Method Of System Reliability And Safety Evaluation For High-speed Train Based On Topology

The traditional identification methods of components characteristics and system reliability and safety assessment methods have lots of defects,such as ignoring system topology and influence of fault propagation mechanism among components,separating relationship between system reliability and safety.Considering above shortcomings,based on high-speed train system,system topology is analyzed and a holistic topological network model is constructed.Besides,from the perspective of combination of function and topology,an identification method for critical components is proposed based on aggregation operator,and then an approach for assessing weak components is also discussed based on interval-valued intuitionistic hesitant fuzzy Choquet integral.Based on the research idea of from components to the system,a failure propagation model is introduced in view of improved polychromatic sets.Finally,a system reliability analysis method based on system effectiveness measure is presented,and an evaluation method of system safety based on the distance measure is put forward.Main research contents are as follows:(1)A modeling method for holistic topological network based on structureThrough analysis of the physical structure of high-speed train system,the extraction rules of components and the classification of relationships among components are given.Based on these,several hierarchical topological network models are constructed to describe different physical structures,including mechanical topological network,electrical mechanical topological network and information mechanical topological network.Finally,the holistic topological network,which reflects the global structure,is proposed by integrating all those hierarchical topological network models.(2)An identification method for critical components based on aggregation operatorIn the light of the holistic topological network and failure data,several generalized importance measures are presented in combination with system function and topology based on fuzzy integral.Due to the uncertainty and subjective influence,an aggregation operator is constructed.In order to identify critical components in the system,the comprehensive importance is proposed through integrating multi-generalized importance measures based on aggregation operator.(3)An identification method for weak components based on intuitionistic fuzzy theoryBased on analysis of failure data and identification results of critical components,weak component and related basic concepts are introduced.For high-speed train,the quantity of failure data is very large.However,the amount of failure data for a component on one car of one train is small.According to the characteristics of system structure and failure data,interval-valued intuitionistic hesitant fuzzy of nodes measures are proposed.Besides,the weakness measure is constructed to assess weak components based on interval-valued intuitionistic hesitant fuzzy Choquet integral.(4)A system fault propagation model based on improved polychromatic setsFor nodes in the holistic topological network,the failure pervasion intensity of nodes is improved in combination with comprehensive importance of nodes and failure data.Considering characteristics of high-speed train system,an improved polychromatic sets theory is introduced.Finally,a system fault propagation model based on the improved polychromatic sets is constructed according to the failure pervasion intensity.(5)A system reliability analysis method based on system effectiveness measurementBased on the generalized importance measure-the shortest function-path length,the global efficiency reliability is improved and the system effectiveness measurement is given.The system reliability is constructed based on system effectiveness measurement and fuzzy integral,since there may be multiple fault propagation paths at the same time in the system.Finally,the system reliability is compared with the system reliability threshold pc,which is calculated by the percolation threshold based on percolation theory.(6)An evaluation method of system safety based on hierarchical fuzzy integralThrough integrating the operation fault data,the severity of failure consequence is quantified and interval-fuzzified.The calculation of failure consequence,caused by failure propagation path,is introduced based on Choquet integral,since failure consequence based on failure data is not caused by the failure of an independent component.Then,the system safety is constructed based on failure consequence and Sugeno integral.At last,system safety is compared with the system safety threshold.Finally,taking the bogie system of high-speed train as an example,we apply the proposed methods and models to case study,which includes the holistic topologicalnetwork model,the identification methods for components features and the evaluation methods for system reliability and safety.The results show that the proposed methods are able to be applied to high-speed train system and other complex electromechanical systems.Meanwhile,the correctness and effectiveness of the proposed methods are also verified.