Generative Design Case Study for Eco-Efficient Vehicles Lower Emissions and Greater Range

EasyChair Preprint 15478

Abstract

The application of AI, or artificial intelligence, offers a novel opportunity to advance generative design in a number of mobility-related fields. The method allows for the creation of parts to be completed in a shorter time frame and has the objective of achieving multiple sustainability goals with minimal information provided at the outset. This article will present a practical case study of this method on an automotive Engine mounting bracket, a complex part that must meet a range of requirements, including those related to rigidity, strength; and sustainability. The initial 3D model will define the space constraints (design space), operating conditions, and objectives to be achieved. These may include, for example, improvements in rigidity, lightweighting and fatigue, reductions in CO2 impact, and improvements in autonomy. The result is a set of designs that respect the constraints and maximize the objectives. Subsequently, the designer may select the design that is most aligned with their preferences, taking into account additional considerations such as manufacturing simplicity, or cost. A comparative analysis was conducted between generative design and traditional design, which relies on human expertise, CAD (computer-aided design) tools, and algorithms. The weight, deformation, stress, fatigue, CO2 emissions and self-sufficiency performance of the two methods were evaluated. It was found that generative design allows for a 30% reduction in the weight of the part compared to traditional design, while simultaneously improving its rigidity and strength. The total weight reduction of many brackets for the same vehicle also results in a decrease in CO2 by 3.5 g/km and an increase in autonomy by 2.8 km.

Keyphrases: Computer Aided Design, Engine mounting bracket, FEA, Lightweighting, Sustainability, fuel consumption, retro-engineering, topology optimization

@booklet{EasyChair:15478,
  author    = {Hassan Rouane and Souad Tayane and Mohamed Ennaji and Jaafar Gaber},
  title     = {Generative Design Case Study for Eco-Efficient Vehicles Lower Emissions and Greater Range},
  howpublished = {EasyChair Preprint 15478},
  year      = {EasyChair, 2024}}