Methods for predicting fracture failure and methods for operating continuous casting machines.

Abstract

The method for predicting fracture failure according to the invention includes: a step of receiving input of the dimensions of the molded fiber drawn out of the mold located in the continuous molding machine; a step of detecting the mold temperatures with multiple thermometers embedded in the mold; a step of performing interpolation corresponding to the dimensions of the molded fiber at the detected temperatures detected by the multiple thermometers; a step of calculating temperature variations by comparing them with the temperatures before the first stage; a step of collecting the standard deviation of the temperature variations of the thermometers in the second stage; a step of normalizing the temperature variations by dividing the temperature variations by the standard deviation; a step of calculating the component in the direction perpendicular to the influence coefficient vector obtained by principal component analysis, which is the degree of deviation from the normal operating point when no fracture failure occurs, based on the temperature variation calculated by performing interpolation;and the step of predicting fracture failure based on the degree of deviation.

Description

Breakout prediction method and continuous casting machine operation method

[0001] The present invention relates to a breakout prediction method and a continuous casting machine operation method.

[0002] Patent Documents 1 and 2 disclose methods for predicting breakouts by detecting characteristic temperature behavior using a temperature sensor placed in a mold, while Patent Document 3 discloses a method for calculating the degree of deviation from the mold temperature during normal operation when no breakout has occurred, and predicting a breakout based on the calculated degree of deviation.

[0003] Japanese Patent Publication No. 63-47545 Japanese Patent Laid-Open No. 2005-296979 Japanese Patent No. 6950860

[0004] Breakouts are broadly divided into two types: binding breakouts and foreign matter breakouts. In binding breakouts, the slab sticks to the mold, often resulting in relatively large temperature fluctuations. In the case of foreign matter breakouts, the slab is caused by the casting process progressing with foreign matter such as powder adhering to the surface of the slab, so the range and magnitude of the change in mold temperature is extremely small.

[0005] The methods disclosed in Patent Documents 1 and 2 detect breakouts by examining the temporal change and spatial distribution of mold temperature, based on the characteristic temperature at the time of actual breakout occurrence. Restrictive breakouts and foreign object breakouts are detected by different methods. Restrictive breakouts, which have relatively large temperature changes, are identified by the temperature relationship between different temperatures at different casting heights (usually higher temperatures above the casting direction, and higher temperatures below the casting direction in abnormal cases). However, foreign object breakouts are difficult to detect because the temperature changes are small, so they are identified by the magnitude of temperature fluctuations and the downward propagation of the temperature change. However, because mold temperatures do not have a uniform distribution during operation and vary depending on the conditions inside the mold, false detections and missed detections are common.

[0006] In addition, the method disclosed in Patent Document 3 predicts breakouts based on the degree of deviation from the mold temperature during normal operation. Therefore, breakouts caused by binding, such as when a slab seizes in the mold and a relatively large temperature fluctuation is observed, can be adequately detected. On the other hand, breakouts caused by foreign matter such as powder adhering to the surface of the slab during casting cannot be adequately detected because the range and magnitude of the change in mold temperature are extremely small.

[0007] The present invention has been made in view of the above-mentioned problems, and an object of the present invention is to provide a breakout prediction method and a continuous casting machine operation method that are capable of detecting foreign object entrapment breakouts with sufficient detection accuracy, even when the range of change or temperature fluctuation in the mold temperature is small.

[0008] In order to solve the above-mentioned problems and achieve the object, (1) a breakout prediction method according to the present invention is characterized by comprising the steps of: inputting dimensions of a cast piece to be withdrawn from a mold in a continuous casting machine; detecting the temperature of the mold using a plurality of thermometers embedded in the mold; performing an interpolation process on the temperatures detected by the plurality of thermometers according to the dimensions of the cast piece; calculating a temperature change by comparing the detected temperatures with the temperature before a first period; obtaining a standard deviation of the temperature change for each thermometer within a second period; standardizing the temperature change by dividing the temperature change by the standard deviation; calculating a component in a direction orthogonal to an influence coefficient vector obtained from principal component analysis based on the temperature change calculated by performing the interpolation process as a deviation from normal operation in which no breakout has occurred; and predicting the breakout based on the deviation.

[0009] (2) The breakout prediction method according to the present invention is characterized in that, in the invention (1) above, in the step of performing the interpolation process, an interpolation process is performed at the center point of each of a plurality of calculation cells equally divided according to the dimensions of the slab for the temperature detected by each of the plurality of thermometers to calculate a temperature.

[0010] (3) The breakout prediction method according to the present invention is characterized in that, in the invention (2) above, the number of calculation cells is kept constant even if the dimensions of the slab are changed.

[0011] (4) In the breakout prediction method according to the present invention, in the invention of (2) or (3) above, the step of calculating the deviation degree includes finding an average value of the temperatures of the plurality of calculation cells that are positioned at the same distance from the upper end of the mold in the pouring direction of molten steel into the mold, finding a difference between the average value and the temperature of each of the plurality of calculation cells, and calculating the deviation degree from the found difference using the influence coefficient vector.

[0012] (5) The breakout prediction method according to the present invention is characterized in that, in the invention of (4) above, in the step of predicting a breakout, the breakout is predicted when the calculated individual deviation of the computational cell exceeds a predetermined first threshold value, or when the calculated total deviation of the computational cells exceeds a predetermined second threshold value.

