Application of the hottest crystallizer expert sys

Application of mold expert system in continuous caster

Abstract: mold expert system is a tool to allocate, collect, store, compare and visualize all data parameters of mold. Compared with the set value, these data can help analyze the problems in the casting process or generate alarm information or change the casting process. The operation of the crystallizer expert system is very beneficial for the operator to monitor the casting process. Because of its visibility, employees can find the best and most stable casting process

keyword: mold expert system breakout prediction mold heat flow

preface

in the continuous casting process of molten steel, the mold is a very important equipment in the caster equipment, and the working condition of the mold will directly affect the slab quality. In order to ensure the stability of production and improve the slab quality, we should try our best to get the data parameters related to the mold work, especially the data that directly affects the equipment and slab quality, and improve and optimize them to make the operation of caster equipment more stable and coordinated. Breakout prediction technology has only detected breakouts from the early stage. After studying the causes and mechanism of breakouts, detecting the signs of breakouts, alarming and controlling, establishing prediction mathematical models, and using artificial intelligence technology and visualization technology, the breakout rate has been greatly reduced

at present, iron and steel enterprises require high output in design and can ensure the quality of billets. Continuous casters play a key role in the output and quality of steel grades. The breakout event in the casting process will cost a lot of money to repair the equipment and affect production at the same time. In order to avoid breakout, it is necessary to detect the breakout when it is possible to remind the operator or automatically reduce the drawing speed. Mold experts are a very useful tool for developing new steel grades and optimizing the pouring process. Mold expert system includes two software packages: mold thermodynamics software package and mold vibration software package

1. Mold thermodynamics software package

the monitoring of mold thermodynamics software package is mainly based on the temperature measured by the thermocouple installed in the copper plate

1.1 temperature monitoring

the temperature change behavior of molten steel at the early stage of solidification in the mold can be obtained, so we can monitor the temperature change inside the mold, From the temperature monitoring, we can know:

· heat flow distribution on the surface of narrow and wide edge copper plate

· influence of mold flux on temperature distribution

· instantaneous behavior of heat flow (slope, speed change)

· shrinkage behavior (when the shell leaves)

· observe the dangerous area

· improve the quality of slab and the safety of operation

· improve the practical ability of pouring

temperature monitoring depicts the temperature value measured by the thermocouple installed in the mold, the thermocouple embedded in the mold copper plate, and transmits the detected copper plate temperature to the system. According to the detected signal and the installation position of each thermocouple, The temperatures of various parts of the billet in the mold can be depicted in different colors. The upper part of Figure 1 depicts all temperatures in different colors, which are updated once a second. The following figure shows the temperature on the inner surface of the mold, reflects the temperature distribution on the surface of the slab, and specifically depicts the historical temperature records of the upper row of thermocouples

Figure 1 temperature monitoring distribution diagram

1.2 breakout prediction

thermocouples are embedded in the mold copper plate to give important temperature information before molten steel solidification. These data are used for breakout prediction and also used as the basis to clarify the heat transfer inside the mold. Breakout prediction system is used as a tool to distribute, collect, store, compare and visualize all data parameters of crystallizer (Fig. 2). These data can help analyze the problems in the casting process or generate alarm information or change the casting process. The operation of the breakout prediction system is very beneficial for the operators to monitor the casting process. Due to its visibility, the employees can find the best and most stable casting process

Figure 2 system structure diagram

1.2.1 breakout prediction principle

the billet shell breaks beyond its mechanical strength for some reason, and the requirements for the experimental process are also very high. It solidifies and breaks repeatedly according to the vibration cycle of the mold. If the billet shell fails to solidify, the crack will expand downward and horizontally at a speed lower than the blank drawing speed. When the billet shell breaks to the bottom of the mold, it will leak. When the billet shell breaks, The molten steel is in direct contact with the mold wall, so there will be a temperature rise there. In this way, if many thermocouples are arranged in the width direction and casting direction of the mold, the rupture and expansion of the shell can be known from the temperature measured by the thermocouples. As shown in the figure, when the rupture of the shell reaches the upper row of thermocouples, the temperature of the upper row of thermocouples rises. When the rupture of the shell moves down to the lower row of thermocouples after a certain time, the temperature of the lower row of thermocouples rises as much as that of the upper row of thermocouples. Identifying this temperature rise can predict the leakage (Figure 3)

