.png){kind=icon}
.png){kind=icon}
Return
Distillation process optimization: A screening-clustering assisted kriging optimization method
.png){kind=icon}
PRE
AI
DOI：10.1016/j.ces.2021.116597.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Economic optimization is an important engineering aspect of modern distillation. A screening-clustering assisted kriging optimization (SCAKO) method is proposed herein to optimize the economics of the distillation process. The SCAKO consists of a kriging surrogate model, an expected improvement sampling approach, a screening-clustering operation, and a quantum-behaved particle swarm optimization algorithm. The main feature of the SCAKO method is the combination of an effective search domain contraction approach and the kriging surrogate model. The insignificant sampled points are deleted from the dataset, and the remaining sampled points are divided into a series of clusters. The search domain is then divided into several sub-domains according to the information of the points in the clusters. Kriging surrogate model is constructed to represent the variation trend of the optimization objective in each sub domain. Case studies were performed to validate the computational effectiveness and efficiency of the SCAKO. (c) 2021 Elsevier Ltd. All rights reserved.
Keywords:
Optimization of distillation process
Kriging surrogate model
Contraction of search domain
Expected improvement sampling approach
Screening-clustering Operation
Quantum-behaved particle swarm& nbsp
optimization