[0013] (6) A breakout prediction method according to the present invention is any one of the inventions (1) to (5) above, characterized in that the influence coefficient vector is a sensitivity coefficient vector whose components are sensitivity coefficients for each of the plurality of temperature change amounts.

[0014] (7) A method for operating a continuous casting machine according to the present invention is characterized in that the casting speed is reduced when a breakout is predicted based on the breakout prediction method according to any one of the above (1) to (6) inventions.

[0015] The breakout prediction method and continuous casting machine operation method according to the present invention have the effect of being able to detect foreign object-entrapment-related breakouts, which are caused by small changes in mold temperature or small temperature fluctuations, with sufficient detection accuracy.

[0016] FIG. 1 is a schematic diagram showing the overall configuration of a continuous casting machine according to an embodiment. FIG. 2 is a perspective view showing the overall configuration of a mold with an embedded thermometer in a continuous casting machine according to an embodiment. FIG. 3(a) is a diagram illustrating the state of the molten steel and solidified shell in the mold when a breakout precursor occurs. FIG. 3(b) is a diagram showing the state of a fractured portion of the solidified shell when a breakout precursor occurs. FIG. 4(a) shows the temperature distribution of the mold at the moment when seizure occurs. FIG. 4(b) shows the temperature distribution of the mold 10 seconds after the seizure occurs. FIG. 5 shows a foreign object entrapment breakout. FIG. 6 is a flowchart showing an example of the procedure for a breakout prediction method according to an embodiment. FIG. 7 shows the correlation of temperatures detected by thermometers during normal operation when no breakout occurs. FIG. 8 shows the correlation of temperatures detected by thermometers when precursors of breakout, such as seizure, occur. FIG. 9(a) shows the relationship between the temperatures detected by the thermometers and the temperatures at which interpolation processing is performed when a wide slab is removed from the bottom end of the mold. FIG. 9(b) shows the relationship between the temperature detected by the thermometer and the temperature obtained by the interpolation process in a case where the width of the slab withdrawn from the bottom end of the mold is narrow. FIG. 10 shows the positional relationship between the thermometer and the calculation cell, which are located at the same distance from the top end of the mold. FIG. 11(a) shows the time series change in deviation in a case where seizure (restrictive breakout) was confirmed. FIG. 11(b) shows the change in deviation at each calculation cell position in a case where seizure (restrictive breakout) was confirmed. FIG. 12(a) shows the time series change in deviation in a case where foreign object entrapment (foreign object entrapment breakout) was confirmed. FIG. 12(b) shows the change in deviation at each calculation cell position in a case where foreign object entrapment (foreign object entrapment breakout) was confirmed. FIG. 13(a) shows a case where the total deviation has a peak, but no abnormalities were found in the slab inspection results. FIG. 13(b) is a diagram showing the deviations in each calculation cell in a case where the total deviation has a peak but no abnormality was found as a result of the slab inspection.FIG. 14 is a diagram showing the results of breakout prediction performed using the breakout prediction method according to the embodiment and a conventional method.

[0017] Hereinafter, embodiments of the breakout prediction method and the continuous casting machine operation method according to the present invention will be described, although the present invention is not limited to these embodiments.

[0018] Fig. 1 is a schematic diagram showing the overall configuration of a continuous casting machine 1 according to an embodiment. As shown in Fig. 1, the continuous casting machine 1 according to an embodiment includes a tundish 3 into which molten steel 2 is poured, a copper mold 5 that cools the molten steel 2 poured from the tundish 3 through an immersion nozzle 4, a plurality of strand support rolls 7 that transport a semi-solidified strand 6 withdrawn from the mold 5, and a determination unit 20 that determines a breakout precursor phenomenon based on the temperature detected by a thermometer 8 embedded in the mold 5. Note that, although a thermocouple is used as the thermometer 8 in this embodiment, the present invention is not limited to this.

[0019] FIG. 2 shows the thermometer 8 in the continuous casting machine 1 according to the embodiment. １，１ ~8 ｍ，ｎ 2 is a perspective view showing a schematic configuration of a mold 5 in which a cooling water passage (not shown) is embedded. As shown in Fig. 2, the mold 5 includes a pair of long-side cooling plates 5a and a pair of short-side cooling plates 5b, and is formed in a generally rectangular cylindrical shape that penetrates in the vertical direction. Inside the long-side cooling plates 5a and the short-side cooling plates 5b, cooling water passages (not shown) are formed along the inner wall surfaces, and cooling water is circulated through these cooling water passages to cool the molten steel 2.

[0020] In addition, a thermometer 8 is installed inside the long side cooling plate 5a of the mold 5. １，１ ~8 ｍ，ｎ is embedded at a predetermined depth from the outer wall surface of the long-side cooling plate 5a. １，１ ~8 ｍ，ｎ When there is no particular distinction between the thermometer 8 and the thermometer 8, the thermometer 8 is also simply referred to as the thermometer 8. １，１ ~8 ｍ，ｎ The thermometer 8 in the first stage is arranged in a configuration of three or more stages in the pouring direction A. １，１ ~8 １，ｎ , second stage thermometer 8 ２，１ ~8２，ｎ , and the n-th stage thermometer 8 ｍ，１ ~8 ｍ，ｎ In this embodiment, the pouring direction A is the direction in which the molten steel 2 is poured from the tundish 3 through the submerged entry nozzle 4 into the mold 5, and is the same direction as the direction in which the slab 6 is withdrawn from the lower end of the mold 5.

