Enhanced damage-coupled viscoplastic constitutive modeling with advanced meta-heuristic algorithm-based automated parameter inversion

Accurate prediction of high-temperature component behavior under low-cycle fatigue (LCF) and creep-fatigue interaction (CFI) conditions requires an advanced constitutive model and a robust parameter identification approach. This study introduces a damage-coupled unified viscoplastic constitutive model (UVCM) incorporating cyclic softening, transient Bauschinger effects, and strain range dependency to characterize life-cycle deformation behavior. To overcome limitations in traditional parameter calibration, a metaheuristic black-winged kite algorithm (MBKA) is developed by combining hybrid initialization strategies, swarm diversity-driven adaptation, and golden sine search. Simulation results indicate that MBKA exhibits superior convergence accuracy over tested renowned algorithms in solving 23 CEC2005 functions, practically in multimodal optimization scenarios. When applied to 2.25CrMoV steel at 455 °C, the UVCM-MBKA framework successfully replicates the observed experimental phenomena, including strain amplitude-dependent cyclic deformation, transient Bauschinger effects, dwell time-induced decelerated stress relaxation, and directional sensitivity of CFI. Furthermore, the model demonstrates accurate fatigue life prediction across 16 experimental cases, with numerical results closely matching observed continuous cyclic softening and three-stage decelerated relaxation. Validation confirms the framework's precision and robustness in predicting cyclic deformations and fatigue life.