Research On Health Status Perception Of Railway Vehicle Communication Network

Railway Vehicle is a large and complex vehicle engineering system,and it is the core component of the high-speed railway system.Its operation status is related to the safety of passengers and property.As the nervous system of the railway vehicle,the communication network plays an important role in the operation of the railway vehicle.The use of the communication network to realize the real-time control of the train and the transmission of various information is the key to ensuring the safe and reliable operation of the railway vehicle.During the operation of railway vehicle,once the communication network fails or its performance is degraded,it may cause system failure such as network data packet loss,and in the worst case,it may affect the safe operation of the railway vehicle.Therefore,it is of great practical value to perceive the health status of communication network systems,and it has important reference value and significance for the health status perception research of other complex engineering systems.The perception of the health status of the railway vehicle communication network is a part of Prognostics and Health Management(PHM).Under the support of NSFC project and key scientific and technological research and development project of Jilin Provincial Science and Technology Department.This article mainly discusses and explores the problems in fault diagnosis,health evaluation and health status prediction of communication networks.By analyzing the health status perception methods of common complex engineering systems,a semi-quantitative information-based belief rule base(BRB)model is proposed to solve the problems existing in each link of the railway vehicle communication network health status perception.To solve the fault diagnosis problem of the railway vehicle communication network,due to the complexity of railway vehicle communication network,there are many fault indicators when building BRB fault diagnosis model,too many antecedent attributes and reference levels will lead to the problem of "combination explosion" of rule parameters,which makes it difficult to effectively use limited prior knowledge of experts to establish the initial rules of large-scale BRB model.In order to solve this problem,a method of automatically generating initial parameters of large-scale BRB model based on cloud model is proposed.Firstly,the standard rules are determined according to the experience of experts,and then the rest rules are generated by using the uncertainty transformation between the qualitative concept and quantitative value of cloud model.Taking the node fault diagnosis of wireless sensor network in the intelligent perception network of railway vehicle as an example,the experimental results show that the proposed method can make full use of expert knowledge to build rule base,and has higher diagnostic accuracy than other models.To solve the problem of assessing the health status of railway vehicle communication networks,based on an in-depth analysis of the characteristic quantities affecting the health status of communication networks,in order to make full use of quantitative data information and qualitative expert knowledge,a BRB model-based health evaluation of railway vehicle communication networks This model builds a belief rule base,combines the activated rules with an ER iterative algorithm,and finally uses the resulting confidence to measure the health of the communication network.A data set was established based on the fault simulation of the MVB bus of the railway vehicle communication network,and the rationality of the proposed method was verified by experiments.To solve the problem of predicting the health status of the railway vehicle communication network,an Improved CBRB model is proposed to predict the health status of the railway vehicle communication network.In this method,the cloud model is used to replace the BRB prediction model conclusion reference level method,which makes the model’s ability to deal with uncertain information further strengthened;at the same time,in view of the unreasonable calculation of matching degree in the original CBRB prediction model,the prediction accuracy of the model is further improved.Finally,the results of health assessment are used as training and testing data,and the effectiveness of the proposed method is proved through experiments.