A novel identification method for hybrid (N)PLS dynamical systems with application to bioprocesses

Stosch, Moritz von; Oliveira, Rui; Peres, J.; Azevedo, S. Feyo de

doi:10.1016/j.eswa.2011.02.117

Cited by 37 publications

(25 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…early-stopping is used to avoid over-fitting (Simutis and Luebbert, 1997). In order to escape from local minima during the identification, several random weight initializations (10 in the cases presented below) are performed and the best parameters are chosen using cross-validation (Simutis and Luebbert, 1997, Oliveira, 2004, von Stosch et al, 2011.…”

Section: Training-parameter Identificationmentioning

confidence: 99%

See 1 more Smart Citation

An efficient method for the numerical integration of measured variable dependent ordinary differential equations

Azevedo

Peres

Stosch

2015

Engineering Applications of Artificial Intelligence

Self Cite

View full text Add to dashboard Cite

Section: Training-parameter Identificationmentioning

confidence: 99%

“…the lsqnonlin function of the Matlab Optimization toolbox was used. The gradients are derived using the sensitivities approach (Oliveira, 2004;von Stosch et al, 2011)…”

Section: Training-parameter Identificationmentioning

confidence: 99%

An efficient method for the numerical integration of measured variable dependent ordinary differential equations

Azevedo

Peres

Stosch

2015

Engineering Applications of Artificial Intelligence

Self Cite

View full text Add to dashboard Cite

“…Another emerging area for the application of hybrid semi-parametric models is systems biology see for instances (Carinhas et al, 2011;von Stosch et al, 2010). Hybrid semi-parametric modeling is attractive in this area since it can help to link the different scales of cell modeling and can account for unknown or uncertain parts.…”

Section: Concluding Remarks and Outlookmentioning

confidence: 99%

“…for MLP the number of hidden layers and the therein covered numbers of nodes, or in case of Partial Least Square/Projection to Latent Structures (PLS) the number of latent variables) can be addressed with the Akaike Information Criterion or Bayesian Information Criteria, the latter being more suitable for models with large numbers of parameters (Lee, Vanrolleghem, & Park, 2005;Peres, Oliveira, & de Azevedo, 2008;von Stosch, Peres, de Azevedo, & Oliveira, 2010). Also other statistical criteria can be applied (Bollas et al, 2003;Kim & Chang, 2000) to evaluate the estimations obtained with different sized nonparametric models.…”

Section: Nonparametric Model Discriminationmentioning

confidence: 99%

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

Stosch¹,

Oliveira²,

Peres³

et al. 2014

Computers & Chemical Engineering

311

198

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

“…This training data will be obtained after following the same procedure presented in [39] and then applied in [40,41]. A similar procedure was also used in [42].…”

Section: Neural State Estimatormentioning

confidence: 99%

Neural Network-Based State Estimation for a Closed-Loop Control Strategy Applied to a Fed-Batch Bioreactor

et al. 2017

View full text Add to dashboard Cite

The lack of online information on some bioprocess variables and the presence of model and parametric uncertainties pose significant challenges to the design of efficient closed-loop control strategies. To address this issue, this work proposes an online state estimator based on a Radial Basis Function (RBF) neural network that operates in closed loop together with a control law derived on a linear algebra-based design strategy. The proposed methodology is applied to a class of nonlinear systems with three types of uncertainties: (i) time-varying parameters, (ii) uncertain nonlinearities, and (iii) unmodeled dynamics. To reduce the effect of uncertainties on the bioreactor, some integrators of the tracking error are introduced, which in turn allow the derivation of the proper control actions. This new control scheme guarantees that all signals are uniformly and ultimately bounded, and the tracking error converges to small values. The effectiveness of the proposed approach is illustrated on the basis of simulated experiments on a fed-batch bioreactor, and its performance is compared with two controllers available in the literature.

show abstract

A novel identification method for hybrid (N)PLS dynamical systems with application to bioprocesses

Cited by 37 publications

References 25 publications

An efficient method for the numerical integration of measured variable dependent ordinary differential equations

An efficient method for the numerical integration of measured variable dependent ordinary differential equations

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

Neural Network-Based State Estimation for a Closed-Loop Control Strategy Applied to a Fed-Batch Bioreactor

Contact Info

Product

Resources

About