[0021] 2 is merely one example for explaining the present invention, and it is sufficient to arrange thermometers 8 on at least one of the pair of long side cooling plates 5a, at least one of the pair of short side cooling plates 5b, or all of the pair of long side cooling plates 5a and the pair of short side cooling plates 5b of the mold 5. Of these, it is preferable to arrange thermometers on all of the pair of long side cooling plates 5a and the pair of short side cooling plates 5b. Furthermore, thermometers 8 can be arranged in the mold 5 in a multi-stage arrangement with more than three stages in the pouring direction A, or in a single stage.

[0022] Next, the precursory phenomena of breakout will be explained. The breakouts targeted in this proposal can be broadly classified into two types. They are called restrictive breakouts and foreign object entrapment breakouts. First, regarding restrictive breakouts, Fig. 3(a) is a diagram illustrating the state of the molten steel 2 and the solidified shell 10 in the mold 5 in a precursory phenomenon of restrictive breakout. Fig. 3(b) is a diagram showing the state of a fractured portion 11 in the solidified shell 10 in a precursory phenomenon of restrictive breakout.

[0023] As shown in Figures 3(a) and 3(b), in a precursor phenomenon to a restrictive breakout, seizure occurs in the mold 5 for some reason, and the solidified shell 10 is constrained by the mold 5. Meanwhile, because the slab 6 is withdrawn from the bottom end of the mold 5 in the same direction as the pouring direction A shown in Figure 3(b), a fractured portion 11 is formed in the solidified shell 10 directly below the seizure. At this fractured portion 11 of the solidified shell 10, the mold 5 comes into contact with the molten steel 2, causing further seizure. As the above phenomenon is repeated, the fractured portion 11 of the solidified shell 10 moves downward, and the solidified shell 10 above the fractured portion 11 becomes thicker. Finally, when the fractured portion 11 passes the bottom end of the mold 5, the molten steel 2 leaks from the fractured portion 11, causing a restrictive breakout.

[0024] In the fractured portion 11, the molten steel 2 and the mold 5 are in contact with each other, so the temperature of the mold 5 rises locally. Therefore, for example, as shown by the arrow B in FIG. 3(b), the fractured portion 11 moving downward is detected by the thermometer 8. ｍ’，１ ~8 ｍ’，ｎ When passing the placement position of the thermometer 8 ｍ’，１ ~8 ｍ’，ｎ After that, the solidified shell 10 above the rupture portion 11 is restrained by the mold 5 and continues to cool, so the temperature detected by the thermometer 8 ｍ’，１ ~8 ｍ’，ｎ The detected temperature of the solidified shell 10 decreases monotonically. On the other hand, the fractured portion 11 propagates not only downward but also laterally, so that the fractured portion 11 expands in a V-shape as shown in FIG. 3(b). ｍ’，１ ~8 ｍ’，ｎ If the temperature is below 8°C, ｍ’，１ ~8 ｍ’，ｎ Since the passage of the rupture portion 11 does not occur at the position ｍ’，１ ~8 ｍ’，ｎ Only a decrease in the detected temperature is observed.

[0025] Figure 4(a) shows the temperature distribution of the mold 5 at the moment when seizure occurred. Figure 4(b) shows the temperature distribution of the mold 5 10 seconds after the moment when seizure occurred. From the temperature distributions of the mold 5 shown in Figures 4(a) and 4(b), it can be seen that the V-shaped high-temperature portion is propagating downward and laterally.

[0026] Next, we will explain the problem of foreign-object entrapment breakout. FIG. 5 illustrates a foreign-object entrapment breakout. Reference numeral 18 in FIG. 5 denotes the meniscus (molten metal surface) of the molten steel 2. As shown in FIG. 5( a), mold powder 23 is used as a lubricant to prevent the above-mentioned seizure between the mold 5 and the slab 6 during casting. After being supplied to the molten metal surface from the top of the mold 5, the mold powder 23 is heated by the molten steel 2, melts, and flows into the gap between the mold 5 and the slab 6, acting as a lubricant to prevent seizure. However, as shown in FIG. 5( b), the mold powder 23 supplied to the molten metal surface of the molten steel 2 may be drawn into the mold 5 as unmelted, clumped foreign objects 22 (called bare powder). This creates unexpected thermal resistance in the gap between the mold 5 and the slab 6. As a result, the solidified shell 10 cannot grow sufficiently at the location where the foreign object 22 is entrapped, and reaches the bottom of the mold 5, as shown in FIGS. 5( c) and 5( d). In this case, as shown in Figure 5(e), after the foreign object 22 passes the lower end of the mold 5 while the local shell thickness is thin, the solidified shell 10 is unable to withstand the static pressure of the molten steel, causing the solidified shell 10 to break, resulting in the molten steel 2 leaking out. In this case, the temperature distribution in the mold significantly differs from that observed when seizure occurs. First, since the size of the trapped foreign object 22 is considered to be approximately 100 mm or less, the mold temperature changes only by the size of the foreign object 22 as the foreign object 22 passes through the mold 5. In other words, the temperature after the foreign object 22 passes and in the surrounding area where the foreign object 22 has passed remains unchanged from before the occurrence of breakout. Furthermore, since the temperature change is relatively small compared to that caused by seizure, this type of breakout is generally difficult to detect and often results in false positives.

