Development of prediction models for deviating, thermo-elastohydrodynamically lubricated contacts (DeviaTEHD)

点击翻译按钮获取中文摘要

The reduction of friction losses in lubricated contacts of machine elements is an important strategy for the development of energy-efficient and wear-resistant technical products. In addition to the lubrication condition, the type of motion or the geometry, one cause of friction losses is deviations of the surface topography of the components, which occur during manufacturing. The deviations of the surface can contribute to friction losses due to the waviness and structure of the surface to varying degrees. In order to enable the development of energy-efficient machine elements with lubricated contacts, an exact calculation of the existing thermo-elastohydrodynamic (TEHD) contacts is necessary for the design, taking into account the deviations from production.Extensive, analytically solvable approximations for the calculation of lubrication film and pressure parameters for infinite line and three-dimensional point contacts already exist. However, these do not take sufficiently into account manufacturing deviations that occur in every component. At the same time, there are reliable but time-consuming simulation tools for TEHD contact calculation. In the context of the product development process and when used in multi-body or system simulations, however, a simple and time-efficient application is an essential requirement for the used calculation model.Due to missing approximations for the exact calculation of deviated contacts, the aim of the planned research project is the development of predictive models for the determination of the central and minimum lubrication film thickness as well as the maximum pressure and temperature in concentrated contacts under consideration of surface deviations. For this purpose, wave-shaped deviations of the surface are parametrically integrated into TEHD simulation models for the infinite two-dimensional line contact as well as the three-dimensional point contact. Using the chair's own TEHD simulation tool, TriboFEM, these two models are analyzed to determine the effects on the formation of the lubricant film and the distribution of pressure in the contact and to generate a database. Based on this database, predictive models for the infinite line and three-dimensional point contact are developed by introducing an additional correction factor to the approximations of DOWSON/HIGGINSON or BLOK/MOES, by more complex methods such as mathematical regression and modern methods of machine learning. By evaluating the prediction models on criteria such as accuracy, usability and time efficiency, a suitable prediction model is selected for line and point contact. At the end of this research project, an application example will be used to demonstrate its applicability in product development.