Adaptive Damage Accumulation and Remaining-Service-Life-Prediction for Gearboxes

当今运行的大多数机器和机械部件都是为有限的使用寿命而设计的。一方面，这为提高效率、降低成本和减少碳足迹创造了巨大的潜力。另一方面，这会导致故障风险随着机器年龄的增加而增加。尤其是如果失败可能对人类或环境造成严重破坏，或者可能导致严重的经济损失，这一事实就会产生目标冲突。一方面，机器应该只在必要时进行维护或更换，另一方面，故障的可能性增加。因此，有一种方法是可取的，使之能够尽可能少地预测剩余的使用寿命和健康状态。变速箱是许多技术系统的基本部件，例如机器人、风力涡轮机和带有电动或内燃机的汽车。变速箱的核心部件是齿轮。这些部件的故障通常会导致整个变速箱出现故障。疲劳寿命分析涉及根据预期载荷和所需使用寿命来确定齿轮的尺寸。不幸的是，目前在运行期间验证技术设计的可能性很小。因此，本研究方案的目标是创造一种方法，能够预测齿轮在运行过程中的剩余使用寿命和健康状态。由于传感器在机器中的分布越来越广，记录的操作数据越来越多，因此设计方法的目的是使结果能够轻松地传输到其他机器元件。这项研究计划的目标之一是调查使用大数据和机器学习方法分析这些数据的潜力。计划进行实验研究，以验证主要基于机器学习的剩余寿命预测，并研究机器学习在传统疲劳寿命分析中的机会。最后，根据这些原理建立的剩余使用寿命预测可以减少资源浪费，并进一步提高机器的安全性。

Most machines and machine elements operating today are designed for a limited service-life. On the one hand, this creates significant potential for increasing efficiency, decreasing cost and reduction of the carbon-footprint. On the other hand, this causes a rising risk of failure with increasing age of the machine. Especially if a failure can cause serious damage to humans or the environment or can result in a high economic loss, this fact creates a conflict of goals. On the one hand, the machine should only be maintained or replaced when necessary and on the other hand, the probability of a failure increases. Therefor a method is desirable, making it possible to predict the remaining service-life and state of health with as little effort as possible.Gearboxes are an elementary component of many technical systems, for example robots, wind turbines and cars with electric or combustion engine. Centerpiece of gearboxes are the gears. A failure of these components usually causes the whole gearbox to fail. The fatigue life analysis deals with the dimensioning of gears according to the expected loads and the required service-life. Unfortunately, there is very little possibility to validate the technical design during operation at the moment. Hence, the goal of this research proposal is to create a method, enabling the prediction of the remaining-service-life and state-of-health of gears during operation. It is planned to design the method in a way, enabling an easy transfer of the results to other machine elements.Because of the increasing spread of sensors in machines, more and more operating-data is recorded. One goal of this research proposal is to investigate the potential of analyzing this data with big-data- and machine learning methods. Experimental investigations are planned to validate a remaining-service-life-prediction, mostly based on machine learning and to investigate the opportunities of machine learning for the traditional fatigue life analysis. Finally, a remaining-service-life-prediction created according to these principals can reduce waste of resources and can furthermore increase the safety of machines.