周龙,李乐,王立勇

• 1:北京信息科技大学机电工程学院

摘要(Abstract)：

为提升对超越离合器关键部件磨损状态的评估精度，提出了基于粒子群优化-反向传播（particle swarm optimization-back propagation, PSO-BP）模型的超越离合器磨损预测算法。首先，基于ABAQUS有限元仿真平台，结合Archard磨损准则并引入用户子程序UMESHMOTION，建立了磨损有限元模型。通过将磨损量映射至网格节点位移，实现了材料去除效应的几何边界演化与磨损深度的动态更新，并结合试验验证了模型的准确性。其次，在此基础上，进一步提取累计滑移距离、瞬时滑移距离与接触应力等关键参数作为输入，以磨损深度作为输出，构建PSO-BP预测模型，实现对楔块与内环接触区域磨损深度的高精度预测。对比分析表明，与传统BP神经网络和GA-BP神经网络相比，PSO-BP模型的拟合精度为99.838%，其预测误差更集中于零误差区域，体现出较高的准确性。

关键词(KeyWords)： 超越离合器;磨损深度;有限元分析;磨损预测

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金项目（52175074）

作者(Author): 周龙,李乐,王立勇

DOI: 10.16508/j.cnki.11-5866/n.2026.01.008

参考文献(References)：

• [1]刘娟，沈火明.扭动微动摩擦力数值分析[J].机械设计，2015,32(5):78-81.LIU J,SHEN H M. Numerical simulation analysis for torsionalfretting friction force[J]. Journal of Machine Design,2015,32(5):78-81.(in Chinese)
• [2]余江鸿，马文涛，易永胜，等.基于Archard模型的飞行器折叠翼含间隙铰链磨损特性研究[J/OL].润滑与密封，2024:1-10.(2024-10-30)[2025-07-23]. https：//link. cnki. net/urlid/44. 1260. th. 20241029. 0926. 002.YU J H,MA W T,YI Y S,et al. Research on wear characteristics of folding wings with clearance hinge in aircraft based on Archard model[J/OL]. Lubrication Engineering,2024:1-10.(2024-10-30)[2025-07-23]. https：//link. cnki. net/urlid/44. 1260. th. 20241029. 0926. 002.(in Chinese)
• [3]韩怡茗，李欣，刘乐强，等.微动磨损形貌与接触应力场耦合机制仿真分析[J].机床与液压，2024,52(19):135-140.HAN Y M,LI X,LIU L Q,et al. Simulation analysis of coupling mechanism between fretting wear scar and contact stress field[J]. Machine Tool&Hydraulics,2024,52(19):135-140.(in Chinese)
• [4]黄帅宗.基于粗糙面的切向微动磨损数值分析[D].成都：西南交通大学，2020.HUANG S Z. Numerical analysis of tangential fretting wear based on rough surface[D]. Chengdu:Southwest Jiaotong University,2020.(in Chinese)
• [5]李娟，夏建中，孙志礼，等.磨损预测模型的试验研究[J].现代制造工程，2011(5):117-119.LI J,XIA J Z,SUN Z L,et al. Experimental study of wear rate prediction model[J]. Modern Manufacturing Engineering,2011(5):117-119.(in Chinese)
• [6]宋剑，王松，廖振华，等.人工神经网络方法在摩擦学领域的应用研究进展[J].润滑与密封，2016,41(11):117-123.SONG J,WANG S,LIAO Z H,et al. Research progress of artificial neural network application in tribology field[J].Lubrication Engineering,2016,41(11):117-123.(in Chinese)
• [7]范杰.地铁弓网系统载流摩擦磨损试验研究及其磨损率预测模型[D].成都：西南交通大学，2022.FAN J. Experimental study on the current carrying friction and wear performance of a metro pantograph-catenary system and their wear prediction model[D]. Chengdu:Southwest Jiaotong University,2022.(in Chinese)
• [8]吕德峰，左洪福，蔡景，等.一种磨损预测的优化算法研究[J].摩擦学学报，2008,28(6):562-566.L??D F,ZUO H F,CAI J,et al. Research on a optimal algorithm for the prediction of wear loss[J]. Tribology,2008,28(6):562-566.(in Chinese)
• [9]于康宁.基于机器视觉的微磨具磨损在位检测与预测方法[D].长沙：湖南大学，2022.YU K N. In-situ detection and prediction method of microgrinding tools wear based on machine vision[D]. Changsha:Hunan University,2022.(in Chinese)
• [10]田苗，于康宁，任莹晖，等.基于GA-BP神经网络的微磨具磨损预测研究[J].金刚石与磨料磨具工程，2024,44(3):363-373.TIAN M,YU K N,REN Y H,et al. Wear prediction of microgrinding tool based on GA-BP neural network[J]. Diamond&Abrasives Engineering,2024,44(3):363-373.(in Chinese)
