Research On Fault Diagnosis System Of Subway On-board Signal Equipment

With the rapid development of urban rail transit construction in Beijing, Beijing subway line now covers the most parts of the city. Bringing a lot of convenience for people to travel, Beijing subway has become the most favorable means of transport for more people to travel. In order to ensure the efficient transportation at daily peak, subway lines tend to be pulled up to ultimate bearing capacity during operation. However, under the condition of high-intensity and overtime running, failure occurs and affects the transport efficiency and the safety of train inevitably.For the subway on-board signal equipment is a complex system which has diversity and uncertainty and the fault samples are not complete, there is a great need for a fault diagnosis system which can quickly deal with the train faults, ensure the security and efficient of the train’s operation. Based on Bayesian network, this thesis develops the diagnostic system of subway on-board signal equipment, improves the equipment maintenance strategy, saves and uses the fault data in a better way. The main work of this thesis is as follows:Firstly, according to the situation of the subway on-board signal equipment system of Beijing Subway Line 13 nowadays, the fault diagnosis method of subway on-board signal equipment is carried out based on Bayesian network in this thesis. Through the description of subway on-board signal equipment’s faults, resolve methods and data mining of the train on-board equipment fault recording, the relationship of faults is found and fault sample set is generated;on the basis of the information from data mining analysis method based on Bayesian network for subway on-board signal equipment fault diagnosis system is developed.Secondly, using the generated fault samples to study the network structure and combining with the expert knowledge, the Bayesian network model is constructed. And the parameter learning of Bayesian network is completed by using fault samples. A method of subway on-board signal equipment’s operating-life prediction based on Weibull distribution function and Bayesian network inference is proposed to try to improve the repair mode of subway on-board signal equipment from "fault repair" to "state repair".Finally, using MATLAB 2010b with Microsoft Visual Studio 2013 mixed programming, combining with two parties’advantages in algorithm processing, form design and function control design, the fault diagnosis system software is complete designed. And a communication mechanism that can make the use of the failure data among employees more effective is put forward. A fault diagnosis system of subway on-board signal equipment with the capability of on-line information transmission is achieved.The fault diagnosis methods and tools proposed in this thesis can effectively improve the efficiency of fault diagnosis and the efficiency of data recording. It has certain guiding significance in the predication of equipment, which reduces the labor intensity of maintenance personnel, and provides strong technical support for the safe and efficient operation of the subway.