kth.sePublications
Change search
Link to record
Permanent link

Direct link
Alternative names
Publications (10 of 374) Show all publications
Ghosh, A., Abdalmoaty, M., Chatterjee, S. & Hjalmarsson, H. (2024). DeepBayes—An estimator for parameter estimation in stochastic nonlinear dynamical models. Automatica, 159, Article ID 111327.
Open this publication in new window or tab >>DeepBayes—An estimator for parameter estimation in stochastic nonlinear dynamical models
2024 (English)In: Automatica, ISSN 0005-1098, E-ISSN 1873-2836, Vol. 159, article id 111327Article in journal (Refereed) Published
Abstract [en]

Stochastic nonlinear dynamical systems are ubiquitous in modern, real-world applications. Yet, estimating the unknown parameters of stochastic, nonlinear dynamical models remains a challenging problem. The majority of existing methods employ maximum likelihood or Bayesian estimation. However, these methods suffer from some limitations, most notably the substantial computational time for inference coupled with limited flexibility in application. In this work, we propose DeepBayes estimators that leverage the power of deep recurrent neural networks. The method consists of first training a recurrent neural network to minimize the mean-squared estimation error over a set of synthetically generated data using models drawn from the model set of interest. The a priori trained estimator can then be used directly for inference by evaluating the network with the estimation data. The deep recurrent neural network architectures can be trained offline and ensure significant time savings during inference. We experiment with two popular recurrent neural networks — long short term memory network (LSTM) and gated recurrent unit (GRU). We demonstrate the applicability of our proposed method on different example models and perform detailed comparisons with state-of-the-art approaches. We also provide a study on a real-world nonlinear benchmark problem. The experimental evaluations show that the proposed approach is asymptotically as good as the Bayes estimator.

Place, publisher, year, edition, pages
Elsevier Ltd, 2024
Keywords
Deep learning, Dynamical systems, Nonlinear system identification, Parameter estimation, Recurrent neural networks
National Category
Control Engineering Other Electrical Engineering, Electronic Engineering, Information Engineering Probability Theory and Statistics
Identifiers
urn:nbn:se:kth:diva-339038 (URN)10.1016/j.automatica.2023.111327 (DOI)001161034600001 ()2-s2.0-85174673962 (Scopus ID)
Note

QC 20231129

Available from: 2023-11-29 Created: 2023-11-29 Last updated: 2024-02-27Bibliographically approved
Parsa, J., Rojas, C. R. & Hjalmarsson, H. (2023). Application-Oriented Input Design With Low Coherence Constraint. IEEE Control Systems Letters, 7, 193-198
Open this publication in new window or tab >>Application-Oriented Input Design With Low Coherence Constraint
2023 (English)In: IEEE Control Systems Letters, E-ISSN 2475-1456, Vol. 7, p. 193-198Article in journal (Refereed) Published
Abstract [en]

In optimal input design input sequences are typically generated without paying attention to the correlations between the regressors of the model to be estimated. In fact, in many cases high correlations are beneficial. This is in contrast to the requirements in sparse estimation. Mutual coherence is the maximum of these correlations, and in case the parameter vector is known to be sparse, we need a low mutual coherence in order to estimate it accurately. This contribution proposes adding a constraint on the mutual coherence to the optimal input design problem to improve the accuracy of estimated sparse models. The proposed method can be combined with any sparse estimation algorithm to estimate the parameters of a model. However, we focus in particular on the bound on the mutual coherence required for Orthogonal Matching Pursuit (OMP), a well-known algorithm in sparse estimation. Furthermore, we analyze the effect of the proposed method on the required input power. Finally, we evaluate, in a numerical study, the performance of the proposed method compared to state-of-the-art algorithms for input design.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2023
Keywords
Coherence, Matching pursuit algorithms, Degradation, Optimization, Ellipsoids, Mathematical models, Costs, System identification, input design, sparse estimation, mutual coherence
National Category
Probability Theory and Statistics
Identifiers
urn:nbn:se:kth:diva-315885 (URN)10.1109/LCSYS.2022.3187319 (DOI)000824789000003 ()2-s2.0-85133775635 (Scopus ID)
Note

