Research On Risk Assessment Method Of Radio Block Center Based On AR-SPA

Radio Block Center(RBC)is the core ground equipment of Chinese Train Control System at level 3(CTCS-3).Its efficient and safe operation is an essential guarantee for the safe operation of high-speed trains.At present,there are relatively few studies on RBC risk assessment,and the index weights in the traditional assessment methods are influenced by subjective factors,which makes the assessment results deviate from the actual situation.To improve the safety risk management system of RBC and improve the accuracy of RBC risk assessment results,RBC is taken as the research subject in this thesis.The risk assessment on the main functions and related interface units of RBC are conducted to finds out the weak sectors of the system and provide reference for the subsequent system design and maintenance.In this thesis,Association Rule(AR)is used to calculate the weights of RBC risk assessment indicators and Set Pair Analyze(SPA)is used to analyze the risk level of RBC,and a risk assessment method based on AR-SPA is proposed.The main research contents are as follows:Firstly,according to the relational database theory,relevant standards and the actual field situation,the indicators(fault types and fault features)that are representative and can effectively reflect the operation of RBC equipment are selected.The AR support degree formula is used to calculate the support degree of fault features,and the coupling of fault type and fault features is analyzed by the support degree value,and the fault features are reduced according to the support degree threshold to obtain the fault features with strong coupling relationship with the fault type,establishing the risk assessment model.The confidence degree of the reduced fault features is calculated by the AR confidence formula,and the constant weight of each fault feature is determined by the confidence value.Secondly,according to the risk level classification of RBC,the risk level is described by using the five element associative number of SPA.The fault features are evaluated,and the relative scoring values of each fault feature are obtained from the fault feature evaluation values and the simplification formula,and the relative scoring values of the fault features are obtained after using the relative affiliation function in fuzzy theory to construct the fault feature identical discrepancy contrary evaluation matrix,and the fault type identical discrepancy contrary evaluation matrix is derived from fault features identical discrepancy contrary evaluation matrix.The subject identical discrepancy contrary evaluation matrix is obtained after combining the fault feature weights with the fault type identical discrepancy contrary evaluation matrix.The fault type weights are calculated according to the variable weight theory and diversity coefficients are processed by the mean division method to obtain the diversity coefficient matrixes.The identical discrepancy contrary evaluation matrixes,fault type weights and diversity coefficient matrixes of the subject are substituted into the SPA associative number formula to obtain the risk level associative number of the subject that determines the risk level of the subject.Finally,the cloud model is used to obtain the risk level cloud map of the subject,and the cloud map is compared and analyzed with the assessment results of the AR-SPA assessment method to verify the method’s accuracy.Based on the assessment results,the sectors with higher risk levels in the system are identified,and measures for risk controlling are given for those weak sectors.