Reliability Analysis Of Train Control Center Based On Dynamic Fault Tree And Networks Of Stochastic Hybrid Automata

With the rapid development of China High-speed Railway,the speed of trains has been greatly improved.In order to ensure the safety of high-speed railways,China has formulated and equipped the Chinese Train Control System(CTCS).The Train Control Center(TCC)is one of the core components of the ground equipment of the CTCS.The TCC is usually equipped on high-speed rail lines from 200km/h to 250km/h.Its function is to determine the Movement Authority(MA)based on the information such as the occupancy of trains within the scope of management,the interlocking route information,and the line speed limit status,etc.The MA is then transmitted to relevant trains through the track circuit and the balise to realize operation control.Since the reliability and safety of the TCC are related to the normal and stable operation of the train control system,which affects the safety and efficient operation of the entire high-speed railway,it is necessary to carry out qualitative and quantitative reliability analysis and research on the TCC.As the key equipment to ensure the safe operation of trains,the TCC usually adopts redundant structure design.This redundant structure design feature causes the TCC equipment failure to show dynamic failure characteristics and common cause failure(CCF)characteristics.The existing research on the qualitative and quantitative reliability analysis of the TCC usually adopts the method of combining Dynamic Fault Tree(DFT)with Markov chains(MC),static Bayesian networks(BN),or Dynamic Bayesian networks(DBN).These methods cannot simultaneously consider the problems of dynamic failure and CCF or has the problem of the explosion of the parameter combination of the conditional probability table.These shortcomings bring difficulties to the qualitative and quantitative reliability analysis of the TCC.Aiming at the deficiencies of the above research methods,this thesis proposes a new method to conduct reliability research on the TCC.The main work is as follows.1.A new method based on DFT and Networks of Stochastic Hybrid Automata(NSHA)is proposed to conduct qualitative and quantitative reliability analysis.Specifically,a DFT model is constructed by analyzing the system structure and functions.Based on the dynamic fault tree failure mechanism and the attributes of the Stochastic Hybrid Automata(SHA),the rules for converting the DFT model into the NSHA model are proposed.Based on the simulation tool UPPAAL-SMC,a model verification method is proposed.Finally,a reliability analysis framework based on DFT and NSHA is proposed.2.Using the proposed method,a qualitative and quantitative reliability analysis on the TCC was carried out.Firstly,through in-depth research and analysis of the structure and core functions of the TCC,and inviting experts to identify and evaluate the potential risks of the TCC,the DFT model of the TCC is established on considering the characteristics of dynamic failure and CCF,and independent failure parameters and CCF parameters of the TCC equipment are obtained.Secondly,based on the conversion rules,the DFT model is converted into a NSHA model.Thirdly,the model is verified in simulation tool UPPAAL-SMC by combining with the model verification statement.The failure probability of the top event in the DFT model is calculated,and the importance of equipment is analyzed.Finally,the influence of CCF on the reliability analysis result is discussed.The results show that PIO,DY and VC are the objects that need to pay attention to;CCF has a great impact on system reliability and is a factor that cannot be ignored.3.Based on the model conversion rules proposed in this paper,an automatic DFT model conversion tool is designed in the C++ development environment,which can automatically convert the DFT model into a NSHA model and save it as an XML file,which is convenient for analysists to use UPPAAL-SMC to open and perform subsequent simulation verification.