# Contactless Inductive Flow Tomography for Real-Time Control of Electromagnetic Actuators in Metal Casting

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Materials and Methods

#### 2.1. Contactless Inductive Flow Tomography

#### 2.2. Mini-LIMMCAST

^{2}. The SEN has an inner diameter of 12 mm and an outer diameter of 21 mm with two side ports directed downwards at an angle of 15°.

## 3. Compensation for the Impact of the EMBr on the CIFT Measurement System

#### 3.1. Krasnosel’skii–Pokrowski Model of Hysteresis

#### 3.2. Compensation for the Impact of the EMBr on the Flow-Induced Magnetic Field Measurements

#### 3.2.1. Compensation via the Numerical Model of Hysteresis

#### 3.2.2. Congruency-Based Compensation

## 4. Real-Time Reconstruction

## 5. Conclusions

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Conflicts of Interest

## References

- Dauby, P.H. Continuous casting: Make better steel and more of it! Rev. Métall.
**2012**, 109, 113–136. [Google Scholar] [CrossRef] - Hibbeler, L.C.; Thomas, B.G. Mold Slag Entrainment Mechanisms in Continuous Casting Molds. In Proceedings of the AISTech 2013, Pittsburgh, PA, USA, 6–9 May 2013. [Google Scholar]
- Cho, S.-M.; Thomas, B.G. Electromagnetic Forces in Continuous Casting of Steel Slabs. Metals
**2019**, 9, 471. [Google Scholar] [CrossRef] [Green Version] - Schurmann, D.; Glavinić, I.; Willers, B.; Timmel, K.; Eckert, S. Impact of the Electromagnetic Brake Position on the Flow Structure in a Slab Continuous Casting Mold: An Experimental Parameter Study. Metall. Mater. Trans. B
**2019**, 51, 61–78. [Google Scholar] [CrossRef] - Mul, M. Real-time Mould Temperature Control for the Purpose of Consistent Slab Quality. In Proceedings of the METEC & 4th ESTAD, Düsseldorf, Germany, 25–29 June 2019. [Google Scholar]
- Hedin, G.; Kamperman, A.; Sedén, M.; Fröjdh, K.; Pejnefors, J. Exploring opportunities in mold temperature monitoring utilizing Fiber Bragg Gratings. In Proceedings of the SCANMET V 2016, Luleå, Sweden, 12–15 June 2016. [Google Scholar]
- Watzinger, J.; Watzinger, I. Latest advancements in ESP casting technology. In Proceedings of the European Continous Casting Conference 2021, Bari, Italy, 20–22 October 2021. [Google Scholar]
- Thomas, B.G.; Yuan, Q.; Sivaramakrishnan, S.; Shi, T.; Vanka, S.P.; Assar, M.B. Comparison of Four Methods to Evaluate Fluid Velocities in a Continuous Slab Casting Mold. ISIJ Int.
**2001**, 41, 10. [Google Scholar] [CrossRef] [Green Version] - Köhler, K.U.; Andrezejewski, P.; Julius, E.; Haubrich, H. Steel Flow Velocity Measurement and Flow Pattern Monitoring in the Mould. In Proceedings of the 78th Steelmaking Conference Proceedings, Nashville, TN, USA, 2–5 April 1995; pp. 445–450. [Google Scholar]
- Zheng, J.; Liu, R.; Wang, X.; Xu, G.; Lyu, Z.; Kolesnikov, Y.; Na, X. An Online Contactless Investigation of the Meniscus Velocity in a Continuous Casting Mold Using Lorentz Force Velocimetry. Metall. Mater. Trans. B
**2020**, 51, 558–569. [Google Scholar] [CrossRef] - Stefani, F.; Gerbeth, G. A contactless method for velocity reconstruction in electrically conducting fluids. Meas. Sci. Technol.
**2000**, 11, 758–765. [Google Scholar] [CrossRef] - Wondrak, T.; Galindo, V.; Gerbeth, G.; Gundrum, T.; Stefani, F.; Timmel, K. Contactless inductive flow tomography for a model of continuous steel casting. Meas. Sci. Technol.
**2010**, 21, 045402. [Google Scholar] [CrossRef] - Wondrak, T.; Eckert, S.; Gerbeth, G.; Klotsche, K.; Stefani, F.; Timmel, K.; Peyton, A.J.; Terzija, N.; Yin, W. Combined electromagnetic tomography for determining two-phase flow characteristics in the submerged entry nozzle and in the mold of a continuous casting model. Metall. Mater. Trans. B
**2011**, 42, 1201–1210. [Google Scholar] [CrossRef] - Wondrak, T.; Ratajczak, M.; Timmel, K.; Stefani, F.; Eckert, S. Recent progress on contactless inductive flow tomography for continuous casting in the presence of strong static magnetic fields. In Proceedings of the Electromagnetic Processing of Materials 2015 (EPM2015), Cannes, France, 12–16 October 2015; pp. 491–494. [Google Scholar]
- Ratajczak, M.; Wondrak, T.; Stefani, F. A gradiometric version of contactless inductive flow tomography: Theory and first applications. Philos. A
