High-speed Train Braking System Fault Detection Dataset
The braking system is a crucial mission subsystem for the safe operation of High-Speed Train (HST). There are a total 44 monitored variables related to the braking system faults of HST. They include train-level conditions, e.g. speed, operation mode, external power supply, line voltage, line current, and braking system-level conditions, e.g. internal temperature, battery voltage, detected slip or slide, ED brake state, TCL brake state, achieved brake effort, etc. All the nominal variables are transformed into the numeric feature. After data transformation, the dataset contains 46 numeric features, 31 features are binary ones, and 15 features are continuous. Due to commercial agreement constraints, we linearly transformed all data and desensitized feature names. There are 22,368 samples in the processed HST dataset, including 389 faulty samples, and the IR of the dataset is 56.50:1.
Publications making use of the HST datasets are requested to cite the following papers.
M. Qian and Y. -F. Li, "A Novel Adaptive Undersampling Framework for Class-imbalance Fault Detection," in IEEE Transactions on Reliability. (In press)
HST dataset can be downloaded by the following link: