Failure Risk Analysis Of Train Integrity Detecting Based On GNSS

Train Integrity Monitoring System(TIMS)can detect the integrity states and send the detection reports to the train drivers and the ground monitoring staffs dynamically,to avoid train derailment,train crash or other accidents.In our country,train integrity detection technology based on ground equipment is widely used.However,it needs huge expenditures for the deployment and maintenance of ground equipment,which is a non-economical solution.The detection technology based on air pressure checking has a disadvantage of low checking frequency,which is unable to guarantee real time.What's more,some factors such as system hardware faults and environment interference would lead to the failure of integrity detection.As an important part of train operation control system,TIMS is required to realize autonomic,real-time and accurate train integrity checking for the Next Generation of Train Control system.Therefore,a TIMS scheme based on GNSS is designed in this thesis.The risk of component failure and external interference on train integrity detection is also studied by using the fault tree analysis and the probabilistic model checking method.The main research contents of the thesis are listed as following:(1)This thesis constructs the GNSS based TIMS,which combines the advantage of air pressure detection based method and the train length detection based method.(2)This thesis studies the safety performance of the GNSS based TIMS.First,the fault tree analysis method is adopted to qualitatively analyze the faults in TIMS.Then,the minimum cut sets that result in detection failure are listed to find out those factors which influence the detection process most.Considering the randomness and the uncertainty of TIMS,in the quantitative analysis,Markov Decision Processes(MDPs)are used to model the three TIMSs(the air pressure detection based,the train length detection based and the GNSS based).(3)This thesis verifies and quantitatively analyzes the designed system model by using the probabilistic model checking tool PRISM,and the correctness of the designed system model is confirmed.Then,three main modules(power supply module,communication module and detection module)in the GNSS based system are verified and analyzed.Meanwhile,the failure risk probability of the system is calculated by PRISM.Finally,the failure risk levels of the three TIMSs are compared.The results indicate that the proposed train integrity detection failure analysis method can evaluate the risks of system module hardware and external environment.Based on that,the safety performance of detection system is evaluated effectively.Meanwhile,the results indicate that the designed GNSS based TIMS can reach the SIL4 level.The results obtained in this thesis will provide important support for the later optimization of the system security.