**2016**, 374, 20150330. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Ratajczak, M.; Wondrak, T. Analysis, design and optimization of compact ultra-high sensitivity coreless induction coil sensors. Meas. Sci. Technol.
**2020**, 31, 065902. [Google Scholar] [CrossRef] - Ratajczak, M.; Wondrak, T.; Zürner, T.; Stefani, F. Enhancing robustness and applicability of contactless inductive flow tomography. In Proceedings of the 2015 IEEE SENSORS, Busan, Korea, 1–4 November 2015. [Google Scholar]
- Ratajczak, M.; Wondrak, T.; Timmel, K.; Stefani, F.; Eckert, S. Flow Visualization by Means of Contactless Inductive Flow Tomography in the Presence of a Magnetic Brake. J. Manuf. Sci. Prod.
**2015**, 15, 41–48. [Google Scholar] [CrossRef] - Ratajczak, M.; Wondrak, T.; Stefani, F.; Eckert, S. Numerical and experimental investigation of the contactless inductive flow tomography in the presence of strong static magnetic fields. Magnetohydrodynamics
**2015**, 51, 461–471. [Google Scholar] [CrossRef] - Glavinić, I.; Ratajczak, M.; Stefani, F.; Wondrak, T. Flow monitoring for continuous steel casting using Contactless Inductive Flow Tomography (CIFT). In Proceedings of the IFAC 2020 World Congress, Berlin, Germany, 11–17 July 2020. [Google Scholar]
- Glavinić, I.; Stefani, F.; Eckert, S.; Wondrak, T. Real time flow control during continuous casting with contactless inductive flow tomography. In Proceedings of the 10th International Conference on Electromagnetic Processing of Materials 2021, Riga, Latvia, 14–16 June 2021; pp. 170–175. [Google Scholar]
- Glavinić, I.; Muttakin, I.; Abouelazayem, S.; Blishchik, A.; Stefani, F.; Eckert, S.; Soleimani, M.; Saidani, I.; Hlava, J.; Kenjereš, S.; et al. Laboratory Investigation of Tomography-Controlled Continuous Steel Casting. Sensors
**2022**, 22, 2195. [Google Scholar] [CrossRef] [PubMed] - Stefani, F.; Gundrum, T.; Gerbeth, G. Contactless Inductive Flow Tomography. Phys. Rev. E
**2004**, 70, 056306. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Krasnosel’skiǐ, M.A.; Pokrovskiǐ, A.V. Systems with Hysteresis; Springer: Berlin/Heidelberg, Germany, 1989. [Google Scholar]
- Banks, H.T.; Kurdila, A.J.; Webb, G. Identification of Hysteretic Control Influence Operators Representing Smart Actuators Part I: Formulation. Math. Probl. Eng.
**1997**, 3, 287–328. [Google Scholar] [CrossRef] - Jiang, H.; Ji, H.; Qiu, J.; Chen, Y. A Modified Prandtl-Ishlinskii Model for Modeling Asymmetric Hysteresis of Piezoelectric Actuators. IEEE Trans. Ultrason. Ferroelectr. Freq. Control
**2010**, 57, 1200–1210. [Google Scholar] [CrossRef] [PubMed] - Campbell, S.L.; Meyer, C.D., Jr. Generalized Inverses of Linear Transformations; Dover: New York, NY, USA, 1991; ISBN 978-0-486-66693-8. [Google Scholar]
- Mayergoyz, I.D. Mathematical Models of Hysteresis. Phys. Rev. Lett.
**1986**, 22, 603–608. [Google Scholar] - Boyd, S.; Vandenberghe, L. Introduction to Applied Linear Algebra: Vectors, Matrices, and Least Squares, 1st ed.; Cambridge University Press: Cambridge, UK, 2018. [Google Scholar]
- Calvetti, D.; Reichel, L.; Shuibi, A. L-Curve and Curvature Bounds for Tikhonov Regularization. Numer. Algorithms
**2004**, 35, 301–314. [Google Scholar] [CrossRef] - Wondrak, T.; Jacobs, R.T.; Faber, P. Fast Reconstruction Algorithm for Contactless Inductive Flow Tomography. In Proceedings of the 10th International Conference on Advanced Computer Information Technologies, Deggendorf, Germany, 16–18 September 2020; pp. 217–220. [Google Scholar]

