Abstract
The excellent dynamic response performance of the proton exchange membrane fuel cell (PEMFC) system provides an important guarantee for quickly and accurately following the power demand of fuel cell vehicles. Insufficient air supply during load changes is the main reason for poor dynamic response. However, due to the strong coupling characteristics of flow and pressure, the existing technology has shortcomings such as poor anti-interference ability, slow decision-making speed and unstable control performance, which limit the optimization effect. To solve the above problem, this paper proposes a control method for air supply subsystem to optimize PEMFC dynamic response performance based on deep reinforcement learning. First, a model of the air supply subsystem is established, and its accuracy is verified through PEMFC system experimental bench. Second, the relationship law between the control parameters of air supply subsystem and the dynamic response performance of PEMFC is revealed. Finally, an air supply subsystem control architecture based on Soft Actor-Critic (SAC) is proposed. When the load power is stepped up from 58.5 kW to 101.6 kW, the time for the voltage to reach the steady state under intelligent control is reduced from 1.2 s to 0.7 s, and the electrical efficiency is increased from 72.6 % to 79.1 % compared with traditional decoupling control. The dynamic response performance is improved by 41.6 % and the efficiency is improved by 6.5 %.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