1.2.2 detection of mold temperature and logical judgment model

use the detection of mold temperature and logical judgment model to predict the pull-out type, and judge based on the comparison between the measured value of each thermocouple (the temperature difference between the upper and lower rows of thermocouples) and the set value

under normal casting process conditions, due to the growth of the shell, the average temperature of the upper row thermocouple is higher than that of the lower row thermocouple (I). When a part of the mold is broken due to bonding shell (II), and reaches the upper row thermocouple, the temperature rises, and the temperature of the lower row thermocouple has no effect (III). After a period of time, when the crack of the shell moves down to the lower row thermocouple, The temperature of the lower row of thermocouples rises (Ⅳ), while the temperature of the upper row of thermocouples decreases (Ⅴ) because it is not affected by the rupture of the blank shell. The temperature of the upper row decreases so much that it is lower than the temperature of the lower row (Ⅵ), In this way, the breakout prediction alarm can be sent out by monitoring the change of temperature (Fig. 4)

Fig. 4 breakout prediction formation schematic diagram

1.3 calculating the heat flow exchange of the mold

using the calculation of the heat exchange of the mold to prevent leakage is to monitor the total heat absorption through the mold, determine the minimum heat absorption according to the statistical data, and determine a suitable thickness of the billet shell before drawing. In order to predict the leakage, it is only necessary to detect the total heat absorption, which can be defined as "unit heat absorption", and can also be measured by the heat flow through the mold wall (kw/m). This heat flow QA can be expressed as follows:

QA = (CP ρ wW Δ T)/a

where

qa -- heat absorption per unit area

cp -- specific heat of water

ρ W -- density of water

w -- cooling water flow

Δ T -- inlet and outlet temperature of cooling water

a -- exposed surface area of the crystallizer

in order to accurately understand the heat absorption of the crystallizer, it is necessary to measure and monitor the heat flow on each side of the crystallizer respectively, because the heat flow value on one side of the crystallizer may be lower than the dangerous value, and the heat transfer is still greater than the acceptable minimum value. If only the total heat transfer is measured and considered normal, steel leakage will occur. When the heat flow QA exceeds the limit, tell the operator to pay attention and take measures to avoid steel leakage

Figure 5 heat flow monitoring screen

2 mold vibration software package

according to the hydraulic cylinder stroke and pressure feedback of the mold vibration system, the friction between the billet and the mold copper plate is calculated. According to the characteristics of the copper plate temperature rising and the friction increasing at the bonding place before the breakout, the breakout is predicted through the detection device according to a certain judgment model

2.1 mold friction

the friction of the mold is calculated according to the amplitude and pressure feedback of the hydraulic vibration hydraulic cylinder. There are two different operating conditions that must be distinguished: the vibration operation in the casting process is "hot", and the offline vibration is "cold". The difference of friction between the two conditions can lead to the mold friction. This calculation method is based on the off-line cold force and thermal force of physical work, which is very important to obtain the value of friction. The thermal force is calculated from the amplitude and pressure measured by the mold expert system. It includes the off-line cooling force and the friction between copper plate and billet. Therefore, the off-line cooling force must be measured. The offline cooling force is mainly based on the amplitude, and its calculation equation is:

offline cooling force = a + b * a + C * a

A, B, C are coefficient constants

A is amplitude

2.2 friction monitoring

mold friction monitoring screen (Figure 6) is divided into three parts, so that the measurement and calculation of the most important friction can be monitored in one window. The upper left is the display of friction and amplitude; On the upper right is the so-called friction cycle; The working curve, casting speed, mold liquid level, etc. are displayed in the lower area

Figure 6 vibration monitoring screen

3 use a variety of technologies to display the crust of the mold, and predict the pull leakage

this leads to the loss of the customer. Due to the progress of calculation technology, image processing technology and the deepening of mathematical model, continuous casting mold visualization technology has emerged, which combines the mold molten steel level, thermocouple temperature, cooling water temperature and flow, mold width, friction The amplitude and other data are sent to a special visual PC for data processing and numerical simulation calculation to form a visual picture (including the temperature field distribution, solidification and crusting of molten steel in the mold). The early thermal phenomenon model only involves the heat conduction problem, and the thermophysical values related to the heat conduction equation are mostly simplified. With the development of numerical calculation technology and the improvement of computer running speed, The velocity field and temperature field are obtained by coupling fluid flow with heat transfer, making it more practical (Fig. 7)

Figure 7 multi-level control of experimental force, storage and printing of experimental data and other functional technologies show the crusting of mold slab

4 Conclusion

breakout prediction system of slab continuous casting has experienced the process from simple formula calculation to complex mathematical model calculation. Breakout prediction system has more and more powerful functions, more and more mathematical models are used in the system, more and more parameters are involved, and the interface display is easier to understand, see, explain and operate. It is believed that in the near future, there will be more advanced new methods for detecting breakout

author kongdecai

(the third steel plant of Jinan Iron and Steel Group Corporation, Jinan 250101, Shandong)

author unit: the third steel plant of Jinan Iron and Steel Group

address: the third steel plant of Jinan Iron and Steel Group, No. 57, Gongye North Road, Jinan, Shandong 250101

email:kongdecai2004@ (end)