Research On Safety Risk Assessment Of Train Control Center Based On Fuzzy-FMECA Method

Railway signal system is an important safety related system in the railway systems,which has played a vital role in protecting the safety of the train under the high speed. Inorder to manage the railway signal system scientifically, the theory of risk assessment can beadopted to identify the dangers of railway signal system in the design phase, and thesehazards can be assessed and controlled later. Risk assessment can control these hazards in theacceptable range, thus avoiding causing significant casualties and property losses. The traincontrol center of the railway signal system is selected as the research object, the correlationfunctions and interface modules of the train control center have been studied in the riskassessment.Due to the risk assessment is a multi-factor comprehensive evaluation problem, whichcontains a lot of uncertainty factors, therefore, building a scientific assessment model isnecessary. The related devices of railway signal system belongs to the high reliability devices,it is difficult to determine the risk grades of the system or the equipment through the existingfault records because the risk data are incomplete or there is a high level of uncertaintyinvolved in the risk data, it must be estimated to get the risk grade on the basis of expertknowledge and experience or engineering judgment inevitably. However expert scoringmethod is difficult to ensure the objectivity and accuracy of evaluation results, so the fuzzyanalytic hierarchy process method and fuzzy comprehensive evaluation method based oncloud model are applied in dealing with the results of expert scoring respectively by combingthe fuzziness and randomness, so as to make the evaluation more scientific and reasonable,and provide more abundant information for the risk assessment.In proceeding with the risk assessment, first of all, failure modes effects and criticalityanalysis (FMECA) method is used to identify the risk of the functions and interface modulesfor the train control center, by the risk identification, failure modes, failure causes and failureeffects and detection of each function and interface module could be analyzed andsummarized, and then based on the standards of risk assessment, by combing with the expertexperience and spot investigation on the relative importance of each function and interfacemodule and risk rating scores, the temporary speed limit processing functions of the traincontrol center is taken as an example, fuzzy analytic hierarchy process method and cloudmodel method are applied in mapping the investigation data to the corresponding fuzzymatrixes of each failure mode respectively, it also gets the weight distribution of every failuremode, risk assessment is conducted to get the occurrence and severity and risk grade. Finally,the corresponding improvement measures and suggestions are put forward for the failuremodes of functions and interface modules to complete the FMECA analysis table. Risk assessment results show that FMECA method can identify and analyze the weakpoints and key projects in the design of train control center, fuzzy analytic hierarchy processcan effectively make comprehensive decision based on the group evaluation, which reducedthe subjective factors contained in the results of the assessment of experts, the application ofthe cloud model in the reconstruction of the membership functions is favorable to realize thetransition from the accurate data into the quality data by combining the fuzziness andrandomness, so as to make the assessment results of the suggested methods visual anddependable, and increase the visual and virtual intuitiveness of the reference data andinformation of the risk assessment. Eventually the FMECA analysis table can providereferential basis for designers to improve the reliability of the system design and maintenancepersonnel to draw up maintenance plans. The introduction of the cloud model should be ableto develop and improve the risk assessment method for the railway signal system.