Heavy haul train reinforcement learning control method and system

Abstract

The invention relates to a heavy haul train reinforcement learning control method and system, and relates to the technical field of heavy haul train intelligent control, and the method comprises the steps: obtaining the running state information of a heavy haul train at a current moment; according to the running state information of the heavy haul train at the current moment and a heavy haul trainvirtual controller, obtaining a heavy-haul train control instruction at a next moment, and sending the heavy haul train control instruction at the next moment to a heavy haul train control unit to control the heavy haul train to run; the heavy haul train virtual controller is obtained by training a reinforcement learning network according to heavy haul train running state data and an expert experience network, wherein the reinforcement learning network comprises a control network and two evaluation networks, and the reinforcement learning network is constructed according to an SAC reinforcement learning algorithm. According to the invention, the heavy haul train can have the properties of safety, stability and high efficiency in the running process.

Description

technical field [0001] The invention relates to the technical field of intelligent control of heavy-duty trains, in particular to a method and system for intensive learning control of heavy-duty trains. Background technique [0002] At present, China's heavy-haul railway lines are constantly expanding, and heavy-haul railway transportation has a very important strategic and economic position in rail transportation. At present, the operation of heavy-duty trains depends on the driver's experience and technology, and the heavy-duty lines have the characteristics of long distances and complicated line conditions, and heavy-duty trains carry large loads and many combined vehicles, which have a great impact on the driver's control level and mental state. test. In order to make heavy-haul trains run safely and on time, better control strategies are needed to control the running process of heavy-haul trains. Therefore, the modeling and control of heavy-duty trains has become the ...

AI Technical Summary

Problems solved by technology

Tracking controllers often use the classic PID control algorithm, but the PID control algorithm relies on manual adjustment in the selection of parameters, which is not suitable for complex process control of heavy-duty trains

In this regard, some people propose to use the generalized predictive control algorithm to realize the speed tracking control of heavy-duty trains, but the calculation of the generalized predictive control algorithm is tedious and complicated, and it does not perform well in the real-time response heavy-duty train control system

Some people combine the automatic parking control of heavy-duty trains with fuzzy control, but the fuzzy rules and membership functions in fuzzy control are derived from experience, and it is difficult to control and calculate during the operation of heavy-duty trains.

[0004] With the development of artificial neural networks, some people have proposed a data-driven control method for heavy-duty trains. However, ordinary neural network training requires a large amount of actual data, and the uneven distribution of actual data samples will lead to overfitting of the trained controller. There are more changes in the actual scene, and the state of the trained controller is prone to collapse when the sample space is unknown, which is very risky in practical applications

Images