| 李照宇,晋刚,王飞,等.基于多体动力学仿真的发动机活塞裙部曲面多目标优化设计[J].内燃机工程,2025,46(5):121-129. |
| 基于多体动力学仿真的发动机活塞裙部曲面多目标优化设计 |
| Multi-Objective Optimization Design of Engine Piston Skirt Surface Based on Multi-Body Dynamics Simulation |
| DOI：10.13949/j.cnki.nrjgc.2025.05.013 |
| 关键词:柔性多体动力学 活塞裙部曲面设计 响应面模型 多目标优化 |
| Key Words:flexible multi-body dynamics piston skirt surface design response surface model multi-objective optimization |
| 基金项目: |
|
| 摘要点击次数: 417 |
| 全文下载次数: 203 |
| 摘要:针对某款4缸发动机搭建了活塞柔性多体动力学模型，提出活塞裙部曲面多目标优化设计方法。建立活塞温度场并通过试验验证，利用Optistruct对活塞缸套有限元模型进行自由度缩减，通过活塞裙部横向曲线和纵向曲线的关系将活塞裙部曲面参数化，收集相关参数在AVL EXCITE中建立活塞柔性多体动力学模型并通过试验验证模型可靠性。再以一个周期内活塞二阶运动、活塞裙部与缸套间油膜厚度和摩擦损失为优化目标，将活塞裙部曲面参数进行拉丁超立方采样。采用不同响应面模型对仿真模型进行拟合，并选取精度最高的径向基函数神经网络响应面模型，结合非支配排序遗传算法（non-dominated sorting genetic algorithm Ⅱ， NSGA-Ⅱ）算法对活塞裙部曲面进行多目标优化，得到Pareto解集并选取其中所需解进行对比验证。优化后活塞最大径向位移减少1.16%，最大倾斜角减小2.32%，最小油膜厚度增加6.00%，摩擦损失降低2.63%。 |
| Abstract:A flexible multi-body dynamics model of a four-cylinder engine was established， and a multi-objective optimization design method for the piston skirt surface was proposed. The temperature field of the piston was established and verified through experiments. Then the degrees of freedom of the finite element models of the piston and cylinder liner were reduced using OptiStruct， and the surface of the piston skirt was parameterized via the relationship between the transverse and longitudinal curves of the piston skirt. Relevant parameters were collected to establish the flexible multi-body dynamics model of the piston in AVL EXCITE， and the reliability of the model was confirmed by experiments.Taking into account the piston’s secondary motion， oil film thickness， and friction loss between the piston skirt and cylinder liner within a cycle as optimization objectives， the surface parameters of the piston skirt were sampled using Latin Hypercube sampling. Different response surface models were adopted to fit the simulation model， and the radial basis function neural network （RBFN） response surface model with the highest precision was selected. Subsequently， combined with the non-dominated sorting genetic algorithm Ⅱ（NSGA-Ⅱ）， multi-objective optimization was carried out on the piston skirt surface. The Pareto solution set was obtained and the required solutions therein were selected for comparative verification. After optimization， it was found that the maximum radial displacement of the piston was reduced by 1.16%， the maximum tilt angle was reduced by 2.32%， the minimum oil film thickness increased by 6.00%， and the friction loss decreased by 2.63%. |
|
查看全文 HTML
查看/发表评论 下载PDF阅读器 |