Inverse identification framework for cohesive zone model incorporating failure mode based on multi-island genetic algorithm
Composite interfaces are commonly simulated by cohesive zone models with the key challenge being the calibration of interfacial parameters. A new framework is presented in this paper to derive the characteristic of any cohesive zone model. This approach employs the multi-island genetic algorithm to obtain the interface parameters aligning closely with the experimental observations. The introduced framework innovatively formulates an objective function, considering both the congruence of the load-displacement curve and the alignment with the failure mode of model. The framework combines machine learning and multi-objective optimization. A method using the interface debonding length to quantify the failure mode of the model is proposed. To demonstrate the feasibility of the proposed framework, the newly strength-based cohesive zone model is taken as an example, and key parameters and damage evolution are identified accurately. The inverse algorithm is used to identify the interface parameters of both the double cantilever beam experiment and the four-point bending test. The robustness and accuracy of the framework are validated through the double cantilever beam test. The findings indicate that the numerical results align closely with the experimental data, confirming that the interface parameters identified by the proposed framework can reproduce the performance of the adhesive joints.