A Fuel Cell Vehicle Energy Management Method Based on Deep Reinforcement Learning Algorithm

Abstract

The invention relates to a fuel cell vehicle energy management method based on a deep reinforcement learning algorithm, belonging to the field of new energy vehicles. The method includes: S1: Obtaining fuel cell vehicle status information; S2: Building a fuel cell vehicle energy management system model; S3: Using a deep reinforcement learning algorithm to construct a fuel cell vehicle energy management strategy, and solving multiple problems including fuel economy and fuel cell efficiency. Objective optimization problem, so as to obtain the optimal energy allocation result. The invention applies the deep reinforcement learning algorithm to the fuel cell vehicle energy management system, which has good optimization and real-time performance; meanwhile, the work efficiency of the fuel cell is considered in the reward function, providing a new idea for energy management.

Description

technical field [0001] The invention belongs to the field of new energy vehicles, and relates to a fuel cell vehicle energy management method based on a deep reinforcement learning (DQN) algorithm. Background technique [0002] At present, traditional automobiles are facing problems such as environmental pollution, global warming, and limited oil resources, making automobile manufacturers turn their attention to the research of hybrid electric vehicles, electric vehicles and fuel cell vehicles. As a transition model from traditional cars to future clean cars, hybrid vehicles usually consist of energy storage systems, electric motors and internal combustion engines, which still consume fuel oil and generate pollution. At the same time, due to the limited driving distance and long charging time of electric vehicles composed of batteries and electric motors, it has become a major obstacle to their commercialization. Therefore, with the development of fuel cell technology, the ...

AI Technical Summary

Problems solved by technology

However, the energy management strategies mentioned above are difficult to satisfy real-time performance and optimality at the same time. For example, although energy management based on rules and local optimization can be applied to real-time control, its optimality cannot be guaranteed; Although the energy management strategy can obtain the global optimal solution, the amount of calculation is too large to be applied to real-time control of real vehicles

Images