Efficient trust region filter modeling strategies for computationally expensive black-box optimization

Runzhe Liang, Yueheng Han, Haoyang Hu, Bingzhen Chen, Zhihong Yuan, Lorenz T. Biegler

Abstract: Highlights•Refine a new framework to elevate optimization efficiency for chemical systems involving computationally expensive black-box functions.•Improve sampling efficiency is improved by incorporating the known data beyond the trust region to assist in constructing reduced models.•Gaussian process as the reduced model can lower the number of calls to black-box functions by more than half compared to common linear and quadratic models.