Scholarlink

<p>To address the complex parameter tuning in electric linear loading systems under varying loads, a composite control method of deep reinforcement learning and PI algorithm is proposed. An improved feedforward strategy tackles force and nonlinearity while cutting complexity. An adaptive PI algorithm (FC-TD3) combines TD3 and the strategy for online tuning. Using the system as subject, optimal parameters from offline learning are applied. Experiments prove it optimizes parameters and boosts system performance.</p>

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

A deep reinforcement learning-based control method for electric linear loading systems

  • 期刊论文

  • 发表日期：

    2025年12月10日

查看全文
  • Wan Xu,
  • Junqi Wang,
  • Dongting Liu,
  • Tingwei Chen

摘要

To address the complex parameter tuning in electric linear loading systems under varying loads, a composite control method of deep reinforcement learning and PI algorithm is proposed. An improved feedforward strategy tackles force and nonlinearity while cutting complexity. An adaptive PI algorithm (FC-TD3) combines TD3 and the strategy for online tuning. Using the system as subject, optimal parameters from offline learning are applied. Experiments prove it optimizes parameters and boosts system performance.