Surrogate Modeling for Accelerating Optimization of Complex Systems in Chemical Engineering

Given the capability to lower computational load when addressing large-scale, nonlinear, and strongly coupled problems, surrogate modeling has recently gained increased popularity within the realm of chemical engineering. By approximating the complex parts of the system to be optimized, the need for expensive function evaluations can be substantially reduced. In recent decades, a variety of surrogate models, encompassing regression models and interpolation models, have been proposed and deployed in the field of chemical engineering. With the emergence of artificial intelligence (AI), machine learning-based surrogate models have gained traction in research. The demonstrated potential of surrogate modeling reveals its pivotal role in the application of chemical processes. This chapter aims to offer a comprehensive perspective by summarizing widely used surrogate modeling techniques and their applications in the optimization of complex systems in chemical engineering. Drawing on knowledge in the field, this chapter elucidates the practical significance and advancements of surrogate modeling for accelerating the optimization of complex chemical systems.