Multi-rate state estimation applied to a pilot-scale reactive distillation process

Haßkerl, Daniel; Subramanian, Sankaranarayanan; Markert, Steven; Kaiser, Stefanie; Engell, Sebastian

doi:10.1016/j.ces.2018.04.018

Cited by 15 publications

(4 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Automatic differentiation also is used to speed up the state estimation algorithm. In our study, we use the fixed-structure multirate extended Kalman filter that was adapted to differential-algebraic equations; see refs , , and . The EKF reconstructs the state of the column from the available measurements using the same RD model that is used in the optimizing controller.…”

Section: Problem Description and Realization Of The Controllermentioning

confidence: 99%

See 1 more Smart Citation

Dynamic Performance Optimization of a Pilot-Scale Reactive Distillation Process by Economics Optimizing Control

Haßkerl

Lindscheid

Subramanian

et al. 2018

Ind. Eng. Chem. Res.

Self Cite

View full text Add to dashboard Cite

The most prominent example of process intensification by process integration is reactive distillation (RD), which can realize a high reactant conversion and a high purity of the target product in equilibrium-limited reaction systems. Reactive distillation is mostly used for single reactions, but realizations for multiple-reaction systems leading to more than one target product can also be found. In multipurpose chemical processes, the target product of a multiple-reaction system can be switched during the production. When combining fluid separation and multiple-reaction systems, e.g., in a RD column, it is challenging to find operating conditions that lead to the most profitable operation of the process. Economics optimizing control where an economic objective is employed in a model-predictive control framework has the potential to drive the process to the economic optimum during operation based upon a process model and online measurements. However, so far only few applications of such advanced process control (APC) concepts to reactive distillation processes have been realized and these do not concern multiple-reaction systems. In this contribution, we describe the application of economics optimizing control to a multiproduct RD process. The selected case study is a homogeneously catalyzed two-step transesterification reaction that is performed in a RD column in pilotplant scale.

show abstract

Section: Problem Description and Realization Of The Controllermentioning

confidence: 99%

“…Due to using near-infrared spectroscopy, two sampling frequencies have to be considered and the fixed-structure multirate algorithm was found to be the best choice. Furthermore, it was proven to work well in practice; see ref .…”

Section: Problem Description and Realization Of The Controllermentioning

confidence: 99%

Dynamic Performance Optimization of a Pilot-Scale Reactive Distillation Process by Economics Optimizing Control

Haßkerl

Lindscheid

Subramanian

et al. 2018

Ind. Eng. Chem. Res.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Regarding (bio-)chemical processes, a recurring issue lies in the availability and frequency of quality measurements. Hence, dedicated research investigated the effect of low sampling rates of important quality measurements and how these measurements can be incorporated into state estimators [6,7,8]. By now, this has also been introduced into state estimators applicable to highly nonlinear systems [9] and been amended by filtering of gross errors [10].…”

Section: Introductionmentioning

confidence: 99%

Semi-supervised Learning for Data-driven Soft-sensing of Biological and Chemical Processes

Esche¹,

Talis²,

Weigert³

et al. 2021

Preprint

View full text Add to dashboard Cite

Continuously operated (bio-)chemical processes increasingly suffer from external disturbances, such as feed fluctuations or changes in market conditions. Product quality often hinges on control of rarely measured concentrations, which are expensive to measure. Semi-supervised regression is a possible building block and method from machine learning to construct soft-sensors for such infrequently measured states. Using two case studies, i.e., the Williams-Otto process and a bioethanol production process, semi-supervised regression is compared against standard regression to evaluate its merits and its possible scope of application for process control in the (bio-)chemical industry.

show abstract

“…Hasskerl et al 19 used the PF state estimation method for a reactive distillation process and demonstrated that the PF was applicable for this complicated system. Haßkerl et al 20 verified that the multirate state estimation adopting an EKF method was effective. Ritschel et al 21 compared the accuracy and efficiency of the EKF, UKF, and PF to estimate the states of the UV flash process.…”

Section: Introductionmentioning

confidence: 99%

Comparison of Composition Soft Sensor Correction Methods for a Kaibel Dividing-Wall Column

Qian

Jia²,

Huang

et al. 2021

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Though a Kaibel dividing-wall column (DWC) is able to save approximately 40% operation fees, the fear of controllability issues of the Kaibel DWC obstructs its large-scale industrialization. In our previous study, a composition/temperature cascade MPC-PI scheme was proposed with proportional–integral (PI) controllers stabilizing the operations and model predictive control (MPC) improving the performance. In the current paper, the MPC-PI scheme in combination with the data-driven soft sensor model replaces detecting compositions by detecting auxiliary variables, which is more applicable and practical. Although the same soft sensor model is applied, different Kalman filters used for error corrections will result in different composition estimation performances, which will consequently affect control performances for the Kaibel DWC. Soft sensors with different nonlinear Kalman filters have been studied and compared for the control scheme, containing the extended Kalman filter (EKF), unscented Kalman filter (UKF), and particle filter (PF). Research shows that the EKF can reach the setpoints of the product compositions faster than the UKF, but there are oscillations with reciprocating motion using the EKF. The speed of the PF to achieve stability is the fastest; however, the steady-state offsets remain. Therefore, the UKF is the most appropriate filter in composition soft sensors for the Kaibel DWCs, as its corresponding control performances are stable without the oscillations using the EKF and the steady-state offsets employing the PF.

show abstract

Multi-rate state estimation applied to a pilot-scale reactive distillation process

Cited by 15 publications

References 47 publications

Dynamic Performance Optimization of a Pilot-Scale Reactive Distillation Process by Economics Optimizing Control

Dynamic Performance Optimization of a Pilot-Scale Reactive Distillation Process by Economics Optimizing Control

Semi-supervised Learning for Data-driven Soft-sensing of Biological and Chemical Processes

Comparison of Composition Soft Sensor Correction Methods for a Kaibel Dividing-Wall Column

Contact Info

Product

Resources

About