[0027] Therefore, in the breakout prediction method according to the embodiment, in order to accurately detect the two types of breakouts described above, the accuracy of breakout prediction is improved by using the amount of temperature change that is the difference from a certain period of time ago (a first period ago) measured by a thermometer installed inside the mold 5, and by standardizing the amount of temperature change by the standard deviation within the certain period of time (a second period). Below, the breakout prediction method according to the embodiment based on the above technical concept will be described in detail.

[0028] The breakout prediction method according to the embodiment includes a step of inputting dimensions of a slab 6 to be withdrawn from a mold 5 in a continuous casting machine 1. The breakout prediction method according to the embodiment also includes a step of detecting the temperature of the mold 5 using a plurality of thermometers 8 embedded in the mold 5. The breakout prediction method according to the embodiment also includes a step of performing an interpolation process on the temperatures detected by the plurality of thermometers 8 according to the dimensions of the slab 6. The breakout prediction method according to the embodiment also includes a step of calculating a temperature change amount by comparing the detected temperatures with temperatures from a certain period ago. The breakout prediction method according to the embodiment also includes a step of acquiring a standard deviation of the temperature change amount for each thermometer 8 within the certain period. The breakout prediction method according to the embodiment also includes a step of standardizing the temperature change by dividing the temperature change amount by the standard deviation. The breakout prediction method according to the embodiment also includes a step of calculating, based on the temperature change amount calculated by performing the interpolation process, a component in a direction orthogonal to the influence coefficient vector obtained from principal component analysis as a deviation degree from normal operation when no breakout has occurred. The breakout prediction method according to the embodiment also includes a step of predicting a breakout based on the deviation degree.

[0029] 6 is a flowchart showing an example of the procedure of a breakout prediction method according to an embodiment. The breakout prediction method shown in this flowchart is executed by the determination unit 20 shown in FIG. 1. The determination unit 20 has at least the functions of the interpolation processing execution means, temperature change amount standardization means, deviation degree calculation means, and breakout prediction means of the present invention. Details of each step in the flowchart shown in FIG. 6 will be described later as appropriate.

[0030] In the breakout prediction method according to the embodiment, the determination unit 20 determines the temperature of the thermometer 8 during normal operation (hereinafter also referred to as normal operation) when no breakout has occurred. １，１ ~8 ｍ，ｎ A sensitivity coefficient for the standardized temperature change for the temperature is calculated (step S1). Here, this sensitivity coefficient is calculated using a standardized temperature change (standardized temperature change) for commonly processing different casting conditions such as casting speed, steel type, mold powder 23, etc., after interpolation processing to accommodate casting of different widths and thermometer failures, as will be described later. Note that this sensitivity coefficient may change as the surface condition of the mold 5 changes during operation, so it is preferable to update it at an appropriate time, such as between castings. Next, the judgment unit 20 calculates the sensitivity coefficient for the temperature change for the temperature of the thermometer 8. １，１ ~8 ｍ，ｎ The temperature T of the mold 5 is continuously １，１ ~T ｍ，ｎ Next, the determination unit 20 detects the temperature of the thermometer 8 １，１ ~8 ｍ，ｎ The detected temperature is input to a calculation cell 12 equally divided according to the dimensions of the slab 6 to be withdrawn from the mold 5 (for example, the width or thickness of the slab 6), which are input by an operator through an input device (not shown) such as a personal computer or the like provided in the continuous casting machine 1. １，１ ~12 ｋ，ｐ Next, the determination unit 20 performs an interpolation process for the temperature of the mold 5 at the center point of the temperature T′ of the mold 5 obtained by the interpolation process (step S3). １，１ ~T' ｋ，ｐ That is, the determination unit 20 performs average bias removal on the temperature T′ of the mold 5 obtained by the interpolation process.１，１ ~T' ｋ，ｐ In T' １，１ ~ΔT' ｋ，ｐ , the calculation cell 12, which is at the same position as the upper end of the mold 5. １，１ ~12 １，ｐ The temperature change amount ΔT' １，１ ~ΔT' １，ｐ , and computational cell 12 ２，１ ~12 ２，ｐ The temperature change amount ΔT' ２，１ ~ΔT' ２，ｐ , T' ｋ，１ ~T' ｋ，ｐ , the average value and the temperature change ΔT' in each calculation cell １，１ ~ΔT' ｋ，ｐ The standard deviation (of the amount of temperature change) within a certain period is then calculated for the calculation cell 12. １，１ ~12 １，ｐ The temperature change amount ΔT' １，１ ~ΔT' １，ｐ Difference from the average value of 、 and calculation cell 12 ２，１ ~12 ２，ｐ The temperature change amount ΔT' ２，１ ~ΔT' ２，ｐ The difference from the average value of the temperature change values ​​is divided by the standard deviation to obtain the standardized temperature change (steps S4 to S6). Next, the determination unit 20 calculates the deviation using the sensitivity coefficient for the calculated standardized temperature change (step S7).

