Application Of Fault Prognostics And Health Management Technology In EMU Operation And Maintenance

In recent years,the high-speed railway has gradually become a sign of industrial achievement of China.Travelling by high-speed rail has changed the way of domestic travelling and even the lifestyle.Comfort,convenience and speed have also become the hallmarks of the high-speed train.As an important carrier of passenger transport,the trains are a major part of the fleet.The number of rolling stock in operation has increased years.The number of trains has now exceeded 3200.The maintenance of rolling stock is still mainly planned preventive maintenance.The conflict is more pronounced between this maintenance pattern and the volume of work.It is urgent for the railway industry to realize the revolution of maintenance scheme so as to achieve intelligent operation and management of Electric multi-units(EMUs).Prognostics and Health Management(PHM)of EMU is an important method to change the existing maintenance mode and implement the reform of maintenance schedule.Through PHM technology,we can use the big data analysis mode to obtain EMU life cycle data and model calculation.Putting forward the maintenance strategy for a part combined with the actual situation,to improve EMU maintenance quality and efficiency and reduce maintenance cost.Through the advanced sensor real-time or asynchronous read running state data,and combined with the system running environment or historical factors,through big data analysis technology and neural network to predict the future state of the system operation,effectively evaluate the health of the system operation,realize the real-time monitoring system,and predict the fault in advance.Afterwards health management strategy is deduced to optimize or maintain the system,and solve and prevent the failure in advance.For the application research of PHM system in EMU Operation and maintenance,the main research contents of this paper are as follows:The first is to explain the background and significance of the research,and point out the importance of fault Prognostics and health management technology in EMU operation and maintenance.The second is to explain the related theory of PHM.Firstly,it introduces the basic definition of PHM,followed by explanation of the theory and implementation method of PHM.On this basis,it explains the composition of PHM system and the basic working principle of PHM system.The third is the research on the EMU PHM fault model.Firstly,the model of traction motor fault Prognostics is explained.Then,the two algorithms are combined,and the improved algorithm principle and operation mode are pointed out.Then,the CRH6 A EMU traction motor is used to build the fault Prognostics and fault diagnosis model,and the model is trained and tested with field data to verify the feasibility of the model Sex.The fourth is a brief introduction of high-voltage power supply and traction drive system,and select CRH6 A EMU traction motor historical data into the model for prediction,through the change of ambient temperature to predict the motor temperature alarm cycle,and use the field operation to verify the results.The fifth is to describe the key problems of PHM technology in the process of EMU maintenance.Through the EMU data source,transmission,as well as the characteristics of each EMU subsystem corresponding algorithm,to solve these key problems.Finally is to establishes the EMU PHM model management platform,clarifies the construction goal of PHM model management platform,and then explains and studies the system design method from three aspects of functional architecture,logical architecture and technical architecture,and studies the function realization method,which proves the effectiveness of this method.In the process of studying CRH6 A EMU traction motor fault prediction model,this paper selects the real historical parameters of the motor,and uses the prediction function of the model to predict the motor temperature,and applies the prediction results to the actual production,which verifies the feasibility of PHM technology in condition-based maintenance to a certain extent.