Robust Bayesian optimization for flexibility analysis of expensive simulation-based models with rigorous uncertainty bounds
Robust Bayesian optimization for flexibility analysis of expensive simulation-based models with rigorous uncertainty bounds
复制标题
DOI：
10.1016/j.compchemeng.2023.108515
复制
发表时间：
2023-11
期刊：
影响因子：
--
通讯作者：
Akshay Kudva;Wei-Ting Tang;J. Paulson
中科院分区：
文献类型：
--
作者：
Akshay Kudva;Wei-Ting Tang;J. Paulson
新兴生化系统的性能取决于其快速准确地适应不确定性的潜力。柔性分析是一个定量的框架，用于确定系统是否能够在不确定性的情况下保持安全可行的运行。大多数可用的方法假设访问面向方程的模型，这在实践中可能很难获得。在本文中，我们提出了一个连续的黑盒灵活性分析方法，BoFlex，克服了这一挑战，通过构建概率代理模型的不确定性和追索权变量的联合空间。BoFlex是基于一个特殊的交替的置信区间的过程，我们表明，在温和的假设下，未知函数收敛到一个正确的解决方案。我们还建立了严格的上界的代理模型的最大信息增益的收敛速度。BoFlex的优势在几个案例研究中得到了证明，包括热交换器网络和鼓泡塔反应器。
The performance of emerging biochemical systems rests on their potential to adapt to uncertainties quickly and accurately. Flexibility analysis is a quantitative framework for determining if a system can maintain safe and feasible operation despite uncertainty. Most available methods assume access to equation-oriented models, which can be difficult to obtain in practice. In this paper, we propose a sequential black-box flexibility analysis method, BoFlex, that overcomes this challenge by constructing probabilistic surrogate models over the joint space of uncertain and recourse variables. BoFlex is based on a special alternating confidence bound procedure, which we show finitely converges to a correct solution under mild assumptions on the unknown functions. We also establish a rigorous upper bound on the convergence rate in terms of the maximum information gain of the surrogate model. The advantages of BoFlex are demonstrated on several case studies including a heat exchanger network and a bubble column reactor.