【作者】 张宇；

【导师】 杨靖；

【作者基本信息】 湖南大学 ， 动力工程及工程热物理， 2018， 硕士

【摘要】 内燃机作为将化学能转化为机械能的动力机械,其能源利用率的高低除了和缸内燃烧情况有关以外,还和机体内各种摩擦副的工作状况有关。曲轴主轴承是发动机内最主要的摩擦副之一,其润滑性能的好坏直接影响着整机的油耗和工作寿命。本文以某船用高转速汽油机合作项目为依托,基于AVL EXCITE多体动力学仿真软件,结合正交实验设计和智能遗传算法,研究影响曲轴主轴承润滑性能的主次因素并进行多目标优化。本文的主要研究工作及创新点如下:(1)搭建该船用高转速汽油机AVL EXCITE多体动力学模型,通过台架试验,获取模型的边界条件,将仿真实验结果和台架试验后的拆机结果对比,仿真结果中主轴承轴瓦磨损位置和区域与台架试验后的轴瓦磨损区域及状况相同,从而验证了本次AVL EXCITE仿真模型的正确性,增加了后续优化分析的可信度。(2)针对影响主轴承润滑特性的各因素,例如供油压力、机油温度、轴瓦宽度、半径间隙、轴瓦油槽宽度、轴瓦表面粗糙度和主轴颈表面粗糙度等,设计正交实验,采用极差分析法得出影响主轴承润滑性能的主次因素和最优参数组合。(3)结合modeFRONTIER优化软件,针对影响主轴承润滑性能的四个最主要因素进行实验设计和样本数据构建。以最小油膜厚度和最大油膜压力的可靠范围为约束条件,总摩擦功损失和总机油流量最小为目标函数,基于径向基函数神经网络建立响应面近似模型。最后,运用带精英策略的非支配排序遗传算法(NSGA-II)对主轴承润滑性能进行多目标优化。结果表明,采用响应面近似模型得到的润滑优化结果,相比于原机结果以及正交实验得到的优化结果,可以更大幅度地提升主轴承的润滑性能,降低摩擦功损耗,并经过试验测试确定主轴承获得了理想的润滑效果。更多还原

【Abstract】 Internal combustion engine is the power machinery that converts chemical energy into mechanical energy.The level of its energy efficiency is not only related to the combustion conditions in the cylinder but also related to the working conditions of various friction pairs in the engine.The crankshaft main bearing is one of the most important friction pairs in the engine.Its lubricating perf ormance directly affects the fuel consumption and working life of the whole machine.The paper is based on the AVL EXCITE multi-body dynamics simulation software and combined with orthogonal experimental design and intelligence genetic algorithm.It analyzes the primary and secondary factors affecting the lubrication performance of the main bearing of crankshaft and optimizes the multi-objective based on a marine high-speed gasoline engine cooperation project.The main research work and innovation of this paper are as follows:(1)The marine high-speed gasoline engine AVL EXCITE multi-body dynamics model was built and the boundary condition of the model was obtained through bench test.Comparing the simulation results with the disassemble results after the b ench test,the abrasion position and area of the main bearing bush in the simulation were the same as those of the bearing bush after the bench test,which verifie d the correctness of the AVL EXCITE simulation model and increased the credibility of subsequent optimization analysis.(2)Orthogonal test was designed according to the factors affecting the lubrication characteristics of the main bearing,such as oil supply pressure,oil temperature,bearing width,radial clearance,bearing groove width,bush surface roughness and main journal surface roughness.And the primary and secondary factors that affect the lubrication performance of the main bearing and the optimal combination of parameters were obtained through range analysis.(3)Experimental design and sample data construction were conducted for the four most important factors affecting the lubricity of the main bearing with ModeFRONTIER software.An approximate model was established based on the RBFN neural network.It took the reliable range of the minimum oil film thickness and the maximum oil film pressure as the constraint condition and the total frictional power loss and the minimum oil flow rate as the objective functions.Finally,the multi-objective optimization of main bearing lubrication performance was carried out by NSGA-II with elitist strategy.The results showed that the lubrication results obtained by the RSM approximate model greatly improved the lubrication performance of the main bearing and reduced the frictional work loss,compared with the results of the original machine and the optimization results obtained by the orthogonal test.And through the bench test,the main bearing lubrication achieved a desired effect.更多还原