This content is not included in
your SAE MOBILUS subscription, or you are not logged in.
Hybrid Modeling of a Catalyst with Autoencoder Based Selection Strategy
Technical Paper
2020-01-2178
ISSN: 0148-7191, e-ISSN: 2688-3627
This content contains downloadable datasets
Annotation ability available
Sector:
Language:
English
Abstract
Two substantially different methods have become popular in building fast computing catalyst models: physico-chemical approaches focusing on dimensionality reduction and machine learning approaches. Data driven models are known to be very fast computing and to achieve high accuracy but they can lack of extrapolation capability. Physico-chemical models are usually slower and less accurate but superior regarding robustness. The robustness can even be reinforced by implementing an extended Kalman filter, which enables the model to adapt its states based on actual sensor values, even if the sensors are drifting.
The present study proposes a combination of both approaches into one hybrid model, keeping the robustness of the physico-chemical model in edge cases while also achieving the accuracy of the data based model in well-known regimes. The output of the hybrid model is controlled by an autoencoder, utilizing methods well known from the field of anomaly detection.
It was validated that the autoencoder has comparable extrapolation properties as the neural network representing the catalyst. Hence, the reconstruction error of the autoencoder can be used as an indicator for the data driven model’s performance. The physico-chemical model is applied when the reconstruction loss surpasses a chosen threshold and a low performance of the machine learning approach is anticipated due to extrapolation.
The present study proves that an increase of accuracy and robustness can be achieved by the combination of both modeling approaches.
Authors
Citation
Kühne, J., März, C., Werfel, J., Gelbert, G. et al., "Hybrid Modeling of a Catalyst with Autoencoder Based Selection Strategy," SAE Technical Paper 2020-01-2178, 2020, https://doi.org/10.4271/2020-01-2178.Data Sets - Support Documents
| Title | Description | Download |
|---|---|---|
| Unnamed Dataset 1 | ||
| Unnamed Dataset 2 | ||
| Unnamed Dataset 3 |
Also In
References
- Gundlapally , S. , Papadimitriou , I. , Wahiduzzaman , S. , and Gu , T. Development of ECU Capable Grey-Box Models from Detailed Models Application to a SCR Reactor Emission Control Science and Technology 124 136 2016
- Hsieh , M.F. , and Wang , J. Diesel Engine SCR Systems: Modelling, Measurements and Control New York Springer 2014
- Rumelhart , D.E. and McClelland , J.L. Learning Internal Representations by Error Propagation Parallel Distributed Processing: Explorations in the Microstructure of Cognition: Foundations 1987 318 362
- Hinton , G.E. , and Salakhutdinov , R.R. Reducing the Dimensionality of Data with Neural Networks Science 28 07 504 507 2006
- Hinton , G.E. , Osindero , S. , and Te , Y.-W. A Fast Learning Algorithm for Deep Belief Nets Neural Computation 06 1527 1554 2006
- Kullback , S. , and Leibler , R.A. On Information and Sufficiency The Annals of Mathematical Statistics 03 79 86 1951
- Zong , B. , Song , Q. , Min , M.R. , Cheng , W. et al. Deep Autoencoding Gaussian Mixture Model for Unsupervised Anomaly Detection ICLR 2018 2018
- Malhotra , P. , Vig , L. , Shroff , G. , and Agarwal , P. Long Short Term Memory Networks for Anomaly Detection in Time Series Conference: 23rd European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning 2015
- Guo , Y. , Liao , W. , Wang , Q. , Yu , L. et al. Multidimensional Time Series Anomaly Detection: A GRU-Based Gaussian Mixture Variational AutoencoderApproach ACML 2018
- DeVries , T. and Taylor , G. 2018
- Henriksson , J. , Berger , C. , Borg , M. , Tornberg , L. et al. Towards Structured Evaluation of Deep Neural Network Supervisors 2019 IEEE International Conference On Artificial Intelligence Testing (AITest) 2019
- Krizhevsky , A. 2009
- Russakovsky , O. , Deng , J. , Su , H. , Krause , J. et al. ImageNet Large Scale Visual Recognition Challenge International Journal of Computer Vision (IJCV) 211 252 2015
- LeCun , Y. and Cortes , C. 2010 http://yann.lecun.com/exdb/mnist/
- Pedregosa , F. , Varoquaux , G. , Gramfort , A. , Michel , V. et al. Scikit-learn: Machine Learning in Python Journal of Machine Learning Research 2825 2830 2011
- https://pytorch.org/docs/stable/index.html#pytorch-documentation 2020
- März , C. , Werfel , J.F.-J. , Kühne , J. , and Scholz , R. Approaches for a New Generation of Fast Computing Catalyst Models Emission Control Science and Technology 01 2020
- https://www.exothermia.com/exhaust-overview 2020
- Gelbert , G. , Friedrichs , O. , Heß , D. , and Lars , H. NH3 Filling Level Control on the Basis of Real-Time Capable Physical Model MTZ 2 60 65 2017
- Faghihi , E.M. , and Shamekhi , A.H. Development of a Neural Network Model for Selective Catalytic Reduction (SCR) Catalytic Converter and Ammonia Dosing Optimization Using Multi Objective Genetic Algorithm Chemical Engineering Journal 165 2 508 516 2010
- Schürholz , K. , Brückner , D. , and Abel , D. Modelling the Exhaust Gas Aftertreatment System of a SI Engine Using Artificial Neural Networks Topics in Catalysis 62 1-4 288 295 2018
- Hagan , M.T. , and Menhaj , M. Training Feed-Forward Networks with the Maquardt Algorithm IEEE Transactions on Neural Networks 5 989 993 1999
- Paszke , A. , Gross , S. , Massa , F. , Lerer , A. et al. PyTorch: An Imperative Style, High-Performance Deep Learning Library Advances in Neural Information Processing Systems 32 Curran Associates, Inc. 2019 8024 8035
- Bai , J. , Lu , F. , and Zhang , K. 2019 https://github.com/onnx/onnx
- Exothermia S.A. 2015 E22 E45
- Ciliberti , P.D. Development of NOx Estimator ECU Models for SCR After Treatment System in a Diesel Engine Bari Politecnico di Bari 2017