QC 20220728

Available from: 2022-07-28 Created: 2022-07-28 Last updated: 2023-08-25Bibliographically approved
Parsa, J., Rojas, C. R. & Hjalmarsson, H. (2023). Coherence-Based Input Design for Nonlinear Systems. IEEE Control Systems Letters, 7, 2934-2939
Open this publication in new window or tab >>Coherence-Based Input Design for Nonlinear Systems
2023 (English)In: IEEE Control Systems Letters, E-ISSN 2475-1456, Vol. 7, p. 2934-2939Article in journal (Refereed) Published
Abstract [en]

Many off-the-shelf generic non-linear model structures have inherent sparse parametrizations. Volterra series and non-linear Auto-Regressive with eXogeneous inputs (NARX) models are examples of this. It is well known that sparse estimation requires low mutual coherence, which translates into input sequences with certain low correlation properties. This letter highlights that standard optimal input design methods do not account for this requirement which may lead to designs unsuitable for this type of model structure. To tackle this problem, this letter proposes incorporating a coherence constraint to standard input design problems. The coherence constraint is defined as the ratio between the diagonal and non-diagonal entries of the Fisher information matrix (FIM) and can be easily added to any input design problem for nonlinear systems, while the resulting problem remains convex. This letter provides a theoretical analysis of how the range of the optimal objective function of the original problem is affected by the coherence constraint. Additionally, this letter presents numerical evaluations of the proposed approach's performance on a Volterra series model in comparison to state-of-the-art algorithms.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2023
Keywords
System identification, input design, nonlinear system, sparse estimation, mutual coherence
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-334291 (URN)10.1109/LCSYS.2023.3291230 (DOI)001033525300013 ()2-s2.0-85163782955 (Scopus ID)
Note

QC 20230818

Available from: 2023-08-18 Created: 2023-08-18 Last updated: 2023-08-23Bibliographically approved
Pasquini, M. & Hjalmarsson, H. (2023). E2-RTO: An Exploitation-Exploration Approach for Real Time Optimization. In: IFAC-PapersOnLine, 22nd IFAC World Congress: . Paper presented at 22nd IFAC World Congress, Jul 9 2023 - Jul 14 2023, Yokohama, Japan. (pp. 1423-1430). Elsevier BV, 56
Open this publication in new window or tab >>E2-RTO: An Exploitation-Exploration Approach for Real Time Optimization
2023 (English)In: IFAC-PapersOnLine, 22nd IFAC World Congress, Elsevier BV , 2023, Vol. 56, p. 1423-1430Conference paper, Published paper (Refereed)
Abstract [en]

In Real-Time Optimization, the problem of optimizing the operating conditions of a plant is solved through iterative methods that directly use plant measurements. In this paper, a novel Exploitation-Exploration approach to solve this problem is proposed. Differently from a previously proposed Two-Step approach, consisting in strict exploitation and consecutive model update, in the Exploitation-Exploration approach an additional exploration step is included, in which an exploration input is constructed so that the updated model would approximately minimize the expectation of the objective function value for the new exploitation step. The proposed algorithm is described and its validity is shown through a numerical example, where it is compared to some common Real-Time Optimization schemes.

Place, publisher, year, edition, pages
Elsevier BV, 2023
Keywords
Real-Time Optimization
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-343700 (URN)10.1016/j.ifacol.2023.10.1819 (DOI)2-s2.0-85184958399 (Scopus ID)
Conference
22nd IFAC World Congress, Jul 9 2023 - Jul 14 2023, Yokohama, Japan.
Note

Part of ISBN: 978-171387234-4

QC 20240228

Available from: 2024-02-22 Created: 2024-02-22 Last updated: 2024-02-28Bibliographically approved
Wang, Y., Pasquini, M., Colin, K., Mäkinen, M., Schwarz, H., Chotteau, V., . . . Jacobsen, E. W. (2023). Model-based Medium Optimization Methodologies in High-cell Density Perfusion Culture. In: : . Paper presented at Cell Culture Engineering XVIII, Cancun, Mexico, April 23-28 2023.
Open this publication in new window or tab >>Model-based Medium Optimization Methodologies in High-cell Density Perfusion Culture
Show others...
2023 (English)Conference paper, Poster (with or without abstract) (Refereed)
Keywords
Perfusion cultures, Medium optimization, Model-based techniques
National Category
Control Engineering Bioprocess Technology
Identifiers
urn:nbn:se:kth:diva-329571 (URN)
Conference
Cell Culture Engineering XVIII, Cancun, Mexico, April 23-28 2023
Note

Yu Wang and Mirko Pasquini contributed equally to this work

QC 20230704

Available from: 2023-06-21 Created: 2023-06-21 Last updated: 2023-07-04Bibliographically approved
Bombois, X., Colin, K., Van den Hof, P. M. J. & Hjalmarsson, H. (2023). On the informativity of direct identification experiments in dynamical networks. Automatica, 148, 110742, Article ID 110742.
Open this publication in new window or tab >>On the informativity of direct identification experiments in dynamical networks
2023 (English)In: Automatica, ISSN 0005-1098, E-ISSN 1873-2836, Vol. 148, p. 110742-, article id 110742Article in journal (Refereed) Published
Abstract [en]

Data informativity is a crucial property to ensure the consistency of the prediction error estimate. This property has thus been extensively studied in the open-loop and in the closed-loop cases, but has only been briefly touched upon in the dynamic network case. In this paper, we consider the prediction error identification of the modules in a row of a dynamic network using the full input approach. Our main contribution is to propose a number of easily verifiable data informativity conditions for this identification problem. Among these conditions, we distinguish a sufficient data informativity condition that can be verified based on the topology of the network and a necessary and sufficient data informativity condition that can be verified via a rank condition on a matrix of coefficients that are related to a full-order model structure of the network. These data informativity conditions allow to determine different situations (i.e., different excitation patterns) leading to data informativity. In order to be able to distinguish between these different situations, we also propose an optimal experiment design problem that allows to determine the excitation pattern yielding a certain pre-specified accuracy with the least excitation power.

Place, publisher, year, edition, pages
Elsevier BV, 2023
Keywords
Dynamic network identification, Data informativity, Optimal experiment design
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-324820 (URN)10.1016/j.automatica.2022.110742 (DOI)000928279800010 ()2-s2.0-85143489645 (Scopus ID)
Note

QC 20230317

Available from: 2023-03-17 Created: 2023-03-17 Last updated: 2023-03-17Bibliographically approved
Colin, K., Hjalmarsson, H. & Bombois, X. (2023). Optimal exploration strategies for finite horizon regret minimization in some adaptive control problems. In: 22nd IFAC World Congress Yokohama, Japan, July 9-14, 2023: . Paper presented at 22nd IFAC World Congress, Yokohama, Japan, Jul 9 2023 - Jul 14 2023 (pp. 2564-2569). Elsevier BV, 56
Open this publication in new window or tab >>Optimal exploration strategies for finite horizon regret minimization in some adaptive control problems
2023 (English)In: 22nd IFAC World Congress Yokohama, Japan, July 9-14, 2023, Elsevier BV , 2023, Vol. 56, p. 2564-2569Conference paper, Published paper (Refereed)
Abstract [en]

In this work, we consider the problem of regret minimization in adaptive minimum variance and linear quadratic control problems. Regret minimization has been extensively studied in the literature for both types of adaptive control problems. Most of these works give results of the optimal rate of the regret in the asymptotic regime. In the minimum variance case, the optimal asymptotic rate for the regret is log(T) which can be reached without any additional external excitation. On the contrary, for most adaptive linear quadratic problems, it is necessary to add an external excitation in order to get the optimal asymptotic rate of √T. In this paper, we will actually show from a theoretical study, as well as, in simulations that when the control horizon is pre-specified a lower regret can be obtained with either no external excitation or a new exploration type termed immediate.

Place, publisher, year, edition, pages
Elsevier BV, 2023
Series
IFAC-PapersOnLine, ISSN 2405-8963 ; 56
Keywords
adaptive control, linear quadratic regulator, linear systems, minimum variance controller, Regret minimization
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-343694 (URN)10.1016/j.ifacol.2023.10.1339 (DOI)2-s2.0-85184960468 (Scopus ID)
Conference
22nd IFAC World Congress, Yokohama, Japan, Jul 9 2023 - Jul 14 2023
Note

QC 20240222

Part of ISBN 9781713872344

Available from: 2024-02-22 Created: 2024-02-22 Last updated: 2024-02-22Bibliographically approved
Pasquini, M., Colin, K., Chotteau, V. & Hjalmarsson, H. (2022). A Lyapunov based heuristic to speed up convergence of a feedback optimization framework with experiment batches-application to bioprocess manufacturing. In: IFAC PAPERSONLINE: . Paper presented at 9th IFAC Conference on Foundations of Systems Biology in Engineering (FOSBE), AUG 28-31, 2022, Cambridge, MA (pp. 135-140). Elsevier BV, 55(23)
Open this publication in new window or tab >>A Lyapunov based heuristic to speed up convergence of a feedback optimization framework with experiment batches-application to bioprocess manufacturing
2022 (English)In: IFAC PAPERSONLINE, Elsevier BV , 2022, Vol. 55, no 23, p. 135-140Conference paper, Published paper (Refereed)
Abstract [en]

In this work a heuristic to speed up the convergence of a feedback-based optimization scheme, when experiments can be run in batches, is discussed. The proposed approach allows to select the most promising experiment in the batch, as the one maximising the decrease of an associated Lyapunov function, and to define the inputs for the next batch, based on this. We suggest the application of the scheme to a biological setting, with the goal of maximizing the concentration of a product of interest in a bioreactor under a continuous perfusion framework, while at the same time minimizing the yield of a toxic byproduct. The potential of the approach is exposed by means of a simple synthetic example. 

Place, publisher, year, edition, pages
Elsevier BV, 2022
Keywords
Dynamics and control of biological systems, Systems biology for (red, green, blue, white) biotechnology, Next generation methods and tools for systems and synthetic biology
National Category
Control Engineering
Research subject
Biotechnology
Identifiers
urn:nbn:se:kth:diva-327447 (URN)10.1016/j.ifacol.2023.01.029 (DOI)000968848300009 ()2-s2.0-85162006176 (Scopus ID)
Conference
9th IFAC Conference on Foundations of Systems Biology in Engineering (FOSBE), AUG 28-31, 2022, Cambridge, MA
Note

QC 20230529

Available from: 2023-05-29 Created: 2023-05-29 Last updated: 2024-02-07Bibliographically approved
Colin, K., Hjalmarsson, H. & Chotteau, V. (2022). Gaussian process modeling of macroscopic kinetics: a better-tailored kernel for Monod-type kinetics. In: 10th Vienna International Conference on Mathematical Modelling MATHMOD 2022 Vienna Austria, 27–29 July 2022: . Paper presented at 10th Vienna International Conference on Mathematical Modelling (MATHMOD), JUL 27-29, 2022, Tech Univ Wien, ELECTR NETWORK (pp. 397-402). Elsevier BV, 55(20)
Open this publication in new window or tab >>Gaussian process modeling of macroscopic kinetics: a better-tailored kernel for Monod-type kinetics
2022 (English)In: 10th Vienna International Conference on Mathematical Modelling MATHMOD 2022 Vienna Austria, 27–29 July 2022, Elsevier BV , 2022, Vol. 55, no 20, p. 397-402Conference paper, Published paper (Refereed)
Abstract [en]

