Mixing Models to Improve Gauge Prediction for Cold Rolling Mills

Ettler, Pavel; Andrýsek, Josef

doi:10.3182/20070821-3-ca-2919.00061

Cited by 3 publications

(3 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ettler, Kárný, and Nedoma (2007) proposed several methods including the predictors-as-regressors (PR) approach, which consists of recursively estimating each candidate model as above, and then running another recursive estimation with the predictors from each model as regressors. Practical aspects of the approach were elaborated in Ettler and Andrýsek (2007).…”

Section: Discussionmentioning

confidence: 99%

Online Prediction Under Model Uncertainty via Dynamic Model Averaging: Application to a Cold Rolling Mill

2010

Self Cite

View full text Add to dashboard Cite

We consider the problem of online prediction when it is uncertain what the best prediction model to use is. We develop a method called Dynamic Model Averaging (DMA) in which a state space model for the parameters of each model is combined with a Markov chain model for the correct model. This allows the "correct" model to vary over time. The state space and Markov chain models are both specified in terms of forgetting, leading to a highly parsimonious representation. As a special case, when the model and parameters do not change, DMA is a recursive implementation of standard Bayesian model averaging, which we call recursive model averaging. The method is applied to the problem of predicting the output strip thickness for a cold rolling mill, where the output is measured with a time delay. We found that when only a small number of physically motivated models were considered and one was clearly best, the method quickly converged to the best model, and the cost of model uncertainty was small; indeed DMA performed slightly better than the best physical model. When model uncertainty and the number of models considered were large, our method ensured that the penalty for model uncertainty was small. At the beginning of the process, when control is most difficult, we found that DMA over a large model space led to better predictions than the single best performing physically motivated model. We also applied the method to several simulated examples, and found that it recovered both constant and time-varying regression parameters and model specifications quite well.

show abstract

Section: Discussionmentioning

confidence: 99%

Online Prediction Under Model Uncertainty via Dynamic Model Averaging: Application to a Cold Rolling Mill

2010

Self Cite

View full text Add to dashboard Cite

show abstract

“…where v r denotes the ratio of the input and output strip speeds, h 1 is the deviation of the input strip thickness from its nominal value and z stands for the so called uncompensated rolling gap. See (Ettler and Andrýsek, 2007) for details.…”

Section: Real Data Examplementioning

confidence: 99%

“…See, e.g., (Bohlin, 1991) for remarks on black and grey box modelling. On-line mixing of several proven process models -which can be considered as continuous decision making -turns out to be a solid compromise solution of all three requirements, at least for the process of cold strip rolling from which comes the motivation for this research (Ettler and Andrýsek, 2007).…”

Section: Introductionmentioning

confidence: 99%

ADAPTIVE CONTINUOUS HIERARCHICAL MODEL-BASED DECISION MAKING - For Process Modelling with Realistic Requirements

Dedecius¹,

Ettler²

2011

Proceedings of the 8th International Conference on Informatics in Control, Automation and Robotics

View full text Add to dashboard Cite

Industrial model-based control often relies on parametric models. However, for certain operational conditions either the precise underlying physical model is not available or the lack of relevant or reliable data prevents its use. A popular approach is to employ the black box or grey box models, releasing the theoretical rigor. This leads to several candidate models being at disposal, from which the (often subjectively) prominent one is selected. However, in the presence of model uncertainty, we propose to benefit from a subset of credible models. The idea behind the multimodelling approach is closely related to hierarchical modelling methodology. By using several modelling levels, it is possible to achieve relatively high quality and robust solution, providing a way around typical constraints in industrial applications.

show abstract