• [11]郭宏，徐延，伊亚聪，等.特征融合与BP神经网络结合的刀具磨损预测[J].机械设计与制造，2025,407(1):108-111.GUO H,XU Y,YI Y C,et al. Tool wear prediction based on feature fusion and BP neural network[J]. Machinery Design&Manufacture,2025,407(1):108-111.(in Chinese)
• [12]张永芳，侯丕鸿，杨鑫亮，等.基于多尺度-组合式循环神经网络的内燃机缸套表面磨损量预测方法[J].摩擦学学报（中英文），2025,45(2):315-324.ZHANG Y F,HOU P H,YANG X L,et al. A prediction method of surface wear in internal combustion engine cylinder liners using multi-scale combined recurrent neural networks[J].Tribology,2025,45(2):315-324.(in Chinese)
• [13]LIN Z W,FAN Y K,TAN J L,et al. Tool wear prediction based on XGBoost feature selection combined with PSO-BP network[J]. Scientific Reports,2025,15(1):3096.
• [14]SHI J,ZHANG Y Y,SUN Y H,et al. Tool life prediction of dicing saw based on PSO-BP neural network[J]. The International Journal of Advanced Manufacturing Technology,2022,123(11):4399-4412.
• [15]DZIERWA A,PAWLUS P. Wear of a rough disc in dry sliding contact with a smooth ball:experiment and modeling[J].Archives of Civil and Mechanical Engineering,2021,21(2):85.
• [16]KASAREKAR A T,SADEGHI F,TSEREGOUNIS S. Fretting fatigue of rough surfaces[J]. Wear,2008,264(7/8):719-730.
• [17]PHAM-BA S,MOLINARI J F. Adhesive wear regimes on rough surfaces and interaction of micro-contacts[J]. Tribology Letters,2021,69(3):107.
• [18]MCCOLL I R,DING J,LEEN S B. Finite element simulation and experimental validation of fretting wear[J]. Wear,2004,256(11-12):1114-1127.
• [19]MAGAZINER R,JIN O,MALL S. Slip regime explanation of observed size effects in fretting[J]. Wear,2004,257(1-2):190-197.
• [20]HEGADEKATTE V,KURZENH??USER S,HUBER N,et al. A predictive modeling scheme for wear in tribometers[J].Tribology International,2008,41(11):1020-1031.
• [21]JAIN M,SAIHJPAL V,SINGH N,et al. An overview of variants and advancements of PSO algorithm[J]. Applied Sciences,2022,12(17):8392.
• [22]陈永杰，周新聪，贺石中，等.基于U-Net和LSTM的在线磨粒图像分割和磨损预测[J].润滑与密封，2025,50(5):103-108.CHEN Y J,ZHOU X C,HE S Z,et al. Online debris image segmentation and wear prediction based on U-Net and LSTM[J].Lubrication Engineering,2025,50(5):103-108.(in Chinese)
• [23]赵紫岩.基于DOE和代理模型的结构优化设计方法及应用研究[D].西安：西安建筑科技大学，2017.ZHAO Z Y. Research on the application and structural optimization design method based on DOE and surrogate model[D]. Xi′an:Xi′an University of Architecture and Technology,2017.(in Chinese)
• [24]杨智勇，陶源彪，陈奕帆，等.基于GA-BP神经网络的G20CrNi2MoA轴承套圈渗碳热处理工艺优化[J/OL].清华大学学报（自然科学版），2025:1-11.(2025-12-29)[2026-01-07]. https：//doi. org/10. 16511/j. cnki. qhdxxb. 2026. 27. 013.YANG Z Y,TAO Y B,CHEN Y F,et al. Optimization of carburizing heat treatment process for G20CrNi2MoA bearing ring based on genetic algorithm-backpropagation neural network[J/OL]. Journal of Tsinghua University(Science and Technology)2025:1-11.(2025-12-29)[2026-01-07]. https：//doi. org/10. 16511/j. cnki. qhdxxb. 2026. 27. 013.(in Chinese)
• [25]刘畅，王立勇，吴健鹏，等.基于BAS-BP的永磁同步电机损耗预测方法[J].北京信息科技大学学报（自然科学版），2025,40(5):45-53.LIU C,WANG L Y,WU J P,et al. Loss prediction method for permanent magnet synchronous motors based on BAS-BP[J].Journal of Beijing Information Science&Technology University(Science and Technology Edition),2025,40(5):45-53.(in Chinese)