Fault Detection Method of Local Outlier Factor for High-Speed Train of Running Gear Based on Inter-Variable Variance

A fault detection method of local outlier factor for high-speed train of running gear based on inter-variable variance is proposed which aims at solving the problems of multiple data sources, high dimension and complex actual working conditions in the process of high-speed train of fault detection in this paper. Firstly, considering the high complexity of space-time computation in big data, IVV(Inter-Variable Variance) algorithm is used to fuse data tracks with similar running states to reduce the computational complexity. Secondly, taking into the fact that the data are more affected by external factors, the linear threshold is replaced by the non-linear threshold as the standard for fault detection by horizontally comparing the other non-fault components of the same high-speed train. Lastly, in the case of determining the position of the fault components of the train initially, the LOF(Local Outlier Factor) algorithm is used to calculate the number of outliers of a single measure point, so as to determine the exact position of the fault and improve the accuracy of the algorithm. The simulation experiment is carried out in combination with the actual running data of high-speed trains, which verify the validity of the algorithm in the fault detection of high-speed train of running gear.