A high-speed train intelligent driving control method based on historical operation data

Abstract

The invention provides a high-speed train intelligent control method based on historical operation data, and the method comprises the following steps: carrying out the attribute definition of the operation states of a train, and carrying out the clustering analysis of the historical operation data of the high-speed train according to the defined attributes; finding out a data set similar to the inter-station operation process of the current train from the historical operation data according to the operation state and the similarity index of the current train; converting the future inter-station operation process of the current train into a Markov decision process according to the driving task information of the current train; searching a plurality of historical operation states with the highest similarity of the future inter-station operation process of the current train according to the weighted Euclidean distance; using a Monte Carlo method for sampling to obtain a reference runningtrack, using a reinforcement learning method to optimize the reference running track of the train in the future inter-station whole running process, and obtaining a train intelligent control strategyto intelligently control the train. The method can realize automatic driving of the train and improve the operation service quality.

Description

technical field [0001] The invention relates to the technical field of rail transit train automatic driving, in particular to an intelligent driving control method for high-speed trains based on historical operating data. Background technique [0002] In recent years, high-speed railways have been widely recognized by countries all over the world for their significant technical and economic advantages such as high speed, large capacity, all-weather, low energy consumption, and low emissions. The construction of high-speed railways in my country is also in full swing. As of the end of 2017, the country's railway operating mileage was 127,000 kilometers, of which 25,000 kilometers were high-speed railways. [0003] The train operation control system based on manual driving, referred to as the train control system, plays an important role in ensuring the operation safety and transportation efficiency of high-speed railways. At present, the typical train control systems used in...

AI Technical Summary

Problems solved by technology

[0005] Therefore, although the urban rail ATO system has been promoted in practical applications, if the existing traditional ATO algorithm is directly introduced into the operation control of high-speed railway trains, it will face many technical problems: First, the traditional ATO algorithm uses PID (proportional -Integral-Derivative controller, Proportion-Integral-Derivative controller) control, so that although the target curve can be accurately tracked under the condition of ensuring time on time, due to the frequent conversion of traction braking working conditions during the tracking process, the passenger comfort is poor and the system High energy consumption; second, the operating environment of high-speed railways is complex, and the operating parameters of high-speed trains are time-varying and strongly nonlinear. The traditional driving control method based on determining the train model has a relatively small cumulative error during long-term operation. large, and the control accuracy is not high, which affects the punctuality of high-speed train driving control; third, due to the poor intelligence of the existing ATO algorithm, the control strategy cannot be adjusted online. , the train control mode must be switched to manual driving to ensure that the train runs on time

Images