Combining first principles with black-box techniques for reaction systems

Chen, Libei; Hontoir, Y.; Huang, Dexian; Zhang, Jie; Morris, A.J.

doi:10.1016/j.conengprac.2003.09.006

Cited by 32 publications

(8 citation statements)

References 10 publications

(10 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Process integration, e.g., reactive distillation, imposes additional complexity for the modeling of distillation units. Different types of ANNs have been used to replace expensive calculations in reactive distillation models , , . Furthermore, in practice, composition measurement devices can be very expensive to purchase and maintain, and they usually have a time‐delay response that prevents proper control of the distillation column.…”

Section: Overview Of the Use Of Hybrid Models In Separation Processesmentioning

confidence: 99%

Hybrid Semi‐parametric Modeling in Separation Processes: A Review

McBride

Medina

Sundmacher

2020

Chemie Ingenieur Technik

View full text Add to dashboard Cite

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. Separations of mixtures play a critical role in chemical industries. Over the last century, the knowledge in the area of chemical thermodynamics and modeling of separation processes has been substantially expanded. Since the models are still not completely accurate, hybrid models can be used as an alternative that allows to retain existing knowledge and augment it using data. This paper explores some of the weaknesses in the current knowledge in separations design, simulation, optimization, and operation, and presents many examples where data-driven and hybrid models have been used to facilitate these tasks.

show abstract

Section: Overview Of the Use Of Hybrid Models In Separation Processesmentioning

confidence: 99%

Hybrid Semi‐parametric Modeling in Separation Processes: A Review

McBride

Medina

Sundmacher

2020

Chemie Ingenieur Technik

View full text Add to dashboard Cite

show abstract

“…(1) adaptive systems attempts to estimate the influent disturbances [26], the parameters [7,12], the state variables [9,10] and/or kinetics [11,14,15] on-line. With the information estimated, a PID like, or more sophisticated control strategies can be readily applied; (2) robust and optimal control [13,28], on the other hand, is based only on simplified linear models used to develop a feedback law to treat nonlinearities, parameter variations and load disturbances, while (3) feedback linearization or similar [4,24,29] is employed to compensate the known nonlinear kinetics through certain implied information contained in (on-line) measurable parameters such as the off gases; and (4) intelligent control system [17,18,30] utilises the experiences gained from the reactor operation.…”

Section: Introductionmentioning

confidence: 99%

Bifurcation and stability analysis of an anaerobic digestion model

et al. 2006

View full text Add to dashboard Cite

This paper presents the dynamic behaviour of the anaerobic digestion process, based on a simplified model. The hydraulic, biological and physicochemical processes such as those which underpin anaerobic digestion have more than one stable stationary solution and they compete with each other. Further, the attractive domains of the stable solutions vary with the key parameters. Thus, some initial transient process moving toward one stable solution could suddenly move towards another solution, at which a so-call catastrophe takes places (e.g. washout). The paper systematically analyses the stationary solutions with their associated stability, which provides insight and guidance for anaerobic digestion reactor design, operation and control.

show abstract

“…Compared with the parallel model, it makes full use of the a priori process knowledge and has better interpretability. Secondly, the relationships between the kinetic reaction rates and their influencing factors are easily discovered by the data‐driven models without specific model structures considered . Thirdly, in the serial hybrid model, only parts of the whole process model require training, which eases the difficulty of training models and significantly reduces the possibility of getting stuck in a local minimum .…”

Section: Serial Hybrid Modelmentioning

confidence: 99%

Serial Hybrid Modelling for a Gold Cyanidation Leaching Plant

et al. 2015

View full text Add to dashboard Cite

In this paper, a serial hybrid model of a gold cyanidation leaching process is proposed. The serial hybrid model consists of mass conservation equations of gold and cyanide as well as two kernel partial least square (KPLS) models, which are used as the estimators of the unknown kinetic reaction rates without the complicated kinetic model structures considered. The proposed serial hybrid model makes full use of both the a priori process knowledge and the ability of a data-driven model to discover the information behind data sets. Moreover, before training the KPLS models, the proposed estimation strategy based on Tikhonov regularization is used to estimate the kinetic reaction rates, which can mitigate the effect of measurement noise on the estimation results effectively. The proposed serial hybrid model has been applied to a gold cyanidation leaching plant to predict the gold leaching rate. The prediction results show that the proposed serial hybrid model can track the real leaching rate of gold closely and has the best prediction accuracy at both dynamic and steady states compared with the pure KPLS and mechanistic models, thereby laying an important foundation for the successful implementation of optimization and control of the leaching process.Keywords: leaching process, mechanistic model, hybrid model, kinetic rate, Tikhonov regularization INTRODUCTION C yanidation has been the main method to recover gold from ores, concentrates, and tailings for the past century. It has the advantages of being a mature technology, having high gold recovery, having a low production cost, and being applicable to a diverse variety of ore types. [1,2] Usually, a gold cyanidation leaching process occurs in an aerated continuous stirred tank reactor (CSTR) containing a solution of sodium cyanide and oxygen. The undissolved gold in the solid is transformed into the easy-to-dissolve gold cyanide complex by means of reaction with both sodium cyanide and oxygen. [2,3] In practice, a cascade of leaching tanks in series are configured in order to increase the residence time of slurry and hence ensure a longer reaction time. [4] As the most important procedure in hydrometallurgy, the performance of the leaching process has a significant impact on the subsequent procedures and ultimately determines the gold recovery. The most important quality index for the leaching process is the leaching rate, which is directly related to gold recovery, production efficiency, and overall yield. The ideal leaching rate of 100 % is required in all plants, however this is not realistic in practice due to the impossibilities of infinitely small ore particles, infinitely long residence time, and complete reaction. [5] In industrial applications, the lowest production cost is usually required under the condition that the leaching rate is ensured to be equal to or greater than the index determined by the plant-wide decision. The above optimal operation can usually be obtained by solving the model-based optimization problem of the leaching process -namely, dete...

show abstract

Combining first principles with black-box techniques for reaction systems

Cited by 32 publications

References 10 publications

Hybrid Semi‐parametric Modeling in Separation Processes: A Review

Hybrid Semi‐parametric Modeling in Separation Processes: A Review

Bifurcation and stability analysis of an anaerobic digestion model

Serial Hybrid Modelling for a Gold Cyanidation Leaching Plant

Contact Info

Product

Resources

About