Nonlinear process modeling and optimization based on Multiway Kernel Partial Least Squares model

Abstract

MKPLS (Multiway Kernel Partial Least Squares) methods are used to model the batch processes from process operational data. To improve the optimization performance, a batch-to-batch optimization strategy is proposed based on the idea of the similarity between the iterations during numerical optimization and successive batch runs. SQP (Sequential Quadratic Programming) coupling with MKPLS model is used to solve the optimization problem, and the plant data, instead of the MKPLS model predictions, are used in gradient calculation. The proposed strategy is illustrated on a simulated bulk polymerization of styrene. The results demonstrate that the optimization performance has been improved in spite of the model-plant mismatches.

Combining Kernel Partial Least-Squares Modeling and Iterative Learning Control for the Batch-to-Batch Optimization of Constrained Nonlinear Processes

“…The five layer neural network method was used to solve the nonlinear issue by Krammer . The principle component curve-neural network algorithm was proposed by Dong et al The basic idea of the kernel method is to first map the input space into a feature space via a nonlinear map and then extract the dominant components in the feature space. − , Sequential quadratic programming coupling with multiway kernel partial least-squares (MKPLS) model is used to solve the optimization problem without constraints . The kernel partial least-squares regression issue is investigated in reproducing the kernel Hilbert space .…”

“…[20][21][22][23]40 Sequential quadratic programming coupling with multiway kernel partial leastsquares (MKPLS) model is used to solve the optimization problem without constraints. 38 The kernel partial least-squares regression issue is investigated in reproducing the kernel Hilbert space. 39 With the simple computing, the kernel methods are available for the optimal problems.…”