活塞-缸套弹流碰撞行为的动态流-固耦合建模及状态监测研究

To implement the non-intrusive and online monitoring of the lubrication conditions between the piston and cylinder assembly in an internal combustion engine, the project proposes a novel method of extracting tribo-dynamic parameters: stiffness and damping based on identifying and extracting the dynamic parameters of the lubricating film by analyzing the vibration response of elasto-hydrodynamic (EHD) impacts. Firstly, the evolutional rule of the dynamic characteristics of the lubricating film under the coupled action of the second-order movement of structure surface and various physical and chemical factors is studied. Combining structural dynamics, computational fluid dynamics and molecular dynamics, a cross-scale fluid-solid coupling model is established to describe and study the nonlinear dynamic characteristics of oil film in the EHD contact of the friction surfaces, and to reveal the influence of structural kinematic parameters on dynamic characteristics of oil film. Then, the influence of oil film dynamic properties on the response characteristic of impact-induced vibration is studied. The dynamic fluid-structure interaction model of the piston-cylinder impaction is established to study the mapping relationship between the dynamic properties of oil films and the characteristics of external vibration responses. Finally, based on the physical model, the clustering analysis rules are formulated to aggregate the responses into a number subgroups for extending average correlation signals based stochastic subspace identification (ACS-SSI) method to the nonlinear system identification, thus reliably extracting oil film dynamic parameters from noisy vibration signals for monitoring tribological behaviour. Through the research of the project, the coupling characteristics of structural vibration response with oil film dynamic behaviour will be understood, which will overcome the main difficulties that limit the use of vibration analysis technique in the field of lubrication condition monitoring. It has a great significance for lubrication improvement, reduction of frictional losses in the reciprocating machinery and the efficiency promotion of energy utilisation.

为了实现内燃机活塞-缸套组件润滑状态的非介入式监测，本项目提出一种通过分析弹流碰撞振动响应，识别提取润滑油膜动力学参数，实现摩擦副间润滑状态监测的方法。首先，探索结构二阶运动与多种理化因素耦合作用下油膜动力学特性的演化规律。结合结构动力学、连续流体力学和分子动力学构建跨尺度流-固耦合模型，研究揭示结构运动参数对油膜动力学特性的影响机理。然后，探索油膜动力学特性对碰撞振动响应特征的影响。建立活塞-缸套弹流碰撞的动态流-固耦合模型，研究明确油膜动力学参数与振动响应特征之间的映射关系。最后，基于物理模型制定振动响应聚类分析规则，改进随机子空间模态特征识别方法的普适性，实现油膜动力学参数的精确识别与提取。通过本项目研究，将在阐明结构振动响应与油膜动力学行为耦合作用机理的基础上，攻克目前制约碰摩振动分析在润滑状态监测领域应用的主要难点，对往复机械改善润滑，减摩增效，提高能源利用率，具有重要的意义。

为实现内燃机活塞-缸套组件润滑状态的非介入式在线监测，项目研究了基于弹流碰撞振动响应分析进行润滑油膜动力学参数识别与活塞-缸套润滑状态监测的方法。首先，对多因素耦合作用下润滑油膜动力学特性进行了研究，将无量纲径隙比参数引入了弹流碰撞动力学建模，着重研究了碰撞速度、等效接触半径、油液黏度等因素对油膜阻尼特性的影响。基于球体撞击油膜高速摄影试验对影响润滑油膜动力学特性的因素进行了对比分析。然后，结合有限元法和动力学微分方程建立了活塞-缸套瞬态流-固耦合模型，就油膜挤压阻尼效应对缸套振动响应特性的影响规律进行了理论研究和试验验证。最后，针对非稳态振动信号开展了时频分割降噪和特征降维筛选，以识别表征润滑状态的局域特征参数。研究了内燃机运行工况、缸内燃烧状态、润滑油膜厚度等因素对缸套振动响应特性和摩擦润滑状态的影响规律。. 通过本项目研究，对润滑油膜与振动表面之间的动态流-固耦合机理有了比较客观的认识，掌握了油膜动力学特性对活塞-缸套弹流碰撞振动响应的影响规律，开发了润滑状态相关振动特征参数识别方法，实现了基于振动信号分析的摩擦副间润滑状态非介入式监测。项目的研究成果为内燃机润滑状态监测和磨损故障预警系统开发提供了理论支撑。