Inferential sensor-based adaptive principal components analysis of mould bath level for breakout defect detection and evaluation in continuous casting

Bouhouche, Salah; Mentouri, Zoheir; Ziani, Said; Bast, Jürgen

doi:10.1016/j.asoc.2015.04.042

Cited by 21 publications

(4 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the ever-increasing demands on the steel industry to lower emissions and increase efficiency comes the equally demanding requirements for a higher quality product. This is leading to an increase in the number of sensors and monitoring devices used in the steelmaking process [1][2][3], whereby particular focus has been during the rolling process, where roll force, strip thickness, and even phase transformation after rolling (austenite to ferrite) can be monitored [4][5][6][7][8]. However, the conditions for higher temperature process monitoring are more challenging, for example, during casting or thermo-mechanical processing, but there is demand for real-time feedback to allow consistent and high quality steel to be produced.…”

Section: Introductionmentioning

confidence: 99%

Characterisation of the Solidification of a Molten Steel Surface Using Infrared Thermography

Slater

Hechu

Davis

et al. 2019

Metals

View full text Add to dashboard Cite

Infrared thermography provides an option for characterising surface reactions and their effects on the solidification of steel under different gas atmospheres. In this work, infrared thermography has been used during solidification of Twin Induced Plasticity (TWIP) steel in argon, carbon dioxide and nitrogen atmospheres using a confocal scanning laser microscope (CSLM). It was found that surface reactions resulted in a solid oxide film (in carbon dioxide) and decarburisation, along with surface graphite formation (in nitrogen). In both cases the emissivity and, hence, the cooling rate of the steel was affected in distinct ways. Differences in nucleation conditions (free surface in argon compared to surface oxide/graphite in carbon dioxide/nitrogen) as well as chemical composition changes (decarburisation) affected the liquidus and solidus temperatures, which were detected by thermal imaging from the thermal profile measured.

show abstract

Section: Introductionmentioning

confidence: 99%

Characterisation of the Solidification of a Molten Steel Surface Using Infrared Thermography

Slater

Hechu

Davis

et al. 2019

Metals

View full text Add to dashboard Cite

show abstract

“…They proposed expectation robust algorithm to improve the conventional robust PCA when incomplete data and missing information exist. Salah et al developed an adaptive PCA (APCA) for breakout defect detection and evaluation in a continuous casting process. They compared PCA, APCA, neural network, and SVR methods and found that the APCA exhibits the best performance.…”

Section: State Of the Artmentioning

confidence: 99%

Soft sensor design and fault detection using Bayesian network and probabilistic principal component analysis

Mohammadi

Zarghami

Lefebvre

et al. 2019

J Adv Manuf & Process

View full text Add to dashboard Cite

Bayesian network (BN) and probabilistic principal component analysis (PPCA) are powerful tools in artificial intelligence. In this study, two intelligent soft sensors, designed based on BN and PPCA, were proposed for fault detection and data prediction in chemical processes. A gas sweetening process was considered as a case study in which the existing data for H 2 S concentration in sweet gas stream were incomplete. In order to detect faults during the operation, a soft sensor was designed using BN and PPCA for predicting H 2 S concentration in the sweet gas stream. Then, another soft sensor was developed based on the BN for fault detection, considering the Gaussian mixture model with hidden nodes. The Tennessee-Eastman challenge process was used to assess the efficiency of the fault detection method. Results showed that the predicted values of the soft sensors are very close to real data. The fault occurring in this process was detected in early stages, demonstrating the good performance of the proposed fault detection system. It was also shown that the performance of the BN is better than the PPCA in prediction of incomplete data. Moreover, the confidence interval can be evaluated for the predicted values when using the BN. An uncertainty analysis was performed to assess the quality of predicted data, and it was observed that the error magnitude of predicted data is smaller when using the BN compared with the PPCA. In particular, the results showed that the BN model is capable of estimating H 2 S concentration with nearly 96.1% accuracy, whereas the accuracy of the PCA-based method was 93.2%. K E Y W O R D SBayesian network, fault detection, gas sweetening process, probabilistic principal component analysis, soft sensor, Tennessee-Eastman challenge process

show abstract

“…The mold flux with high melting temperature and low melting rate cannot generate sufficient liquid to fill the shell/mold gap, thus the lubrication of the initial shell would be deteriorated, especially during the positive stripping time when the initial shell is continuously drained by the mold oscillation. The second is the casting parameters, such as superheat, casting speed, mold oscillation frequency and stroke, etc . The initial shell will be thin and easy to be torn during the casting process with a high superheat and casting speed.…”

Section: Breakout Versus Heat Transfermentioning

confidence: 99%

Initial Solidification and Its Related Heat Transfer Phenomena in the Continuous Casting Mold

Wang

Zhu

Zhou

2017

steel research int.

View full text Add to dashboard Cite

The initial solidification of molten steel in the mold is very important, as it is the origin of the final slab, which associates with the surface defects and determines the final quality of casting slab. In this article, six main topics, including the initial solidification study method, molten steel solidification, mold flux structure, fluid flow, oscillation marks formation, and breakout are reviewed with the purpose to explore the intrinsic mechanism of the complex initial solidification process and its related multi-phase heat transfer phenomena. The results summarized here can provide a foundational understanding of the initial solidification and heat transfer phenomena occurring in the continuous casting mold.

show abstract

Inferential sensor-based adaptive principal components analysis of mould bath level for breakout defect detection and evaluation in continuous casting

Cited by 21 publications

References 28 publications

Characterisation of the Solidification of a Molten Steel Surface Using Infrared Thermography

Characterisation of the Solidification of a Molten Steel Surface Using Infrared Thermography

Soft sensor design and fault detection using Bayesian network and probabilistic principal component analysis

Initial Solidification and Its Related Heat Transfer Phenomena in the Continuous Casting Mold

Contact Info

Product

Resources

About