The Study And Implement Of The High-speed EMU’s Operation And Maintenance Intelligent Decision Knowledgebase

With the China’s high-speed EMU putting into use in the large scale, the EMU has become an important tool for railway passenger and freight transportation. With the growth of the running time and the running mileage, the operation of EMU has accumulated a large amount of operation data and rich experience in operation and maintenance. But at present the problem we faced is that these data and experience can’t be used directly and effectively by the operation and maintenance personnel. At the same time, failure is unavoidable in the long run of EMU, so how to ensure the efficient and safe operation of the EMU has become the most important task in the operation and maintenance of the EMU.Facing with these problems, this paper studies and implements the operation and maintenance of high-speed EMU intelligent decision-making knowledgebase, this base used to guide the EMU’s operation and maintenance decisions. The establishment of knowledgebase enables knowledge input, sharing knowledge exchange and so on normally.In view of the current situation of massive data and the characteristics of the rapid response of operation and maintenance requirements, this paper makes the following improvements to the key technologies used in the knowledgebase. (1) Improve the existing Fp-growth algorithm in the process of knowledge discovery,and improve the shortcomings of the existing algorithm.(2) The improved Fp-growth algorithm is Map-Reduce parallelized, and the frequent pattern tree is constructed at each node, and then the local frequent item sets are combined to get the global frequent item sets. (3) Improve the knowledge retrieval technology improved. The forward maximum matching word segmentation algorithm based on dictionary is adopted and the hash mechanism is used to organize thesaurus in this paper, and eventually the Chinese word segmentation module is designed and implemented.The experimental results show that the improved Fp-growth algorithm has been improved in two aspects:time and space performance. The Map-Reduce parallel algorithm improves the processing speed of the data, and has a good speedup and good scalability. At the same time, the use of Chinese word segmentation technology can help the operation and maintenance personnel to acquire knowledge from the knowledge base efficiently.This paper establishes rule base and fact base of the knowledgebase. This system includes inference model knowledge search model、knowledge update model and fault statistics and mining model. Apply the knowledge base technology in the operation and maintenance of EMU intelligent decision, use the visualization technology to help the operation and maintenance staffs get the optimal operation and maintenance plan quickly and effectively, help the EMU’s operation and maintenance more scientific and efficient, then reduce the occurrence probability of EMU failure and reduce EMU operation and maintenance costs, improve the life of the EMU, and provide protection for the EMU’s operation safety.