Constrained Hybrid Neural Modelling of Biotechnological Processes

Karama, Asma; Bernard, Olivier; Gouzé, Jean‐Luc

doi:10.2202/1542-6580.2117

Cited by 4 publications

(7 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First principles constrained loss function (FPCLF) [86], [48], [13], [47], [91] First principles constrained model design (FPCMD) [50], [92], [13], [93], [94], [95], [28], [96], [97] First principles constrained output models (FPCOMs) [57], [58] [98], [99], [57] First principles initialized models (FPIMs) [90] [ 49] [ 46], [100], [17], [101], [63], [90], [102], [103], [24] uncertain FPs model are further split into categories due to the nature of the uncertainty. The uncertainty enters in the form of parametric uncertainty in lumped parameter models while in residual models it enters as structural uncertainty of FP model.…”

Section: Cstr Modelingmentioning

confidence: 99%

Integration of machine learning and first principles models

et al. 2022

View full text Add to dashboard Cite

Model building and parameter estimation are traditional concepts widely used in chemical, biological, metallurgical, and manufacturing industries. Early modeling methodologies focused on mathematically capturing the process knowledge and domain expertise of the modeler. The models thus developed are termed first principles models (or white‐box models). Over time, computational power became cheaper, and massive amounts of data became available for modeling. This led to the development of cutting edge machine learning models (black‐box models) and artificial intelligence (AI) techniques. Hybrid models (gray‐box models) are a combination of first principles and machine learning models. The development of hybrid models has captured the attention of researchers as this combines the best of both modeling paradigms. Recent attention to this field stems from the interest in explainable AI (XAI), a critical requirement as AI systems become more pervasive. This work aims at identifying and categorizing various hybrid models available in the literature that integrate machine‐learning models with different forms of domain knowledge. Benefits such as enhanced predictive power, extrapolation capabilities, and other advantages of combining the two approaches are summarized. The goal of this article is to consolidate the published corpus in the area of hybrid modeling and develop a comprehensive framework to understand the various techniques presented. This framework can further be used as the foundation to explore rational associations between several models.

show abstract

Section: Cstr Modelingmentioning

confidence: 99%

Integration of machine learning and first principles models

et al. 2022

View full text Add to dashboard Cite

show abstract

“…More forming knowledge can also be incorporated using "standard" formulations of the kinetic rates, representing the contained kinetic parameters by nonparametric model expressions (AlYemni, 2003;Bellos, Kallinikos, Gounaris, & Papayannakos, 2005;Kasprow, 2000;Mazutti et al, 2010). In addition, forming knowledge can also lead to Bounded Input Bounded Output (BIBO) stability properties of the model (Karama, Bernard, & Gouz, 2010;Oliveira, 2004), e.g. a reaction can only occur when all reactants are present.…”

Section: Mechanistic Knowledge Incorporationmentioning

confidence: 99%

Hybrid semi-parametric modeling in process systems engineering: Past, present and future

Stosch¹,

Oliveira²,

Peres³

et al. 2014

Computers & Chemical Engineering

358

201

View full text Add to dashboard Cite

Hybrid semi-parametric models consist of model structures that combine parametric and nonparametric submodels based on different knowledge sources. The development of a hybrid semi-parametric model can offer several advantages over traditional mechanistic or data-driven modeling, as reviewed in this paper. These advantages, such as broader knowledge base, transparency of the modeling approach and cost-effective model development, have been widely recognized, not only in academia but also in the industry.In this paper, the most common hybrid semi-parametric modeling and parameter identification techniques are revisited. Applications in the areas of (bio)chemical engineering for process monitoring, control, optimization, scaleup and model-reduction are reviewed. It is outlined that the application of hybrid semi-parametric techniques does not automatically lead into better results but that rational knowledge integration has potential to significantly improve model-based process operation and design.

show abstract

“…In this work, we consider a problem of trajectory tracking control of a fermentation process and we are particularly interested in taking into account the constraints in the control. The constraints are very important in practical situations to keep the physical and biological meaning, such as the positivity of the concentrations and boundness of variables . The introduction of the constraints in the case of T‐S was developed in Reference by considering the norm‐bounded approach of the input and leading to an additional linear matrix inequality (LMI) to be verified.…”

Section: Introductionmentioning

confidence: 99%

“…The constraints are very important in practical situations to keep the physical and biological meaning, such as the positivity of the concentrations and boundness of variables. 12,13 The introduction of the constraints in the case of T-S was developed in Reference 14 by considering the norm-bounded approach of the input and leading to an additional linear matrix inequality (LMI) to be verified. This LMI depends on the value or on the upper bound of the initial states of the system.…”

Section: Introductionmentioning

confidence: 99%

Takagi‐Sugeno tracking control of a fermentation process with respect to asymmetric constraints

Mohamed

Karama

Khallouq

2019

Adaptive Control & Signal

Self Cite

View full text Add to dashboard Cite

Summary This article deals with the tracking problem for nonlinear biological process with asymmetric constraints on the controller. Using the fuzzy logic Takagi‐Sugeno (TS), the nonlinear process is modeled by a fuzzy model and a parallel distributed compensation (PDC) controller with respect asymmetric constraints is designed. The main contribution of this work is to extend the available results, in the context of a TS formulation, for symmetrical constraints to asymmetrical ones. Hence, sufficient conditions are obtained for the first time, in terms of linear matrix inequalities (LMIs), for the existence of an asymmetrical constrained PDC controller. Moreover, fuzzy observer based controller with unmeasured premise variables is designed. The performance and effectiveness of the method are illustrated in simulation by considering a fermentation process.

show abstract

Constrained Hybrid Neural Modelling of Biotechnological Processes

Cited by 4 publications

References 19 publications

Integration of machine learning and first principles models

Integration of machine learning and first principles models

Hybrid semi-parametric modeling in process systems engineering: Past, present and future

Takagi‐Sugeno tracking control of a fermentation process with respect to asymmetric constraints

Contact Info

Product

Resources

About