Reliability optimization method for CFRP control arm with multi-scale parameters

SU Hailiang; AN Deng; QIN Qiulin; HAN Zhe; WEI Tengteng

doi:10.16579/j.issn.1001.9669.2026.08.007

Abstract：[Objective]Multi-scale parameter uncertainties are involved in the design of carbon fiber reinforced polymer(CFRP)control arms.Macro-scale layup optimization is mostly adopted in existing researches,which can hardly balance material micro-property fluctuation and macro structural reliability.To fully exploit the mechanical properties of CFRP,a reliability design scheme with synergistic macro and micro parameters was proposed for CFRP control arms.[Methods]Firstly,the mechanical properties of unidirectional CFRP were predicted by Digimat software,and the prediction accuracy was verified through tensile tests to provide data support for cross-scale parameter transfer.Secondly,structural replacement of the CFRP control arm was completed based on the equal stiffness principle,and a finite element analysis model was constructed to provide a calculation carrier for reliability evaluation.Then,a double-weighted vector-projection response surface method was proposed,and reliability analysis of the control arm under multi-scale uncertain parameters was realized combined with a polynomial surrogate model to balance calculation accuracy and efficiency.Finally,a multi-scale reliability optimization framework was established and solved by integrating the surrogate model,improved response surface method and particle swarm optimization algorithm.[Results]The results show that the reliability calculation error of the double-weighted vector-projection response surface method is less than 0.9％compared with the Monte Carlo method.The optimized CFRP control arm achieves a weight reduction of 41.29％compared with the original steel structure under the constraints of 90％reliability and performance requirements,which can provide references for lightweight reliability design of composite automotive structural parts.