活塞环-缸套动力润滑分析及结构优化设计  

Hydrodynamic lubrication analysis of piston ring-liner system and optimized design of structure

作　　者：张俊红[1] 李周裕[1] 何振鹏[1] 张桂昌[1] 谢伟松[2] 李林洁[1] 

机构地区：[1]天津大学内燃机燃烧学国家重点实验室,天津300072 [2]天津大学理学院,天津300072

出　　处：《机械设计》2013年第10期56-61,共6页Journal of Machine Design

基　　金：国家自然科学基金资助项目(50975192);教育部博士点基金资助项目(20090032110001)

摘　　要：采用混合润滑模型分析了柴油机活塞环-缸套系统的润滑性能。选取影响润滑性能的5个主要参数进行正交试验,通过极差分析确定了各个因素对活塞环-缸套摩擦功耗影响的主次关系。建立遗传算法优化的BP神经网络,利用正交试验结果训练该神经网络,得到活塞环-缸套的摩擦功耗神经网络预测模型,然后利用该模型针对选取的5个主要参数进行了优化设计。结果表明,影响摩擦功耗的主要因素由强到弱为:缸套表面粗糙度、活塞环桶面高度、活塞环桶面偏移、活塞环表面粗糙度和活塞环轴向高度。运用正交试验和遗传算法优化的BP神经网络相结合的方法进行活塞环优化设计,试验工作量大大减少,预测精度较好,为活塞环-缸套摩擦学设计提供了便利。Piston ring/liner lubrication performance was ana- lyzed by using mixed lubrication model. Five major parameters af- fecting lubrication characteristics were selected for orthogonal exper- iment. The primary and secondary relations of factors that influence friction loss were determined based on range analysis method. The BP neural network optimized by genetic algorithm was establish and trained by using the results of orthogonal experiment, then the pre- diction model of friction loss for piston ring/liner was developed. Optimization design of the five major parameters was carried out. Results show that factors influencing friction loss from strong to weak are cylinder liner surface roughness, barrel height of rings, batrel offset of rings, surface roughness of tings and axial height of tings. Experiment workload is reduced greatly and prediction preci- sion is good when orthogonal experiment and BP neural network op- timized by the genetic algorithm were combined for the purpose of optimization design. Thus advantage makes it convenient for tribo- logical design of piston ring/liner.

关 键 词：活塞环 润滑 正交试验 BP神经网络 遗传算法 优化设计 

分 类 号：TH117.2[机械工程—机械设计及理论]