A Monitoring Data Mining based Approach to Measuring and Correcting Timetable Parameters

Calculating train timetable parameters correctly is a critical work in assessing the transport capacity of existing lines accurately. It provides important data to plan more feasible and stable train timetables. Most prior methods of determining train diagram parameters are a combination of computation of train traction and real tests, thus the parameters won’t change once determined. Obviously, there are several major weaknesses in these conventional methods: single fixed data, low accuracy, and in adaptation to dynamic changes in line conditions. Therefore, this paper proposes a new method of checking and correcting train timetable parameters based on locomotive running records. It makes the best of measured data from the Railway Information System. By statistically processing the measured data and synthetically comparing train diagram parameters in each stage, it provides important theories and data supporting for parameter corrections. The key problem of original data acquisition, measurement data processing, and data analysis are solved; and a train diagram parameter correction system is constructed. Furthermore, the DaTong-Jungar railway is taken as an example to illustrate the use of the proposed diagram parameter correction method. The case study results show that the method is simple and easy to operate with high accuracy; and the significant promotion has verified its validity.