly reduce the false detection ratio and improve the detection ratio.(2)Aiming at the fact that the causality relation between fault and symptom in power tele-control system is more and more complicated,the traditional fault location method is difficult to meet the requirement of telecontrol system.A fault location based on Bayesian network model is proposed.In the view of the shortcoming of the traditional algorithms,the concept of symptom interpretation degree is introduced,the candidate fault subsets are heuristically searched based on symptom interpretation degree.The simulation results show that this method can improve the performance of fault localization.(3)In order to verify the feasibility of this method,this paper builds a simulation model of power telecontrol system based on OPNET network simulation software,and then simulates the fault scenarios and analyzes the simulation data.The simulation results show that this method can significantly improve the performance of power tele-control fault diagnosis algorithm.
Keywords/Search Tags:Power Tele-control System, Fault Diagnosis, Bayesian Network, Distance Metric Learning, Symptom Interpretation Degree
PDF Full Text Request
Related items