Research And Application Of Key Technologies For Operations And Maintenance Knowledge Management In EMU

At present, knowledge management has been developed as the highest form of management. In addition, data is an indispensable component in knowledge management, which is also the foundation of the EMU operations. Huge amounts of raw data will produce during the EMU operation. We may consume a lot of human and material resources and get low quality mining results if we directly mine and analysis the raw data.Based on the above problems, this paper designs and implements the system of operations and maintenance knowledge management in EMU on account of the needs of operations and maintenance. This paper designs the overall architecture, mainly, studies the technology of data preprocessing and the technique of knowledge incremental updating. Firstly, this paper puts forward the technology of data preprocessing into the EMU operations, in view of the raw data may contain dirty data. A case study of brake pad in this paper, the system uses Rosetta to finish data cleaning of brake pad. Then, we can get the key factor affecting the life of brake pad. In addition, this paper puts forward the file incremental updating strategy, and make further improvement, in view of the same time may have more than one MID (Mobile Internet Device) request to the server about fault knowledge retrieval. The above strategies solve the big pressure of the server and network, and the long time waiting of MID.Experiments show that the brake wear mainly depends on two factors, that is cut-off velocity and BC. In addition, data preprocessing reduces the workload of data mining largely. Compared to the traditional incremental updating strategy, the improved incremental updating strategy greatly reduces the transmission time of network, at the same time, reduces the pressure of the server and the waiting time of MID, improves the stability of the whole system, provides better protection for the cycle safety of EMU.