Font Size: a A A

The Study Of Short-circuit Fault Detection And Fault Location Method Of Dc Traction Power Supply System

Posted on:2014-01-25Degree:MasterType:Thesis
Country:ChinaCandidate:Y G WangFull Text:PDF
GTID:2232330398476006Subject:Power system and its automation
Abstract/Summary:PDF Full Text Request
Urban rail transit with its comfortable,fast and convenient features,people gradually become the main means of transporthas been rapid development in recent years.DC traction power supply system is the core part of the urban rail transit project, its safety and reliability is a precondition for normal operation of trains.Due to the characteristics of the catenary no alternative,and once it failed, need maintenaners in quickly repair the catenary fault.So need to quickly identify and accurately locate faults, timely maintenance and troubleshooting, turn power back on, reducing power losses, which have important significance to the normal operation of urban rail transit.This article analyzes the24-pulse rectifier units, rail model considering the effects of skin effect, ac drive system, etc.Using PSCAD/EMTDC to build catenary short circuit fault model and normal starting models,through simulation analysis, obtained feeders current characteristics of short-circuit fault and locomotives start.Using the theory of wavelet transform and SOM neural network, combines feeders current characteristics when the locomotive start-up and short circuit fault, researches a method of real time identify two current. Sliding time window was carried out on the feeder current signal wavelet transform for real-time detection feeder electric surge time point, the interception of surge time point after10ms feeder current data, the use of wavelet packet transform to the signal for3layers of decomposition and get the high frequency and low frequency signals, as a feature vector, and clustering analysis into the SOM neural network. Experimental analysis show that the method can real-time detect feeder surge current, valid identification of fault, provide reference for feeder protection and fault location.For traction network parameter deviations on fault location accuracy of double Terminal electrical quantities, research based on catenary parameter adaptive method of two Terminal time domain location, and equation solving fault location using the search method.Experimental data show that in the case of catenary parameters are not accurate, this method is compared with existing double end algorithm has high location accuracy, and the results are not affected by fault location, fault resistance and circuit parametersthe.According to the principle of double-terminal fault location, combined with the actual conditions, adopt modular design concept, developed a programme to research fault location devive for DC traction power supply system. Emphatically introduced upper-computer software design and realization of fault location system, and use of the simulation data test the software functional, verified it operate stability to meet the design requirements.
Keywords/Search Tags:DC Traction power supply system, Fault Identification, Fault location, WaveletTransform, SOM neural network
PDF Full Text Request
Related items