Research On GNSS/INS Based Train Positioning Performance Evaluation Method | | Posted on:2022-03-19 | Degree:Doctor | Type:Dissertation | | Country:China | Candidate:B Q Wu | Full Text:PDF | | GTID:1482306560492764 | Subject:Traffic Information Engineering & Control | | Abstract/Summary: | PDF Full Text Request | | Train control system is a key technology to ensure the safety and reliability of railway transportation.To obtain real-time and accurate train positioning information is the basis for train control system to guarantee the safety of train operation.With the completion of Beidou navigation satellite system,multi-sensors fusion based train autonomous positioning become an important trend for train control system in the future.This paper focus on the GNSS/INS based train autonomous positioning,in order to ensure that train autonomous positioning can provide reliable location services for the train control system,effective performance evaluation methods should be proposed.This paper is aimed to solve the positioning performance evaluation problem for the train control system.By analyzing the processing of train autonomous positioning.the scenarios based weighted measurement noise calculation method and the fault detection and state recognition method based on integrity evaluation are studied.On this basis,the SPN-based train combination location fault propagation model is constructed,and the transition fire rate calculation based on certain information and the interval grey number expression based on uncertain information are proposed,then the availability evaluation method of the train autonomous positioning under the grey SPN model is generated.Furthermore,considering the application of train autonomous positioning in moving block,the safety risk assessment method based on step-size adaptive LSTM neural network is proposed.The innovations of this paper are as follows:(1)Based on the analysis of the positioning performance requirements in existing train control system and railway satellite navigation system research,the global satellite navigation system based train positioning performance requirement architecture is summarized,.(2)An environment scenarios surrounding train track line clustering and dividing method based on PSO and K means algorithm is proposed,and by combining with the distribution characteristics of satellites along the railway,a method for calculating weighted measurement noise in tight-coupled EKF based on environment scenarios dividing results is proposed.Based on the analysis of the tightly-coupled EKF algorithm,a multi-layer fault detection and status identification method is proposed.(3)A grey SPN-based method for evaluating the integrated train positioning availability is proposed.The SPN based train autonomous positioning fault propagation model is constructed,which can describe the processing from information collection to positioning result output in DTM assisted GNSS/INS based train autonomous positioning.The states change process is described,and the interval grey number expression method of the transition fire rate influenced by uncertain information is proposed based on the uncertainty knowledge expression method in grey system theory,.Then the availability is evaluated based on the grey SPN model,which can solve the problem of train autonomous positioning availability under complex railway environment.(4)A risk assessment method for train autonomous positioning based on step-size adaptive LSTM neural network is proposed.Based on the analysis of the implementation of moving block,the situation in which train autonomous positioning may affect the safe tracking operation in moving block is described.Then based on LSTM neural network,a risk assessment model is proposed to describe the relationship between train autonomous positioning information with timing characteristics and the risk.Considering the different time correlation levels of the network inputs in different environment scenarios,a step size adaptive adjustment method is proposed.In order to support the verification of all the methods proposed in this paper,data collected in Beijing-Shenyang railway line are used to ensure the effectiveness of the environment scenarios clustering method,the environment scenarios based measurement noise weighted estimation method,the integrity and availability evaluation method,and safety risk assessment method for moving block.The results of this paper can provide theoretical reference and technical support for promoting the application of satellite navigation system in train control system in the future. | | Keywords/Search Tags: | Train Positioning, GNSS, Performance Evaluation, Fault Detection, SPN, Neural Network, Risk Assessment | PDF Full Text Request | Related items |
| |
|