[0031] One method of determining the sensitivity coefficient vector, which is an influence coefficient vector, is to use principal component analysis. Here, the sensitivity coefficient vector, which is a vector whose components are the sensitivity coefficients, which are influence coefficients, is determined by the temperature of the thermometer 8 during normal operation. １，１ ~8 ｍ，ｎ represents a direction showing the average behavior of the temperature change amount of the calculation cell obtained by the interpolation process. In a vector whose components are the differences from the average value, the component parallel to the direction of the sensitivity coefficient vector is the component of the average behavior, and the component perpendicular to the direction of the sensitivity coefficient vector is the component of the deviation from the average behavior. In addition, the deviation of each calculation cell 12 indicates the component of the difference from the average value for each calculation cell.

[0032] Next, the determination unit 20 determines whether a breakout is predicted (step S8) when the calculated individual deviation of the computational cell 12 (individual deviation) exceeds the threshold Y or when the calculated total deviation of the computational cell 12 (total deviation) exceeds the threshold X. If it is determined that a breakout is not predicted (No in step S8), the determination unit 20 proceeds to step S2.

[0033] On the other hand, if it is determined that a breakout has been predicted (Yes in step S8), the determination unit 20 automatically reduces the casting speed to a predetermined speed (step S9). In this way, by sufficiently reducing the casting speed when the determination unit 20 predicts a breakout, a solidified shell 10 of sufficient thickness is formed in the mold 5 even in areas where seizure or foreign matter has been trapped, and therefore, breakout can be avoided. Thereafter, the determination unit 20 reduces the casting speed to a predetermined value and then returns to the processing routine.

[0034] Next, we will explain the differences between using temperature and temperature change in the sensitivity coefficient used in the breakout prediction method according to the embodiment. In a restrictive breakout, the temperature distribution obtained at the time of occurrence occurs over a very wide range (approximately 100 mm to 500 mm), and the temperature change itself is also very large (20°C to 50°C). Therefore, when calculating the deviation from the sensitivity coefficient, the deviation is large and detection is easy. However, in a foreign object-entrapment breakout, the mold temperature changes only within the range of the size of the foreign object 22 (100 mm or less), and the temperature change itself is very small (20°C or less). Because temperature distribution varies even during normal operation, identifying small temperature changes caused by entrapment as abnormal would allow for false detection, reducing the advantages of this method. In contrast, by calculating the sensitivity coefficient using temperature change and obtaining the deviation, it is possible to obtain a large deviation for sudden temperature changes that do not occur under normal conditions. The breakouts targeted for detection in this study are neither binding breakouts nor foreign object-related breakouts, and abnormal conditions occur suddenly without warning. Therefore, rather than using the temperature as the sensitivity coefficient and expressing the abnormal temperature distribution as the degree of deviation, it is preferable to use the temperature change as the sensitivity coefficient and express the abnormal temperature change as the degree of deviation.

[0035] Next, the temperature change amount used in the calculation is standardized. This standardization method will be explained. When the operating conditions change, for example, when the casting speed increases or decreases, or when the molten steel temperature increases or decreases, the temperature tends to change overall. When predicting a breakout, the overall temperature change caused by a change in operating conditions can become a disturbance and may deteriorate the detection accuracy. Therefore, it is necessary to use the value after excluding the average bias.

[0036] As a method for removing the average bias, for example, １，１ ~8 ｍ，ｎ The detected temperature T １，１ ~T ｍ，ｎ All the average values ​​of T ａｖｅ , and the detected temperature T １，１ ~T ｍ，ｎand the average value T ａｖｅ Another method for removing the average bias is to measure the temperature difference between the temperature of the thermometer 8 and the temperature of the mold 5 at the same distance from the top of the mold 5 in the pouring direction A. ｉ，１ ~8 ｉ，ｎ The detected temperature T ｉ，１ ~T ｉ，ｎ The average value of T ｉ，ａｖｅ , and the detected temperature T ｉ，１ ~T ｉ，ｎ and the average value T ｉ，ａｖｅ A method of calculating the difference between the temperature and the temperature of the thermometer 8 for each of the thermometers 8 located at the same distance can be given.

[0037] In addition, the stability of the mold temperature may change depending on operating conditions such as the oscillation of the casting mold, the casting speed, the type of steel, and the mold powder 23 used. Furthermore, electromagnetic noise may affect the thermometer in use and be interpreted as temperature fluctuations. When predicting breakout based on deviations from temperature fluctuations or when the mold temperature becomes unstable due to certain conditions, there is a high possibility of false detection, and it is necessary to address this. Standardization involves calculating the standard deviation of the temperature change for each thermometer 8 over a certain period and dividing it by the temperature change. This allows for a standardized temperature change that can be used to determine a certain threshold, ignoring differences in temperature change due to differences in processing conditions.

[0038] Another method for determining the sensitivity coefficient vector, which is an influence coefficient vector, is to calculate the sensitivity coefficient vector of each thermometer 8 when the overall temperature changes due to fluctuations in the molten metal surface, for example. １，１ ~8 ｍ，ｎ A method for experimentally determining the ease of temperature transfer of the molten steel 2 in the molten steel 2 can be considered.