**Figure 2.**Experimental setup comprising CIFT coils and sensors as well as the electromagnetic brake (EMBr). Two excitation coils generate a primarily vertical magnetic field. Fourteen gradiometric coils, seven on each narrow side, are used to measure the flow-induced magnetic field. The EMBr generates a strong magnetic field below the submerged entry nozzle (SEN), perpendicular to the wide side of the mould. (

**a**) Photograph of the mould and CIFT coils. (

**b**) Sketch adopted from Schurmann et al. [4].

**Figure 3.**Measurement of the magnetic field for an experiment where the EMBr current ${I}_{B}$ is changed during the measurement for the sensors 1–7. The current was turned on at $t\approx $ 28 s to ${I}_{B}$ = 200 A and turned off at $t\approx 44\text{}\mathrm{s}$ to ${I}_{B}$ = 0 A.

**Figure 4.**Measured offset of the in-phase component of the magnetic field for an experiment where the EMBr current was cycled from 0 A to 600 A and back to 0 A in steps of $\mathsf{\Delta}{I}_{B}$ = 25 A. The plots show only measurements for sensors 1–7. (

**a**) plot over time; (

**b**) plot of offset over EMBr current where the hysteresis can be observed.

**Figure 6.**Generalized output of a single KP operator. (

**a**) Input function ${I}_{B}\left(t\right)$ in relation to parameters ${p}_{1}$, ${p}_{2}$ and a. (

**b**) Generalized output of the KP operator for the given input function. Colours represent the values of the input and the corresponding output of the KP operator.

**Figure 7.**Preisach plane. (

**a**) Discretization scheme of the Preisach plane. (

**b**) Evolution of the shape of the KP operators on the Preisach plane.

**Figure 8.**EMBr current profiles used for KP model training and validation. (

**a**) Current profile used for identification of model weights. (

**b**) Current profile used for validation of the model.

**Figure 9.**Plots of the absolute difference (red) between the measured in-phase magnetic field offset (green +) and the value predicted by the KP model of hysteresis (blue).

**Figure 10.**Congruency-based compensation for the impact of the EMBr current on the flow-induced magnetic field measured by sensors 1–7. (

**a**) EMBr current profile. (

**b**) Mean value and standard deviation of the magnetic field offset after the compensation for the periods between the current changes. (

**c**) Uncompensated in-phase component of the magnetic field. (

**d**) Compensated in-phase component of the magnetic field.

**Figure 11.**Results of congruency-based compensation for sensors 1–7, utilizing the numerical KP model of hysteresis for the experiment from Figure 10. (

**a**) Time plot of the compensated in-phase component of the magnetic field. (

**b**) Mean value and standard deviation of the compensated in-phase component of the magnetic field for periods in which the EMBr current is kept constant.

