Research on Association Rules of Train Communication Network Fault Alarm Based on Data Mining

Abstract

With the continuous development of intellectualization and integration of high-speed train, the system units such as on-board equipment and lines become more and more complex, and the operation status and fault alarm data transmitted and stored in the train communication network are gradually increasing. How to obtain the correlation between fault alarms from fault alarm information data and form the relevant diagnosis knowledge is of great significance to training fault location and safe and efficient operation. FP-Growth algorithm is one of the classical algorithms for mining association rules in data mining methods, which is used to mine frequent item sets of data. Based on the analysis of real fault alarm data in a high-speed railway communication network, the association rules of fault alarm are studied by combining data mining algorithms. Experimental results show that the FP-Growth algorithm can effectively mine the association between train state information and fault information, form an effective alarm rule base, and improve the maintenance efficiency for operators.