Scholarlink

In order to improve the speed stability and dynamic responsiveness of methanol engine, this study proposes an intelligent control strategy based on deep reinforcement learning (DRL). By applying the flexible actor-critic (SAC) algorithm and deep deterministic policy gradient (DDPG) algorithm, a DRL controller was constructed that can replace the traditional PID speed control loop. Based on the MATLAB/Simulink platform, a high-precision crankcase power output model are established by combining the engine experimental data. Simulation results show that the SAC algorithm exhibits faster response time, lower error level and smaller overshoot rate in speed control.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Deep Reinforcement Learning Based Speed Control of Pure Methanol Engine

  • 会议论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年

查看全文
  • Yafeng Ren,
  • Jiaxin Qin,
  • Hong Jia

摘要

In order to improve the speed stability and dynamic responsiveness of methanol engine, this study proposes an intelligent control strategy based on deep reinforcement learning (DRL). By applying the flexible actor-critic (SAC) algorithm and deep deterministic policy gradient (DDPG) algorithm, a DRL controller was constructed that can replace the traditional PID speed control loop. Based on the MATLAB/Simulink platform, a high-precision crankcase power output model are established by combining the engine experimental data. Simulation results show that the SAC algorithm exhibits faster response time, lower error level and smaller overshoot rate in speed control.