**Figure 12.**Reconstruction of the time dependant flow field at $t=$ 17 s. (

**a**) Three-second time average of the velocity field obtained from the numerical simulation. (

**b**) Reconstructed velocity field for the optimal regularization parameter selected via the L-curve method.

**Figure 13.**The L-curve for the Reconstruction of the time dependant flow field at $t=$ 17 s. Annotated (a)–(f) are the regularization parameters for which the velocity field is shown in Figure 14. ${\lambda}_{opt}$ is marked by a red triangle and annotated with (e). (a) $\lambda =$ 1.000 ×${10}^{-8}$. (b) $\lambda =$ 1.000 ×${10}^{-9}$. (c) $\lambda =$ 1.000 ×${10}^{-10}$. (d) $\lambda =$ 2.125 ×${10}^{-11}$. (e) ${\lambda}_{opt}=$ 1.896 ×${10}^{-11}$. (f) $\lambda =$ 1.000 ×${10}^{-12}$.

**Figure 14.**Velocity reconstructions for the regularization parameters annotated in Figure 13. (

**a**) $\lambda =$ 1.000 ×${10}^{-8}$. (

**b**) $\lambda =$ 1.000 ×${10}^{-9}$. (

**c**) $\lambda =$ 1.000 ×${10}^{-10}$. (

**d**) $\lambda =$ 2.125 ×${10}^{-11}$. (

**e**) ${\lambda}_{opt}=$ 1.896 ×${10}^{-11}$, the optimal regularization parameter selected via the L-curve method. (

**f**) $\lambda =$ 1.000 ×${10}^{-12}$.

**Figure 15.**Time evolution plots of error, correlation and optimal regularization parameter. (

**a**) Error and correlation plot of reconstructions for optimally selected regularization parameter ${\lambda}_{opt}$. (

**b**) Temporal evolution of the optimal regularization parameter ${\lambda}_{opt}$.

**Figure 16.**Absolute difference of error and correlation between the velocity fields reconstructed by the standard and the real-time algorithm.

**Figure 17.**(

**a**) Three-second time average of the velocity field obtained from the numerical simulation at $t=$ 40 s. (

**b**) Reconstructed velocity field. (

**c**) Time plot of error and correlation for velocities close to the narrow face of the mould. $\left|x\right|\ge $ 100 mm. y = 0 mm, $z\ge $ 450 mm.

**Figure 18.**Comparison of the jet impingement point obtained from the numerical simulations and the CIFT real-time reconstruction for the same numerical velocity field.

**Table 1.**Reconstruction parameters $\lambda $ and corresponding values of error and correlation for the velocity reconstructions in Figure 14. Reconstruction (e) is done with the optimal regularization parameter calculated by the L-curve method.

Reconstruction | $\mathit{\lambda}$ | Correlation | Error |
---|---|---|---|

(a) | 1.000 × ${10}^{-8}$ | 0.65378 | 0.57316 |

(b) | 1.000 × ${10}^{-9}$ | 0.69040 | 0.52493 |

(c) | 1.000 × ${10}^{-10}$ | 0.70435 | 0.50469 |

(d) | 2.125 × ${10}^{-11}$ | 0.69651 | 0.52070 |

(e) | 1.896 × ${\mathbf{10}}^{-\mathbf{11}}$ | 0.69601 | 0.52175 |

(f) | 1.000 × ${10}^{-12}$ | 0.69084 | 0.53272 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Glavinić, I.; Galindo, V.; Stefani, F.; Eckert, S.; Wondrak, T.
Contactless Inductive Flow Tomography for Real-Time Control of Electromagnetic Actuators in Metal Casting. *Sensors* **2022**, *22*, 4155.
https://doi.org/10.3390/s22114155

**AMA Style**

Glavinić I, Galindo V, Stefani F, Eckert S, Wondrak T.
Contactless Inductive Flow Tomography for Real-Time Control of Electromagnetic Actuators in Metal Casting. *Sensors*. 2022; 22(11):4155.
https://doi.org/10.3390/s22114155

**Chicago/Turabian Style**

Glavinić, Ivan, Vladimir Galindo, Frank Stefani, Sven Eckert, and Thomas Wondrak.
2022. "Contactless Inductive Flow Tomography for Real-Time Control of Electromagnetic Actuators in Metal Casting" *Sensors* 22, no. 11: 4155.
https://doi.org/10.3390/s22114155