Metamodel-based Hybrid Parametric Optimization with Static Prototype Validation for a Six-Axle Wheel Load Cell Design

Abstract

The accurate measurement of forces and torques acting on a vehicle’s chassis, originating from wheel-pavement interactions, is essential for optimizing suspension systems, axles, and other structural components. A critical challenge in optimizing wheel load cell (WLC) design is to enhance sensitivity without compromising mechanical robustness. To address this question, this study developed a six-axis load cell designed to decouple force and torque measurements, achieving high sensitivity and reliability. The objective was to design an optimized WLC for multi-axis force measurement by maximizing sensitivity through a metamodel-based hybrid optimization framework. The methodology began with a fractional factorial Design of Experiments (DOE) to identify key parameters affecting load cell sensitivity and to define the optimization search space. This was followed by a dual-step parametric optimization process, utilizing the finite element method (FEM) to evaluate load cell performance and an inner-point optimization algorithm for convergence. After completing the DOE and FEM stages, the load cell was physically constructed and calibrated using a universal testing machine MTS 810, where experimental data closely matched simulation results, validating the design’s effectiveness. The main findings indicate that the optimized load cell can reliably measure six components, with a sensitivity increase of 14% compared to conventional designs. Numerical results from FEM analyses showed stress levels at 215 MPa, displacements around 0.16 mm, and a first-mode vibration frequency of 1351 Hz, meeting all structural integrity and performance requirements. Calibration confirmed minimal cross-talk effects, supporting the robustness of the final design. The developed load cell met optimal design criteria, showing substantial improvements in sensitivity and accuracy over existing designs, with potential applications in automotive, aerospace, and engineering testing.