Vehicle suspension recommendation system: multi-fidelity neural network-based mechanism design optimization

Abstract Mechanical mechanisms are designed to perform specific functions in a variety of fields. In most cases, there is not a unique mechanism that performs a well-defined function. For example, vehicle suspensions are designed to improve driving performance and ride comfort, but different types are available depending on the environment in which they are used. This variability in design due to different usage environments makes performance comparison difficult. In addition, the industry’s traditional design process is multi-step, gradually reducing the number of design candidates while performing costly analysis to achieve target performances. Recently, artificial intelligence models have been used to replace the computational cost of finite element analysis (FEA). However, there are limitations in data availability and different analysis environments, especially when moving from low-fidelity to high-fidelity analysis. In this paper, we propose a multi-fidelity design framework aimed at recommending optimal types and designs of mechanical mechanisms. As an application, vehicle suspension systems were selected, and several types were defined. For each type, mechanism parameters were generated and converted into 3D CAD models, followed by low-fidelity rigid body dynamic analysis under driving conditions. To effectively build a deep learning-based multi-fidelity surrogate model, the results of the low-fidelity analysis were analyzed using Density-Based Spatial Clustering of Applications with Noise (DBSCAN) and sampled at 5% for the high-cost flexible body dynamic analysis. After training a multi-fidelity model, a multi-objective optimization problem was formulated for the performance metrics of each suspension type. Finally, we recommend the optimal type and design based on the input (sprung mass) to optimize the ride comfort-related performance metrics. Subsequently, to validate the proposed methodology, we extracted basic design rules for Pareto solutions using data mining techniques. We also verified the effectiveness and applicability by comparing the results with those obtained from a conventional deep learning-based design process.