Session: 02-01-02: Product and Process Design

Paper Number: 144707

144707 - Linking Early Robust Design With Quantitative Methods for Enhanced Industrial Applications 

Robust Design (RD) aims to design systems that are insensitive to various sources of deviation. Implementing RD in the early stages can reduce costly iterations during later product development. Despite the numerous approaches that have been developed to improve the robustness of product concepts in the early stages, current research on early RD methods is often based on simplified, largely context-dependent academic case examples, which hinders the transferability of these methods to industrial applications.

This paper addresses this gap by investigating the application of an RD method in an industrial context, focusing on a dual-mass flywheel that is widely used in vehicle transmission systems. The study aims to evaluate the transferability of early RD methods and to identify practical challenges encountered in industrial applications. The Centrifugal Pendulum Vibration Absorber (CPVA) is a critical add-on component in dual-mass flywheels. This is because the absorption of a specific vibration order is closely related to the tuning order of the CPVA. To evaluate the robustness of different CPVA concepts, the RD method with the Embodiment Function Relation and Tolerance (EFRT) model is used. Utilizing this model provides an early robustness evaluation by systematically analyzing the geometric relationships in the product concepts and the system behavior under deviations. The findings are then compared with sensitivity simulations performed in Matlab Simulink to validate the results. The simulation incorporates the dynamic equations of the CPVAs into the vehicle driveline model to evaluate their vibration-damping effect on the transmission input shaft. Statistical analysis of angular velocity oscillations within a defined speed range, using simulated transient engine torque as input, is performed to evaluate the effect of tuning order deviation on the performance of the CPVAs.

The Results with EFRT modeling show the significant impact of the assembly relationship of the End Stop Damper on the robustness of the present CPVA concept, particularly concerning geometric deviations and high-frequency irregularities. The findings lead to the adaptation of the CPVA concept to improve its robustness, while still meeting the functional requirements. The evaluation of the two concepts is carried out with the EFRT model and further confirmed by sensitivity simulations in Simulink. Simulations show that the dual-mass flywheel system equipped with more robust CPVAs is less affected by geometric deviations and less prone to instability under high-frequency excitation.

These results underscore the potential of early robustness analysis to improve the robustness of complex system concepts, especially in high-speed nonlinear dynamic systems. Nevertheless, this study identifies several challenges in the application of early RD methods that require further investigation, e.g., difficulty in using tests to check the accuracy of modeling and analyzing results for systems operating at high frequencies, and inaccuracies in the particularization of indirectly affected parameters.

By investigating the transferability of early RD methods into industrial cases and identifying practical challenges, this work contributes to enhancing the application of early RD methods in industrial products. Addressing these challenges is critical to advancing the development of early RD methods and ensuring their effective application in industrial contexts.

Presenting Author: Markus Doellken Karlsruhe Institute of Technology (KIT), IPEK – Institute of Product Engineering

Presenting Author Biography: Markus Doellken is a researcher at the Karlsruhe Institute of Technology. His research focuses on design methods.

Authors:

Jiahang Li Karlsruhe Institute of Technology (KIT), IPEK – Institute of Product Engineering
Yiming Long Tongji University, School of Automotive Studies
Markus Doellken Karlsruhe Institute of Technology (KIT), IPEK – Institute of Product Engineering
Yizhe Zhang Tongji University, School of Automotive Studies
Guangqiang Wu Tongji University, School of Automotive Studies
Sven Matthiesen Karlsruhe Institute of Technology (KIT), IPEK – Institute of Product Engineering