Fully Automatic Operation System Fault Propagation And Cognition Model

The Fully Automatic Operation(FAO)system has realized the automation of the whole process of train operation,greatly reducing the work pressure of the driver,but also putting forward higher requirements for other operators,especially train dispatchers.When facing sudden failures,train dispatchers have greater cognitive pressure and heavier decision-making responsibilities,and their possible cognitive errors have a more direct impact on train operation.The train dispatchers take direct responsibility for the fault handling in the FAO system.Reducing the cognitive pressure of the train dispatchers and avoiding the cognitive errors are the important for improving the operating efficiency of the FAO system.The main research contents of this paper are as follows:First,according to the characteristics of the FAO system and the traditional CBTC(Communication Based Train Control System)system,starting with the responsibilities and personnel organization structure of the train dispatchers,combined with the STAMP theory,the cognitive information flow of the train dispatchers is analyzed.Comprehensively referring to basic theories including Information Processing Model,Situational Awareness theory and Contextual Cognitive Model,the dispatcher cognitive model is established from the perspective of cognitive process and cognitive level.Secondly,a cognitive ontology model for train dispatchers is constructed to describe the specific cognition of train dispatchers on on-site operation.Combining the theory of fault propagation,the train operation fault propagation model is established,and the sparse representation theory is used to formulate the auxiliary decision-making scheme.The deduction of the train dispatcher’s cognitive ontology model is proposed to express the process of dispatcher’s reasoning and prediction based on the current situation.Taking two alarm events in the Yanfang Line as an example,the actual application was carried out,and the actual data of the VOBC(Vehicle On-Board Controller)system fault handling of the Yanfang Line from 2017 to 2020 was used to verify the auxiliary decision-making scheme.Finally,on the basis of summarizing the cognitive characteristics of train dispatchers,the human controller cognitive model and control structure model in STPA(SystemsTheoretic Process Analysis)theory are extended from the perspective of cognitive process.A method to analyze the casual scenarios of cognitive errors is proposed.Combining the concept of leading indicators in risk management,the identification method of cognitive failure leading indicators is proposed,and the Semantic Web Rule Language(SWRL)is used to realize the automatic identification of the casual scenarios and leading indicators of the dispatchers’ cognitive errors.The improvement strategy in real-time operation is proposed by using the leading indicator.Finally,two cases was used to apply the automatic analysis method,and some improvement strategies for real-time operation are proposed.This thesis contains 57 figures,15 tables and 70 references.