Topology Driven Design of Under-Hood Automotive Components for Optimal Weight and NVH Attributes 2019-01-0834

Weight is a major factor during the development of Automotive Powertrains due to stringent fuel economy requirements. Light weighting constitutes a challenge to the engineering community when trying to deliver quieter powertrains. For this reason, the NVH (Noise Vibration Harshness) CAE engineers are adopting advanced vibro-acoustic simulation methods combined with topology optimization methods to drive the design of the under hood components for Noise Vibration and Harshness. Vibro-acoustic computational methods can be complex and require significant computation effort. Computation of Equivalent Radiated Power (referred to as ERP) is a simplified method to assess maximum dynamic radiation of components for specific excitations in frequency response analysis which in turn affects radiated sound. Topology Optimization is a mathematical technique used to find the best material distribution for structural systems in order to deliver a specific objective under clearly defined constraints.

This paper will showcase the process adopted to optimize the weight of an under-hood automotive component while maintaining the ERP performance across several one third octave bands. In order to address the large eigenvalue problem solved during the ERP load-case, a multi-threaded approximate Eigen solver - AMSES (Automatic Multi-level Sub-structuring Eigen solver Solution) is used.

This paper highlights the methodology and best practices identified during ERP analysis and topology optimization of the front cover of an inline gasoline engine. In order to obtain high fidelity results during the optimization process, the engine model was condensed to a modal superelement model. ERP performance of residual model was validated to match the results of the complete model. Role of design optimization parameters in controlling manufacturability of topology driven concept design process will be illustrated with respect to the system under consideration. Techniques adopted to recover modified geometry resulting from a structural optimization, for further use in the design re-evaluation of the front cover are presented.