[0039] On the other hand, when signs of seizure that lead to a breakout occur, the thermometer 8 ｉ，ｊ１ and 8 ｉ，ｊ２As shown in Fig. 7, the detected temperatures of the thermometer 8 near the position of the fractured portion 11 of the solidified shell 10 are distributed at positions away from the broken line (the line at a 45-degree angle to the right in the example shown in Fig. 7) that indicates the direction of the sensitivity coefficient vector. ｉ，ｊ１ The detected temperature T ｉ，ｊ１ The temperature drops, and a little later the thermometer ｉ，ｊ１ Thermometer 8 located on both sides of ｉ，ｊ１＋１ and thermometer 8 ｉ，ｊ１―１ The detected temperature T ｉ，ｊ１＋１ and the detected temperature T ｉ，ｊ１―１ This is because the

[0040] From the above considerations, thermometer 8 １，１ ~8 ｍ，ｎ The temperature change ΔT １，１ ~ΔT ｍ，ｎ It can be seen that the occurrence of a breakout can be determined by the degree to which the temperature of the thermometer 8 deviates from the dashed line indicating the direction of the sensitivity coefficient vector. １，１ ~8 ｍ，ｎ ΔT calculated from the detected temperature １，１ ~ΔT ｍ，ｎ The component of the temperature change vector, which is a vector with components, perpendicular to the sensitivity coefficient vector is calculated as the deviation. It can be seen that the occurrence of a breakout can be determined based on the calculated deviation.

[0041] For example, in FIGS. 7 and 8, the thermometer 8 ｉ，ｊ１ and thermometer 8 ｉ，ｊ２ In a temperature change vector having the temperature change of the detected temperature as a component, a deviation component is calculated, which is a component in a direction perpendicular to the sensitivity coefficient vector. The occurrence of a breakout is then determined based on this calculated deviation component. Note that in Figures 7 and 8, the direction of the sensitivity coefficient vector is the same as the direction of the first principal component of the temperature change distribution under normal conditions, and the direction perpendicular to the direction of the sensitivity coefficient vector is the same as the direction of the second principal component of the temperature change distribution under normal conditions.

[0042] However, the detected temperature T １，１ ~T ｍ，ｎIf this method itself is used to predict a breakout, there is a risk of false detection. That is, when the pouring width when pouring the molten steel 2 into the mold 5, in other words, the width of the slab 6 withdrawn from the lower end of the mold 5, is changed during operation, there is a risk of falsely predicting that a breakout will occur, even though there are no signs that could lead to a breakout.

[0043] FIG. 9( a ) shows the temperature measurement results of the thermometer 8 in the case where the width (casting width) of the slab 6 removed from the lower end of the mold 5 is wide. ｍ１，ｎ１ ~8 ｍ１，ｎ１＋１８ The detected temperature T ｍ１，ｎ１ ~T ｍ１，ｎ１＋１８ and the temperature T' obtained by performing the interpolation process. ｍ１，ｎ１ ~T' ｍ１，ｎ１＋１８ 9(b) is a diagram showing the relationship between the temperature and the melting point of the thermometer 8 in a case where the width (casting width) of the slab 6 withdrawn from the lower end of the mold 5 is narrow. ｍ１，ｎ１ ~8 ｍ１，ｎ１＋１８ The detected temperature T ｍ１，ｎ１ ~T ｍ１，ｎ１＋１８ and the temperature T' obtained by performing the interpolation process. ｍ１，ｎ１ ~T' ｍ１，ｎ１＋１８ 9(a) and 9(b), the thermometer 8 ｍ１，ｎ１ ~8 ｍ１，ｎ１＋１８ are arranged at the same distance from the upper end of the mold 5 in the pouring direction A. ｍ１，ｎ１ ~T' ｍ１，ｎ１＋１８ is a calculation cell 12 equally divided according to the width of the slab 6. ｍ１，ｎ１ ~12 ｍ１，ｎ１＋１８ At the center point of the thermometer 8 ｍ１，ｎ１ ~8 ｍ１，ｎ１＋１８ The detected temperature T ｍ１，ｎ１ ~T ｍ１，ｎ１＋１８ The estimated temperature of the mold 5 is calculated by performing an interpolation process on the temperature of the mold 5. The method of the interpolation process will be described later.

[0044] When the casting width is changed during casting and the state changes from that shown in FIG. 9(a) to that shown in FIG. 9(b), ｍ１，ｎ１ ~8 ｍ１，ｎ１＋１８ The detected temperature T ｍ１，ｎ１ ~T ｍ１，ｎ１＋１８ When focusing on the detected temperature T ｍ１，ｎ１＋３ and the detected temperature T ｍ１，ｎ１＋１５The temperature change is large only in the case where the temperature is large, and no significant temperature change is observed in the other detected temperatures. Therefore, in the cases shown in FIGS. 9(a) and 9(b), the detected temperature T ｍ１，ｎ１ ~T ｍ１，ｎ１＋１８ If this is used to predict a breakout, there is a risk that it may deviate from the sensitivity coefficient vector and falsely detect the occurrence of a sign that could lead to a breakout.

[0045] On the other hand, when the casting width is changed during casting and the state changes from FIG. 9(a) to FIG. 9(b), the temperature T' obtained by performing the interpolation process while keeping the number of calculation cells 12 (number of cells) constant even when the dimensions of the slab 6 are changed is ｍ１，ｎ１ ~T' ｍ１，ｎ１＋１８ When we focus on the temperature T' ｍ１，ｎ１ ~T' ｍ１，ｎ１＋１８ Therefore, in the cases shown in FIGS. 9A and 9B, the temperature T′ at which the interpolation process is performed is small. ｍ１，ｎ１ ~T' ｍ１，ｎ１＋１８ By using this to predict breakouts, the risk of false positives of the occurrence of signs leading to a breakout can be reduced.

