Hybrid train energy management method and system based on deep reinforcement learning

Abstract

The invention discloses a hybrid train energy management method and system based on deep reinforcement learning. The method comprises: acquiring historical data of the train running speed, the train running environment, the train running energy consumption information and the train running power assembly information of a hybrid train to serve as source data; extracting speed, acceleration and battery capacity from the source data as input, taking an energy management strategy as output, and establishing an energy management strategy model; establishing a train power simulation model, and inputting the energy management strategy into the train power simulation model to obtain a simulation running state and a reward parameter; optimizing the energy management strategy model according to thereward parameters; performing offline training through deep reinforcement learning to obtain an optimized energy management strategy model; and inputting the real-time data of the hybrid train into the optimized energy management strategy model to obtain an optimized energy management strategy. According to the invention, the hybrid power train energy management can be carried out by completely applying a machine learning artificial intelligence means.

Description

technical field [0001] The invention relates to the technical field of energy management of hybrid trains, in particular to an energy management method and system for hybrid trains based on deep reinforcement learning. Background technique [0002] Hybrid Electric Train (HET) is a railway power train that uses a rechargeable energy storage device to assist the traction system. Hybrid rail trains will have a rechargeable energy storage device installed on board, using excess energy from a power source (usually a diesel engine) or electricity recovered from regenerative braking to charge the energy storage device. Hybrid trains have multi-source power sources, which are more complex than traditional trains. Therefore, efficient and energy-saving energy management strategies become the key to achieving hybrid energy conservation and emission reduction. [0003] At present, many simple and regular models are used in commercial applications, and the power consumption is relative...

AI Technical Summary

Problems solved by technology

[0006] The present invention provides a method and system for energy management of hybrid trains based on deep reinforcement learning, which is used to solve the problem of energy management of hybrid trains using a simple rule-based model. Management, technical issues with limited space for fuel economy improvement

Images