中国机械工程 ›› 2025, Vol. 36 ›› Issue (10): 2335-2342.DOI: 10.3969/j.issn.1004-132X.2025.10.022

• 机械基础工程 • 上一篇    

基于强化学习自适应鲁棒控制的异构双阀协调控制策略研究

苏世杰1(), 程泳钦1, 胡毅1,2, 何建辉1, 杨书吉1   

  1. 1.江苏科技大学机械工程学院, 镇江, 212003
    2.航天晨光股份有限公司, 南京, 211100
  • 收稿日期:2024-09-10 出版日期:2025-10-25 发布日期:2025-11-05
  • 通讯作者: 苏世杰
  • 作者简介:苏世杰*(通信作者),男,1981年生,教授、博士。研究方向为电液伺服控制、智能控制以及仿生机械设计等。E-mail:sushijie@just.edu.cn
    第一联系人:闫祖龙,男,1999 年生,硕士研究生。研究方向为基于深度学习的精密加工光学元件的表面形貌预测和性能分析。E-mail:yanzulong@njfu.edu.cn。庞启龙*(通信作者),男,1979 年生,副教授。研究方向为表面微观结构的分析与重构、parylene镀膜工艺与设备、分子动力学仿真。E-mail:qlpang@njfu.edu.cn
  • 基金资助:
    江苏省高等学校基础学科研究重大资助项目(23KJA40005);江苏省镇江市产业前瞻与共性关键技术项目(GY2024003)

Coordination Control of Dual-valve Electrohydraulic Servo Systems Based on Integration of Reinforcement Learning and Adaptive Robust Control

Shijie SU1(), Yongqin CHENG1, Yi HU1,2, Jianhui HE1, Shuji YANG1   

  1. 1.School of Mechanical Engineering,Jiangsu University of Science and Technology,Zhenjiang,Jiangsu,212003
    2.Aerosun Co. ,Ltd. ,Nanjing,211100
  • Received:2024-09-10 Online:2025-10-25 Published:2025-11-05
  • Contact: Shijie SU

摘要:

采用小流量伺服阀和大流量比例阀并联驱动同一个执行元件的异构双阀电液伺服系统具有成本低、流量大、精度高的优势,但当系统参数发生变化或受到内外扰动时,其控制性能和稳定性会急剧下降。为此提出一种融合强化学习SAC(soft actor-critic)算法与自适应鲁棒控制算法(ARC)的异构双阀协调控制策略。该控制策略一方面通过设计的流量分配策略减小比例阀与伺服阀在工作切换中产生的瞬态误差,另一方面通过上层SAC算法学习目标电液伺服系统的动态非线性特性,进而实现对下层ARC算法控制参数的动态调节,以增强系统的控制性能与鲁棒性。该研究为后续的仿真和实验验证提供了坚实的理论基础。

关键词: 电液伺服系统, 双阀并联控制, 流量分配, 伺服控制

Abstract:

The dual-valve electrohydraulic servo systems, which employed a small-flow servo valve and a large-flow proportional valve to drive the same actuator in parallel, offered advantages such as low cost, high flow rate and high accuracy. However, the control performance of the systems was compromised by parameter uncertainties, system nonlinearities and disturbances. To address these issues, a dual-valve coordinated control strategy was proposed which integrates the SAC reinforcement learning algorithm with the ARC algorithm. This control strategy aimed to reduce transient errors generated by the proportional valves and the servo valves during work switching through a specifically designed flow allocation strategy. Additionally, the upper SAC algorithm learned the dynamic nonlinearities of the target electrohydraulic servo systems. Consequently, the control parameters of the lower ARC algorithm were dynamically adjusted, thereby enhancing the system's control performance and robustness.The findings of this study establish a solid theoretical foundation for subsequent simulation and experimental validation.

Key words: electrohydraulic servo system, dual-valve parallel control, flow allocation, servo control

中图分类号: