A hybrid neural network‐first principles approach to process modeling
Abstract
Abstract A hybrid neural network‐first principles modeling scheme is developed and used to model a fedbatch bioreactor. The hybrid model combines a partial first principles model, which incorporates the available prior knowledge about the process being modeled, with a neural network which serves as an estimator of unmeasured process parameters that are difficult to model from first principles. This hybrid model has better properties than standard “black‐box” neural network models in that it is able to interpolate and extrapolate much more accurately, is easier to analyze and interpret, and requires significantly fewer training examples. Two alternative state and parameter estimation strategies, extended Kalman filtering and NLP optimization, are also considered. When no a priori known model of the unobserved process parameters is available, the hybrid network model gives better estimates of the parameters, when compared to these methods. By providing a model of these unmeasured parameters, the hybrid network can also make predictions and hence can be used for process optimization. These results apply both when full and partial state measurements are available, but in the latter case a state reconstruction method must be used for the first principles component of the hybrid model.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Hybrid modeling techniques and applications have been growing in number since the early 90's.
“Indeed, hybrid modeling techniques and applications have been growing in number since the early 90's (Psichogios and Ungar 1992; Rico-Martínez et al. 1992; Thompson and Kramer 1994)”
From Perspectives on the integration between first-principles and data-driven modeling (2022) · cites this paper as Psichogios and Ungar 1992
Cites
Learning Internal Representations by Error Propagation
Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1985) · doi:10.21236/ada164453 · Cited by 16,501 · see citing works
Adaptive filtering, prediction and control
Ioannou, P. (1985) · Automatica · doi:10.1016/0005-1098(85)90014-7 · Cited by 4,487 · see citing works
Adaptive filtering prediction and control
Goodwin, G. C., & Sin, K. S. (1984) · Cited by 3,352 · see citing works
Anderson, B. D. O., Moore, J. B., & Eslami, M. (1982) · IEEE Transactions on Systems Man and Cybernetics · Cited by 3,195 · see citing works
Learning internal representations by error propagation
RUMELHART, D. E., HINTON, G. E., & WILLIAMS, R. J. (1988) · Readings in Cognitive Science · doi:10.1016/b978-1-4832-1446-7.50035-2 · Cited by 3,002 · see citing works
Beyond Regression : "New Tools for Prediction and Analysis in the Behavioral Sciences
Werbos, P. J. (1974) · Medical Entomology and Zoology · Cited by 2,772 · see citing works
Beyond regression : new fools for prediction and analysis in the behavioral sciences
Werbos, P. J. (1974) · Medical Entomology and Zoology · Cited by 2,746 · see citing works
Use of neural nets for dynamic modeling and control of chemical process systems
Bhat, N., & McAvoy, T. J. (1990) · Computers & Chemical Engineering · doi:10.1016/0098-1354(90)87028-n · Cited by 672 · see citing works
A simple procedure for pruning back-propagation trained neural networks
Karnin, E. D. (1990) · IEEE Transactions on Neural Networks · doi:10.1109/72.80236 · Cited by 672 · see citing works
Canonical form observer design for non-linear time-variable systems
Bestle, D., & Zeitz, M. J. (1983) · International Journal of Control · doi:10.1080/00207178308933084 · Cited by 561 · see citing works
Skeletonization: A Technique for Trimming the Fat from a Network via Relevance Assessment
Mozer, M. C., & Smolensky, P. (1988) · neural information processing systems · Cited by 540 · see citing works
Sensitivity analysis of initial value problems with mixed odes and algebraic equations
Caracotsios, M., & Stewart, W. E. (1985) · Computers & Chemical Engineering · doi:10.1016/0098-1354(85)85014-6 · Cited by 414 · see citing works
Nonlinear state feedback synthesis by global input/output linearization
Kravaris, C., & Chung, C. (1987) · AIChE Journal · doi:10.1002/aic.690330408 · Cited by 370 · see citing works
Exponential observers for nonlinear dynamic systems
Kou, S. R., Elliott, D. L., & TARN, T. J. (1975) · Information and Control · doi:10.1016/s0019-9958(75)90382-4 · Cited by 298 · see citing works
Artificial neural networks in process engineering
Willis, M. J., Massimo, C. D., Montague, G. A., et al. (1991) · IEE Proceedings D Control Theory and Applications · doi:10.1049/ip-d.1991.0036 · Cited by 259 · see citing works
Stinchcombe, & White (1989) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.1989.118640 · Cited by 250 · see citing works
Direct and indirect model based control using artificial neural networks
Psichogios, D. C., & Ungar, L. (1991) · Industrial & Engineering Chemistry Research · doi:10.1021/ie00060a009 · Cited by 225 · see citing works
Self-citationOn‐line prediction of fermentation variables using neural networks
Thibault, J., Van Breusegem, V., & Chéruy, A. (1990) · Biotechnology and Bioengineering · doi:10.1002/bit.260361009 · Cited by 220 · see citing works
Biegler, L. T., & Cuthrell, J. E. (1985) · Computers & Chemical Engineering · doi:10.1016/0098-1354(85)80004-1 · Cited by 175 · see citing works
A neural network architecture that computes its own reliability
Leonard, J. A., Kramer, M. A. M., & Ungar, L. (1992) · Computers & Chemical Engineering · doi:10.1016/0098-1354(92)80035-8 · Cited by 146 · see citing works
Self-citationComparison of two approaches to on-line parameter and state estimation of nonlinear systems
Jang, S. S., Joseph, B., & Mukai, H. (1986) · Industrial & Engineering Chemistry Process Design and Development · doi:10.1021/i200034a037 · Cited by 136 · see citing works
Multistep nonlinear predictive controller
Brengel, D. D., & Seider, W. D. (1989) · Industrial & Engineering Chemistry Research · doi:10.1021/ie00096a013 · Cited by 111 · see citing works
Mavrovouniotis, M. L., & Chang, S. (1992) · Computers & Chemical Engineering · doi:10.1016/0098-1354(92)80053-c · Cited by 106 · see citing works
Neural Network Modeling and an Extended DMC Algorithm to Control Nonlinear Systems
Hernandaz, E., & Arkun, Y. (1990) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.23919/acc.1990.4791169 · Cited by 65 · see citing works
A Neural Network Structure for System Identification
Haesloop, D., & Holt, B. R. (1990) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.23919/acc.1990.4791170 · Cited by 49 · see citing works
Application of modern estimation and identification techniques to chemical processes
Wells, C. H. (1971) · AIChE Journal · doi:10.1002/aic.690170433 · Cited by 41 · see citing works
A note on non-linear observers
Banks, S. P. (1981) · International Journal of Control · doi:10.1080/00207178108922522 · Cited by 35 · see citing works
A Comparison of Adaptive Estimation with Neural based Techniques for Bioprocess Application
Lant, P. A., Willis, M. J., Montague, G. A., et al. (1990) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.23919/acc.1990.4791115 · Cited by 28 · see citing works
ADAPTIVE CONTROL OF FEDBATCH BIOREACTORS
Dochain, D., & Bastin, G. L. (1990) · Chemical Engineering Communications · doi:10.1080/00986449008940684 · Cited by 22 · see citing works
Process Identification using Nonlinear Programming Techniques
Sistu, P. B., & Bequette, B. W. (1990) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.23919/acc.1990.4790994 · Cited by 16 · see citing works
Banks, S. P. (1980) · White Rose Research Online (University of Leeds, The University of Sheffield, University of York) · Cited by 12 · see citing works
Application of Dynamic Programming for Fermentative Ethanol Production by Zymomonas mobilis
Rivera, S. L., & Karim, M. N. (1990) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.23919/acc.1990.4791110 · Cited by 10 · see citing works
Multivariable Control of a Continuous Culture
Gostomski, P. A., Clesceri, L. S., & Bungay, H. R. (1996) · Springer eBooks · doi:10.1007/978-94-009-0177-3_3 · Cited by 1 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Raissi, M., Perdikaris, P., & Karniadakis, G. E. (2019) · Journal of Computational Physics · doi:10.1016/j.jcp.2018.10.045 · Cited by 20,384 · see citing works
Interpretable machine learning: Fundamental principles and 10 grand challenges
Rudin, C., Chen, C., Chen, Z., Huang, H., Semenova, L., & Zhong, C. (2022) · Statistics Surveys · doi:10.1214/21-ss133 · Cited by 958 · see citing works
Hybrid semi-parametric modeling in process systems engineering: Past, present and future
von Stosch, M., Oliveira, R., Peres, J., & Feyo de Azevedo, S. (2014) · Computers & Chemical Engineering · doi:10.1016/j.compchemeng.2013.08.008 · Cited by 490 · see citing works
Advances and opportunities in machine learning for process data analytics
Qin, S. J., & Chiang, L. H. (2019) · Computers & Chemical Engineering · doi:10.1016/j.compchemeng.2019.04.003 · Cited by 351 · see citing works
Perspectives on the integration between first-principles and data-driven modeling
Bradley, W., Kim, J., Kilwein, Z., Blakely, L., Eydenberg, M., Jalvin, J., Laird, C., & Boukouvala, F. (2022) · Computers & Chemical Engineering · doi:10.1016/j.compchemeng.2022.107898 · Cited by 231 · see citing works
Machine learning for biochemical engineering: A review
Mowbray, M., Savage, T., Wu, C., Song, Z., Cho, B. A., Del Rio-Chanona, E. A., & Zhang, D. (2021) · Biochemical Engineering Journal · doi:10.1016/j.bej.2021.108054 · Cited by 188 · see citing works
A review and perspective on hybrid modeling methodologies
Schweidtmann, A. M., Zhang, D., & von Stosch, M. (2024) · Digital Chemical Engineering · doi:10.1016/j.dche.2023.100136 · Cited by 178 · see citing works
Noll, P., & Henkel, M. (2020) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2020.10.018 · Cited by 60 · see citing works
- Authors
- Psichogios, D. C., & Ungar, L. H.
- Affiliations
- University of Pennsylvania (US)
- Corresponding author
- Ungar, Lyle H.
- ORCID
- Ungar, Lyle H.
- Venue
- AIChE Journal
- Publisher
- Wiley
- Published
- 1992-10
- Volume
- 38
- Issue
- 10
- Pages
- 1499-1511
- ISSN
- 0001-1541 (ISSN-L)
- 1547-5905
- OpenAlex
- W1968326286
- Cited by
- 982
- References
- 33
- Citations by year
- 2026: 85 · 2025: 76 · 2024: 81 · 2023: 75 · 2022: 69 · 2021: 76 · 2020: 41 · 2019: 30 · 2018: 20 · 2017: 18 · 2016: 10 · 2015: 6 · 2014: 20 · 2013: 18 · 2012: 8
- FWCI
- 16.36
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- AIChE Journal (publishedVersion)
- Topics
- Physical Sciences › Engineering › Control and Systems Engineering › Control Systems and Identification
- Physical Sciences › Engineering › Control and Systems Engineering › Fault Detection and Control Systems
- Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
- Keywords
- process modeling, parameter estimation, state estimation, extended Kalman filter, nonlinear programming, state reconstruction, process optimization
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
