Modeling of laser supported contact pattern analysis

齿轮是驱动技术中的关键部件，因此在国内工业中也是如此。齿轮传动的一个重要质量特征是接触模式，它描述了啮合齿轮副上接触区域的形式和位置。即使在今天，它仍然是用色浆法来测量的，这种方法既耗时又难以自动化。为了克服这一缺陷，将研究一种基于激光的方法。它首先对齿轮进行激光加热，然后将热量传递给接触的牙齿。在接触过程中，受累的牙侧会分别升温和降温，红外摄像机可以记录下它们的接触模式。与以前在80年代和90年代进行的牙齿接触模式的热成像调查相比，预计信噪比将显著提高，从而产生一种适合工业用途的检测方法。本研究项目的目的是从理论上检验和评价新方法的可行性。为此，测量过程被分成几个步骤，依次进行分析(直到牙齿接触时刻的热产生和扩散，到第二个受累齿面的热传递，牙齿接触后的热扩散，得到的牙齿温度分布的红外成像)。温度分布的计算将使用有限元方法进行。最后，考虑到红外摄像机的噪声特性，对温度检测进行了仿真。牙齿接触后的热扩散会使接触图案的红外图像或多或少变得模糊。因此，必须从模拟测量中重建原始的接触模式。这将使用神经网络工具来实现。项目结果将为新检测方法的实验验证提供规划依据。

Gears form a key-component in drive technologies and, by this, in domestic industries. An important quality feature of gear drives is the contact pattern, which describes the form and location of contact areas occurring at the meshing gear pair. Even today it is measured by the colour paste method, which is time consuming and difficult to automatise. In order to overcome this deficit, a laser based approach will be examined. It starts with the laser heating of a gear tooth and the subsequent heat transfer to a contacting tooth. The involved tooth flanks will heat up and cool down respectively during the contact, and their contact pattern can be recorded by infrared cameras. Compared to former thermographic investigations of tooth contact patterns carried out in the 80s and 90s, a significantly improved signal-to-noise ratio is expected, leading to an inspection method suitable for industrial purposes. The objective of the research project is to examine and to evaluate the feasibility of the new method theoretically. For this, the measurement process is divided into several steps which are analysed successively (heat generation and -diffusion until the moment of teeth contact, heat transfer to the second involved tooth flank, heat diffusion after the teeth contact, infrared imaging of the resulting teeth temperature distributions). The calculation of the temperature distribution will be performed using the Finite Elements Method. Finally, the temperature detection will be simulated, considering the noise characteristics of infrared cameras. The heat diffusion after the teeth contact will cause more or less blurred infrared images of the contact patterns. Hence, the original contact patterns have to be reconstructed from simulated measurements. This will be realised using neural network tools. The project results will form a planning basis aiming at an experimental verification of the new inspection method.