[0046] In addition, in FIGS. 9(a) and 9(b), the detected temperature T ｍ１，ｎ１＋７ , detected temperature T ｍ１，ｎ１＋１１ , detected temperature T ｍ１，ｎ１＋１２ , and the detected temperature T ｍ１，ｎ１＋１６ and a thermometer 8 that detects the temperature. ｍ１，ｎ１＋７ , thermometer 8 ｍ１，ｎ１＋１１ , thermometer 8 ｍ１，ｎ１＋１２ , and thermometer 8 ｍ１，ｎ１＋１６ Even when the thermometer 8 having such a temperature detection defect is included, the detected temperature T ｍ１，ｎ１ ~T ｍ１，ｎ１＋１８ If the temperature T′ obtained by the interpolation process is used to predict a breakout, it may deviate from the sensitivity coefficient vector and be erroneously detected as a sign of a breakout. ｍ１，ｎ１ ~T' ｍ１，ｎ１＋１８ In this case, even if the temperature detection section includes a thermometer 8 with poor temperature detection, the risk of false detection of the occurrence of signs that could lead to a breakout can be reduced by using the estimated temperature of the mold 5 in the section with poor temperature detection.

[0047] Next, the interpolation process will be described. ｉ，１ ~8 ｉ，ｊ and calculation cell 12 ｉ，１ ~12 ｉ，ｊ 1 is a diagram showing the positional relationship of

[0048] As shown in FIG. ｉ，１ ~12 ｉ，ｊ is a thermometer 8 located at the same distance from the top end of the mold 5 on the long side cooling plate 5a of the mold 5. ｉ，１ ~8 ｉ，ｊ On the other hand, the section of the long side cooling plate 5a corresponding to the width of the slab 6 (the section sandwiched between the pair of short side cooling plates 5b in the width direction of the mold 5) is equally divided into a certain number of cells. ｉ，１ ~8 ｉ，ｊ The detected temperatures are linearly interpolated to obtain the calculation cell 12. ｉ，１ ~12 ｉ，ｊ The estimated temperatures of the mold 5 (long side cooling plates 5 a) at the positions of the respective center points are calculated. The number of calculation cells 12 for the interpolation process may be the same as or different from the number of thermometers 8 in the vertical and horizontal directions, but is kept constant regardless of fluctuations in the casting width during casting.

[0049] The above-described interpolation process can be applied when determining sensitivity coefficient vectors and calculating deviations using principal component analysis. In this case, the principal component analysis is performed using interpolated temperatures instead of actual detected temperatures. Even when the slab width is changed, the same number of temperature vectors can be used, allowing principal component analysis to be performed including data on different widths. This eliminates the need to determine different influence coefficients for each width, and allows influence coefficient vectors to be determined that include data on different slab widths. Furthermore, deviations can be calculated using influence coefficient vectors calculated based on temperatures obtained by interpolating detected temperatures. Therefore, breakout prediction for different slab widths is possible based on a unified standard. Furthermore, the risk of false detection of signs leading to breakout can be reduced even when the slab width changes during casting.

[0050] The amount of temperature change is calculated from the difference between the temperature detected a certain time ago and the current value. For simplicity, the difference between the temperatures detected by two thermometers at the same location in the width direction but in different casting directions (for example, 8 in the mold thermometer layout diagram in Figure 2) is used. １，２ and 8 ２，２ It can also be used as a substitute for the difference between

[0051] Next, the determination of breakout prediction will be explained. Fig. 11(a) is a diagram showing the time series change in the total deviation in a case where burn-in (restrictive breakout) was confirmed. Fig. 11(b) is a diagram showing the change in the individual deviation at the position of each computational cell 12 in a case where burn-in (restrictive breakout) was confirmed.

[0052] 12(a) is a graph showing the time series change in the total deviation in a case where foreign matter entrapment (foreign matter entrapment breakout) was confirmed, and FIG. 12(b) is a graph showing the change in the individual deviation at the position of each calculation cell 12 in a case where foreign matter entrapment (foreign matter entrapment breakout) was confirmed.

[0053] 13(a) shows a case where the total deviation has a peak, but the slab inspection results show no abnormalities. FIG. 13(b) shows the individual deviations for each calculation cell 12 in a case where the slab inspection results show no abnormalities. Note that FIGS. 13(a) and 13(b) are considered to be cases where noise was introduced into the measurement value of the thermometer 8 due to electromagnetic noise generated near the thermometer.

[0054] The two graphs shown in Figures 11(a), 12(a), and 13(a) are obtained by calculating the total deviation for each of the upper and lower parts of the mold 5. In Figures 11, 12, and 13, the abnormality threshold is 150 for the total deviation and 20 for the individual deviation, and is indicated by the dashed lines in each figure.

[0055] In Figure 11(a), the deviation increases sharply at a certain time during operation. As mentioned above, burn-in causes temperature fluctuations over a relatively wide range, and the temperature fluctuation width is large, so it can be seen that the total deviation has a value that is clearly different from that under normal conditions. Also, as shown in Figure 11(b), it can be seen that within each calculation cell 12, some calculation cells 12 show large deviation values.

