An abnormal detection method for the state of subway electromechanical equipment based on big data analysis

Abstract

The invention discloses a metro electromechanical equipment state anomaly detection method based on big data analysis, and anomaly detection of the metro electromechanical equipment is realized from abrand new perspective of data evolution process and data association. The latent features of historical data are mined through a time series model and an adaptive neural network, and the dynamic change rule of the data to time is represented by a transfer probability sequence. For multi-dimensional monitoring data, the correlation between all the parameters is simplified through an unsupervised clustering method, so that the problem that the correlation between the parameters is difficult to determine is solved. An anomaly detection system is provided, and the anomaly detection system is suitable for state monitoring data flow of metro electromechanical equipment, and the anomaly in the data flow can be rapidly detected.

Description

technical field [0001] The invention belongs to the technical field of neural network equipment faults in a big data environment, and specifically refers to a method for detecting abnormal states of subway electromechanical equipment based on a self-organized neural network (self-organized maps, SOM) algorithm in a big data environment. Background technique [0002] Metro electromechanical equipment will be affected by abnormal events such as overload, overvoltage, internal insulation aging, and natural environment during actual operation. These abnormal operating conditions will lead to equipment defects and failures. Strong necessity. In the actual operation and maintenance of equipment, most of them are based on partial equipment information of a single system, and a simple threshold judgment method is used to detect abnormalities. This traditional threshold judgment has limitations. On the one hand, the utilization rate of equipment information and the correct rate of s...

AI Technical Summary

Problems solved by technology

Due to the slow development of latent faults of subway electromechanical equipment, when the equipment is in an abnormal state, the monitored parameters often do not exceed the limit values ​​in the guidelines or regulations, so it is difficult to detect

According to the above conclusions, it can be seen that for the online monitoring data that does not exceed the limit value of the state quantity, the abnormal state cannot be detected simply by using the AR model.

Images