In bioprocesses, it is important to model the kinetics of the macroscopic rates of reactions since these are required to catch the dynamical aspects of a process. In [Wang et al. 2020], a modeling method involving Gaussian processes has been developed, using a kernel especially designed for the modeling of Monod-type kinetics (activation, inhibition, double component, neutral effect). However, as will be illustrated in this paper, when the number of training data is limited or the metabolite concentration data do not have large variations (which is generally the case for real-life data), this kernel can yield inaccurate models for the kinetics. In this paper, we develop a new kernel better tailored for the modeling of Monod-type kinetics and we show that it has good modeling performances in the case of a limited number of data. The idea is to use the particular structure of Monod-type functions in the design of the kernel, i.e., we incorporate prior knowledge in the modeling.

Place, publisher, year, edition, pages
Elsevier BV, 2022
Series
IFAC PAPERSONLINE, ISSN 2405-8963 ; 55
Keywords
Gaussian process, Nonlinear system identification, Monod model, Kinetics, Macroscopic modeling
National Category
Control Engineering
Research subject
Chemical Engineering
Identifiers
urn:nbn:se:kth:diva-320408 (URN)10.1016/j.ifacol.2022.09.127 (DOI)000860842100067 ()2-s2.0-85142254386 (Scopus ID)
Conference
10th Vienna International Conference on Mathematical Modelling (MATHMOD), JUL 27-29, 2022, Tech Univ Wien, ELECTR NETWORK
Projects
Competence center AdBIOPRO
Note

QC 20221110

Available from: 2022-11-10 Created: 2022-11-10 Last updated: 2024-02-07Bibliographically approved
Parsa, J., Rojas, C. R. & Hjalmarsson, H. (2022). Optimal Input Design for Sparse System Identification. In: 2022 EUROPEAN CONTROL CONFERENCE (ECC): . Paper presented at European Control Conference (ECC), JUL 12-15, 2022, London, ENGLAND (pp. 1999-2004). IEEE
Open this publication in new window or tab >>Optimal Input Design for Sparse System Identification
2022 (English)In: 2022 EUROPEAN CONTROL CONFERENCE (ECC), IEEE , 2022, p. 1999-2004Conference paper, Published paper (Refereed)
Abstract [en]

In this contribution we consider sparse linear regression problems. It is well known that the mutual coherence, i.e. the maximum correlation of the regressors, is important for the ability of any algorithm to recover the sparsity pattern of an unknown parameter vector from data. A low mutual coherence improves the ability of recovery. In optimal experiment design this requirement may be in conflict with other objectives encoded by the desired Fisher matrix. In this contribution we alleviate this issue by combining optimal input design with a recently proposed approach to achieve low mutual coherence by way of a linear coordinate transformation. The resulting optimization problem is solved using cyclic minimization. Via simulations we demonstrate that the resulting algorithm is able to achieve a Fisher matrix which results in a performance close to the performance if the sparsity would have been known, while at the same time being able to recover the sparsity pattern.

Place, publisher, year, edition, pages
IEEE, 2022
National Category
Control Engineering Computer Sciences
Identifiers
urn:nbn:se:kth:diva-320689 (URN)10.23919/ECC55457.2022.9838242 (DOI)000857432300276 ()2-s2.0-85135115246 (Scopus ID)
Conference
European Control Conference (ECC), JUL 12-15, 2022, London, ENGLAND
Note

Part of proceedings; ISBN 978-3-907144-07-7

QC 20221031

Available from: 2022-10-31 Created: 2022-10-31 Last updated: 2023-06-08Bibliographically approved
Organisations
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0002-9368-3079

Search in DiVA

Show all publications