主办：陕西省汽车工程学会
ISSN 1671-7988  CN 61-1394/TH
创刊：1976年
汽车实用技术

汽车实用技术 ›› 2026, Vol. 51 ›› Issue (19): 57-63.DOI: 10.16638/j.cnki.1671-7988.2026.019.008

• 智能底盘集成控制技术 • 上一篇    

基于 PSO-BP 优化模糊 PID 方法的电动汽车 制动动力系统仿真研究

段炆甫 1，付强 1,2，宿达 1   

  1. 1.沈阳航空航天大学 机电工程学院； 2.飞行器快速试制技术研究教育部重点实验室
  • 发布日期:2026-10-08
  • 通讯作者: 段炆甫
  • 作者简介:段炆甫（1999－），男，硕士研究生，研究方向车辆动力学控制
  • 基金资助:
    新能源飞行器喷气动力制动动力学机制及智能控制策略研究（LJKZ0227）

Simulation study of electric vehicle brake power system based on PSO-BP optimized fuzzy PID method

DUAN Wenfu1 , FU Qiang1,2 , SU Da1   

  1. 1.School of Mechatronics Engineering, Shenyang Aerospace University; 2.Key Laboratory of Rapid Development & Manufacturing Technology for Aircraft, Ministry of Education
  • Published:2026-10-08
  • Contact: DUAN Wenfu

PDF

摘要： 针对电动汽车在地面制动过程中存在的非线性、时变性及强耦合特征，文章提出了一 种融合粒子群优化（PSO）改进反向传播（BP）神经网络与模糊比例-积分-微分（PID）的协 同控制策略，用于防抱死制动系统（ABS）/空气反推制动系统（ARTS）复合制动动力系统的 智能控制。该方法利用模糊 PID 实现滑移率的快速响应与粗调，结合 PSO 优化 BP 神经网络 进行精细化增益补偿，有效抑制稳态误差与振荡问题。建立了包括 ARTS 与地面制动力的五 自由度纵向动力学模型，并基于一维定常等熵流动理论构建反推力数学模型。仿真结果表明， 与传统 ABS 系统相比，所提出的控制策略在高附着系数路况下使制动时间缩短了 25.7%，制 动距离缩短了 34.7%，并将滑移率精确稳定在 0.2 附近的理想区域，显著提高了制动效能与稳 定性，验证了该方案在电动汽车地面制动中的应用潜力。

关键词: ABS/ARTS；复合制动系统；模糊 PID；PSO-BP 神经网络

Abstract: To address the nonlinear, time-varying, and strongly coupled characteristics of electric vehicles during ground braking, this paper propsoes a cooperative control strategy integrating particle swarm optimization (PSO)-enhanced back propagation (BP) neural networks with fuzzy proportion-integral-differential (PID) for the intelligent control of anti-lock braking system (ABS)/ air reverse thrust system (ARTS) compsoite braking power systems. This approach utilizes fuzzy PID for rapid response and coarse-tuning of slip rate, while PSO optimizes the BP neural network for fine-tuning gain compensation, effectively suppressing steady-state error and oscillation issues. A five-degree-of- freedom longitudinal dynamic model incorporating the ARTS and ground braking force was established, with a mathematical model for reverse thrust developed based on one-dimensional steady-state isentropic flow theory. Simulation results demonstrate that compared to traditional ABS, the propsoed control strategy reduces braking time by 25.7% and braking distance by 34.7% on high-coefficient-of-adhesion road surfaces. It precisely stabilizes slip rate within the ideal range near 0.2, significantly enhancing braking efficiency and stability. This validates the approach's potential for application in electric vehicle ground braking systems.

Key words: ABS/ARTS; composite braking system; fuzzy PID; PSO-BP algorithm