Railway vehicle running gear anomaly detection method based on symbol regression and generative adversarial network

The invention discloses a railway vehicle running gear anomaly detection method based on symbolic regression and generative adversarial network, comprising the following steps: acquiring sensing data of a railway vehicle running gear, and storing the sensing data in a database; establishing a structure learning model and a generative adversarial network model, and superposing results of the structure learning model and the generative adversarial network model as health baseline data; acquiring sensing data of the running gear of the railway vehicle in the target time period as real-time monitoring data, and calculating a real-time deviation according to the real-time monitoring data and the corresponding health baseline data; and calculating an average error of all the real-time deviations in the target time period, and if the average error is greater than a preset alarm threshold, judging that an abnormality occurs. According to the invention, accurate mechanism analysis can be carried out on the structural