Multi-Objective Optimization of Curing Profiles for CFRP Patch Repair Under Thermochemical Coupling

Abstract

This study addresses the challenge of temperature non-uniformity during carbon fibre-reinforced polymer (CFRP) composite patch repair, which compromises curing quality and process efficiency. A coupled heat transfer–curing kinetics finite element model was developed and experimentally validated to investigate the heat sink effect of support structures. Key findings reveal that temperature differences concentrate near aluminum components and increase with curing temperature. For the present scarf-repair configuration, global sensitivity analysis identified the second-stage holding temperature (T2) and heating rate (r2) as the dominant factors governing temperature uniformity, whereas the holding times (dt1 and dt2) primarily determine the total curing time (ttotal). A novel multi-objective optimization framework combining optimal Latin hypercube sampling, radial basis functions, and NSGA-II was established. The optimized curing profile achieves a surrogate-predicted reduction of 22.5% in maximum temperature difference (22.0% when confirmed by high-fidelity finite element verification) and 36% in total curing time, while maintaining a minimum degree of cure above 0.98. These results provide a validated, surrogate-based framework for designing curing protocols that resolve metal-induced thermal non-uniformity in composite repairs without sacrificing cure quality.