Real-Time Estimation of Molten Steel Flow in Continuous Casting Mold
- PDF / 1,490,051 Bytes
- 8 Pages / 593.972 x 792 pts Page_size
- 17 Downloads / 182 Views
CTION
IN the continuous casting process, molten steel is poured into the mold through the submerged entry nozzle (SEN), and the molten steel is cooled by the mold in which water pipes are embedded, as shown in Figure 1(a). A slab is drawn from the bottom of the mold while solidified shells are formed on the mold plates. To maintain the proper molten steel level, i.e., meniscus level, the flow rate of the molten steel is adjusted by changing the opening of the gate inside the SEN. In addition, thermocouples are installed in the mold to monitor the state of the solidified shell and molten steel flow. It is well known that slab defects derived from mold powder or nonmetallic inclusions are reduced by appropriately controlling the molten steel flow with a magnetic field. For instance, Figure 1(b) shows a static magnetic field[1,2] which is used to stabilize the jet flow from SEN ports for the reduction of inclusions. Another flow control method in which the molten steel is stirred
YOSHINARI HASHIMOTO is with the Steel Research Laboratory, Cyber-Physical System R&D Department, JFE Steel Corp., Fukuyama, 721-8510, Japan. Contact e-mail: [email protected] AKITOSHI MATSUI is with the Steel Research Laboratory, Steelmaking Research Department, JFE Steel Corp., Fukuyama, 721-8510, Japan. TOSHIYUKI HAYASE is with the Institute of Fluid Science, Tohoku University, Sendai, 980-8577, Japan. MANABU KANO is with the Department of Systems Science, Graduate School of Informatics, Kyoto University, Kyoto, 606-8501, Japan. Manuscript submitted September 29, 2019.
METALLURGICAL AND MATERIALS TRANSACTIONS B
by a traveling magnetic field[3] has also been proposed to wash away the inclusions trapped at the interface between the solidified shell and molten steel. Various numerical simulation models have been proposed[4–10] to optimize the magnetic field conditions of the flow control devices. Takatani et al.[4] constructed a flow model that considers the change of the free surface shape. Thomas et al.[5] proposed a model that can simulate the transport and entrapment of inclusions. However, when these conventional models are applied to the actual plants, estimation errors occur due to various disturbances such as the clogging inside the SEN. The nozzle clogging causes the asymmetrical flow,[11,12] in which the flow rates from the left and right SEN ports become uneven. Real-time estimation of changes in the flow pattern caused by the disturbances makes it possible to control the molten steel flow accurately by the dynamic manipulation of the magnetic fields. Suzuki et al.[13] proposed a real-time flow pattern estimation method, in which the mold temperature is converted to the flow velocity. However, its application is limited to the vicinity of the mold plate, and it cannot estimate the three-dimensional flow pattern in the entire mold. On the other hand, Hayase et al.[14] developed a state estimation algorithm that combines the sensor information and model calculation. The algorithm adjusts the virtual external force in the model using measurement
Data Loading...