[0056] 12(a), the total deviation when a foreign object is detected shows a smaller peak than that shown in FIG. 11(a). Also, in FIG. 12(b), it can be seen that calculation cell 12 has only two peaks in the deviation.

[0057] 13(a), the total deviation has a peak. On the other hand, as shown in FIG. 13(b), the individual deviations in each calculation cell 12 are all low. Since almost all thermometers 8 are experiencing temperature changes at the same time, it is estimated that this condition is caused by ambient electromagnetic noise generated near the thermometers 8, and is not an actual temperature change.

[0058] Here, in order to detect seizure and foreign object entrapment, which are signs of breakouts, and to avoid detecting cases in which there is no abnormality, a breakout is predicted based on the following two conditions. The first condition is that an abnormality is detected when the total deviation exceeds a preset threshold X (second threshold). Even in this case, at least one point must be equal to or greater than a preset threshold Y. The second condition is whether or not there is at least one point of deviation that has a value equal to or greater than a preset threshold Z (first threshold).

[0059] FIG. 14 shows the results of breakout prediction using the breakout prediction method according to the embodiment and a conventional method. As shown in FIG. 14 , the conventional methods (conventional examples) based on Patent Documents 1 and 2 predicted breakouts based on the behavior of temperature sensors embedded inside the mold. However, they only detected breakouts with a detection rate of about 50%, resulting in many false positives. On the other hand, as shown in FIG. 14 , the breakout prediction method according to the embodiment (inventive example) significantly reduced the false positive rate. Furthermore, the breakout prediction method according to the embodiment (inventive example) was able to detect abnormalities based on the deviation degree, even if the conventional method (conventional example) did not detect them. Furthermore, the breakout prediction method according to the embodiment (inventive example) often found abnormalities in actual cast slabs that did not result in breakout, making it possible to predict breakouts that the conventional method (comparative example) could not detect.

[0060] The present invention can provide a breakout prediction method and a continuous casting machine operation method that can detect foreign object entrapment breakouts with sufficient detection accuracy, even when the range of change or temperature fluctuation in the mold temperature is small.

[0061] REFERENCE SIGNS LIST 1 continuous casting machine 2 molten steel 3 tundish 4 submerged nozzle 5 mold 6 slab 7 slab support roll 8 thermometer 10 solidified shell 11 fractured portion 12 calculation cell 18 meniscus 20 judgment portion 22 foreign matter 23 mold powder

Claims

1. A method for predicting a breakout comprising the steps of: inputting dimensions of a cast piece to be pulled out of a mold in a continuous casting machine; detecting the temperature of the mold using a plurality of thermometers embedded in the mold; performing an interpolation process on the temperatures detected by the plurality of thermometers according to the dimensions of the cast piece; calculating an amount of temperature change by comparing with the temperature before a first period; obtaining a standard deviation of the amount of temperature change for each thermometer during a second period; standardizing the temperature change by dividing the amount of temperature change by the standard deviation; calculating a component in a direction perpendicular to an influence coefficient vector obtained from principal component analysis based on the amount of temperature change calculated by performing the interpolation process as a deviation from normal operation when no breakout has occurred; and predicting the breakout based on the deviation.

2. The breakout prediction method according to claim 1, characterized in that in the step of performing the interpolation process, an interpolation process is performed at the center point of each of a plurality of calculation cells equally divided according to the dimensions of the cast piece to calculate a temperature for each detected temperature of the plurality of thermometers.

3. The method for predicting a breakout according to claim 2, characterized in that the number of said calculation cells is kept constant even if the dimensions of said slab are changed.

4. The method for predicting a breakout as described in claim 2, characterized in that in the step of calculating the deviation, an average value of the temperatures of the plurality of calculation cells located at the same distance from the top end of the mold in the pouring direction of molten steel into the mold is calculated, a difference between the average value and the temperatures of the plurality of calculation cells is calculated, and the deviation is calculated from the obtained difference using the influence coefficient vector.

5. The method for predicting a breakout as described in claim 3, characterized in that in the step of calculating the deviation, an average value of the temperatures of the plurality of calculation cells located at the same distance from the top end of the mold in the pouring direction of the molten steel into the mold is calculated, a difference between the average value and the temperatures of the plurality of calculation cells is calculated, and the deviation is calculated from the obtained difference using the influence coefficient vector.

6. The breakout prediction method according to claim 4, characterized in that in the step of predicting a breakout, the breakout is predicted when the calculated individual deviation of the computational cell exceeds a predetermined first threshold value, or when the calculated total deviation of the computational cells exceeds a predetermined second threshold value.

7. The breakout prediction method described in claim 5, characterized in that in the step of predicting a breakout, the breakout is predicted when the calculated individual deviation of the calculation cell exceeds a predetermined first threshold value, or when the calculated total deviation of the calculation cells exceeds a predetermined second threshold value.

8. A breakout prediction method according to any one of claims 1 to 7, characterized in that the influence coefficient vector is a sensitivity coefficient vector having as its components the sensitivity coefficients of each of the plurality of temperature change amounts.

9. A method for operating a continuous casting machine, comprising the steps of: reducing the casting speed when a breakout is predicted based on the breakout prediction method according to any one of claims 1 to 7.

10. A method for operating a continuous casting machine, comprising the steps of: reducing the casting speed when a breakout is predicted based on the breakout prediction method according to claim 8.

Images