Iterative Learning Control for Final Batch Product Quality Using Partial Least Squares Models

Flores‐Cerrillo, Jesus; MacGregor, John F.

doi:10.1021/ie048811p

Cited by 66 publications

(72 citation statements)

References 41 publications

(59 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Latent variable methods have proven successful for databased MPC of batch processes, 14,[25][26][27]29,49 A key issue in these contributions is the arrangement of batch data into matrices for latent variable analysis. In many of these contributions, a latent variable model is built by unfolding batch data in the batch-wise manner described in Eqs.…”

Section: Latent Variable Control Analysismentioning

confidence: 99%

See 1 more Smart Citation

Subspace identification for data‐driven modeling and quality control of batch processes

2016

View full text Add to dashboard Cite

In this work, we present a novel, data-driven, quality modeling, and control approach for batch processes. Specifically, we adapt subspace identification methods for use with batch data to identify a state-space model from available process measurements and input moves. We demonstrate that the resulting linear time-invariant (LTI), dynamic, state-space model is able to describe the transient behavior of finite duration batch processes. Next, we relate the terminal quality to the terminal value of the identified states. Finally, we apply the resulting model in a shrinking-horizon, model predictive control scheme to directly control terminal product quality. The theoretical properties of the proposed approach are studied and compared to state-of-the-art latent variable control approaches. The efficacy of the proposed approach is demonstrated through a simulation study of a batch polymethyl methacrylate polymerization reactor. Results for both disturbance rejection and set-point changes (i.e., new quality grades) are demonstrated.

show abstract

Section: Latent Variable Control Analysismentioning

confidence: 99%

“…Between batches, the input trajectory is updated to reject disturbances observed in the preceding batches. A significant number of literature contributions have been made addressing this approach [13][14][15][16][17] , Refs. 18 (This is available in print at the library in storage), 19, and 20.…”

Section: Introductionmentioning

confidence: 99%

Subspace identification for data‐driven modeling and quality control of batch processes

2016

View full text Add to dashboard Cite

show abstract

“…Chin et al [5] developed an improved control framework in which the real-time feedback control and ILC are combined for the repetitive processes. On the basis of the empirical partial least squares models, a terminal ILC approach was proposed for production of nylon by Flores-Cerrillo et al [6]. In Ref.…”

Section: Introductionmentioning

confidence: 99%

Improved state space model predictive fault-tolerant control for injection molding batch processes with partial actuator faults using GA optimization

Zou

Zhang

2018

ISA Transactions

View full text Add to dashboard Cite

“…They attracted considerable interest in profile tracking control for batch processes. Chen and Wang [6] proposed an iterative learning control strategy based on the PLS method for single batch unit only [6]; Flores-Cerrillo and MacGregor [7] using MPLS developed a terminal iterative learning control strategy for batch-to-batch and within-batch control of final product properties [7]. PLS/MPLS takes advantage of the precision improved by…”

Section: Introductionmentioning

confidence: 99%

Development of MBPLS based control for serial operation processes

Chen

Tsai

2009

Korean J. Chem. Eng.

View full text Add to dashboard Cite

A control scheme based on the multiblock PLS (MBPLS) model for multi-stage processes (or serially connected processes) is developed. MBPLS arranges a large number of variables into meaningful blocks for each stage of the large-scale system. Two control design strategies, course-to-course (CtC) and within-stage (WS) controls, are proposed for the re-optimization design in the whole multistage course. In CtC, MBPLS control and optimization are done by applying feedback from the finished output quality when one course for all stages is done. It utilizes the information from the current course to improve quality of the next one. In WS, the MBPLS-based re-optimization strategy is developed to explore the possible adjustments of the future inputs at the rest of the stages in order to fix up the disturbances just in time and to maintain the product specification when the current course is finished. The proposed technique is successfully applied to two simulated industrial problems, including a photolithography sequences and a reverse osmosis desalination process, and the advantages of the proposed method are demonstrated.

show abstract

Iterative Learning Control for Final Batch Product Quality Using Partial Least Squares Models

Cited by 66 publications

References 41 publications

Subspace identification for data‐driven modeling and quality control of batch processes

Subspace identification for data‐driven modeling and quality control of batch processes

Improved state space model predictive fault-tolerant control for injection molding batch processes with partial actuator faults using GA optimization

Development of MBPLS based control for serial operation processes

Contact Info

Product

Resources

About