Bayesian optimization for inverse calibration of expensive computer models: A case study for Johnson-Cook model in machining

Inverse model calibration for identifying the constitutive parameters can be computationally demanding expensive-to-evaluate simulation models. This paper presents a modified Bayesian optimization (BO) method, denoted as BO-bound, that incorporates theoretical bounds on quantity of interest. A case study inverse Johnson Cook (J-C) flow stress is presented using machining (cutting) force data. The results show fast five J-C within 25 simulations. In general, BO-bound method applicable any expensive models well problems with known bounds.