Breakout discrimination method of continuous casting apparatus

Abstract

To provide a breakout discrimination method of a continuous casting apparatus which satisfactorily discriminates breakout caused by a corner part occupying most of breakout, and surely prevents the breakout.SOLUTION: A breakout discrimination method arranges thermocouples 4 at a plurality of positions including two positions #23 and #20 or #15 and #13 sandwiching a corner part on the outer peripheral surface of a mold 3 of a continuous casting apparatus, acquires a temperature signal of each of the thermocouples 4 in casting in a constant cycle, converts the temperature signal into a temperature data sequence Ta, then calculates a difference between one temperature data and temperature data immediately before the one temperature data and obtains a difference data sequence Tad, converts a data position when a difference value of the difference data sequence Tad exceeds a predetermined threshold into an index idx and gives the index idx to the thermocouples 4, and discriminates occurred breakout from breakout caused by the corner part, in a case in which the thermocouples 4 at a head position and a subsequent position are installed in the two positions #23 and #20 or #15 and #13 when the thermocouples 4 are arrayed in ascending order of the index idx.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a method for distinguishing breakouts in a continuous casting machine, and more particularly to a method for distinguishing breakouts caused by corners, which account for the majority of breakouts, from other breakouts. [Background technology]

[0002] If a breakout occurs in a continuous casting device, casting becomes impossible and recovery is costly. Therefore, methods for predicting breakout have been devised, and for example, the method described in Patent Document 1 (conventional method) has been proposed.

[0003] In the conventional method, a large number of thermocouples are embedded in the mold of a continuous casting machine to collect temperature transition data when a breakout occurs and when it does not occur. A temperature change pattern detection network and a breakout detection network, both implemented as neural networks, are trained using this temperature transition data, and the outputs of the temperature change pattern detection network and peak signals obtained by processing the output signals from the thermocouples are input to the breakout detection network to predict a breakout. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 7-178524 Summary of the Invention [Problem to be solved by the invention]

[0005] The breakouts mentioned above can be classified into three types: corner-induced breakouts, in which the solidified shell (solid phase) around the molten steel at the corner of the water-cooled mold solidifies and shrinks, causing it to break, allowing the molten steel (liquid phase) to flow out; deckle-induced breakouts, in which deckles (precipitates) adhere to the solidified shell around the nozzle of the mold and break, causing the molten steel to flow out; and nozzle-clogging breakouts, in which one of the nozzle discharge holes becomes blocked, causing the molten steel to concentrate in the remaining discharge holes, causing the solidified shell to break and allowing the molten steel to flow out. According to the inventor's knowledge, of these breakouts, corner-induced breakouts account for the majority (approximately 85%).

[0006] In this case, specific measures to prevent breakouts differ depending on the type of breakout, so it is necessary to accurately distinguish and predict those caused by corners, which account for the majority of breakouts, and to reliably prevent their occurrence. However, the above-mentioned conventional prediction methods have the problem of being unable to distinguish and predict breakouts.

[0007] The present invention is devised to solve such problems, and has an object to provide a breakout discrimination method for a continuous casting machine that can effectively discriminate breakouts caused by corner portions, which account for the majority of breakouts, and reliably prevent such breakouts. [Means for solving the problem]

[0008] In order to achieve the above object, the first invention is characterized in that temperature detection means are arranged at multiple positions on the outer peripheral surface of the mold of a continuous casting machine, including two positions on either side of a corner, and a temperature differential value is calculated for each of the temperature detection means during casting.If the temperature detection means arranged at the two positions have the earliest and subsequent times at which the temperature differential value exceeds a predetermined threshold, the breakout that has occurred is distinguished as being caused by the corner.

[0009] In the second invention, temperature detection means (4) are arranged at multiple positions on the outer surface of the mold (3) of a continuous casting device, including two positions (#23, #20 or #15, #13) on either side of a corner, and temperature signals are acquired from each of the temperature detection means 4 at regular intervals during casting to create a temperature data sequence (Ta).The difference between one temperature data and the temperature data immediately preceding it is then calculated to obtain a differential data sequence (Tad).The data position at which the differential value of the differential data sequence (Tad) exceeds a predetermined threshold is assigned to the temperature detection means (4) as an index (idx).When the temperature detection means 4 are arranged in ascending order of the index (idx), if the temperature detection means 4 at the leading position and the temperature detection means 4 at the following position are installed at the two positions (#23, #20 or #15, #13), the breakout that occurs is distinguished as being caused by the corner.

[0010] The symbols in parentheses above indicate, for reference, the correspondence with specific means described in the embodiments to be described later. [Effects of the Invention]

[0011] According to the breakout discrimination method for a continuous casting machine of the present invention, it is possible to effectively discriminate breakouts caused by corner portions, which account for the majority of breakouts. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a schematic diagram showing a configuration of a continuous casting device. [Figure 2] FIG. 2 is a schematic perspective view of a mold. [Figure 3] FIG. [Figure 4] 10 is a flowchart showing the procedure of a discrimination method. [Figure 5] 10 is a graph showing data strings of temperature and difference when no breakout occurs. [Figure 6] 10 is a graph showing data strings of temperature and difference when a breakout occurs. [Figure 7]10 is a graph showing data strings of temperature and difference when noise is present; DETAILED DESCRIPTION OF THE INVENTION

[0013] The embodiments described below are merely examples, and various design improvements made by those skilled in the art without departing from the gist of the present invention are also included in the scope of the present invention.

[0014] Figure 1 shows a schematic diagram of a continuous casting machine in which breakout (BO) is to be detected. In Figure 1, molten steel Sm is supplied from a ladle 1 to a tundish 2, and from the tundish 2, through a mold 3, a slab St with unsolidified molten steel Sm surrounded by a thin solidified shell is drawn between a number of support rolls 31.

[0015] Figure 2 shows an overview of the mold 3. In Figure 2, a number of thermocouples 4 serving as temperature detection means are installed on the outer periphery of the side wall of the square-shaped mold 3, and temperature changes in the pouring direction (indicated by the arrow in the figure) are measured over the entire circumference of the mold 3.

[0016] The arrangement of thermocouples 4 on each of the four outer periphery surfaces of the side wall of the mold 3 is as shown in Fig. 3, and in this embodiment, eight thermocouples are provided in three upper and lower rows on each of the opposing surfaces 3a and 3b (see Fig. 2), and six thermocouples are provided in two upper and lower rows on each of the other opposing surfaces 3c and 3d, for a total of 28 thermocouples 4 (numbered #1 to #28). The temperature signal 4a of each thermocouple 4 is input to a computer 5 for carrying out the discrimination method of the present invention, which will be described below.

[0017] In this embodiment, the computer 5 obtains temperature data by sampling the temperature signal from each thermocouple 4 at a frequency of 2 Hz. The computer performs processing on each piece of temperature data as shown in the flowchart of FIG.

[0018] 4, temperature data of each thermocouple is acquired, and initial values ​​are set in step 102. The initial values ​​are the setting of a window size w (w=5 in this embodiment) for moving average processing, which will be described later, and the setting of a threshold value th for difference data after difference processing, which will be described later, (th=0.15 in this embodiment).

[0019] In step 103, padding is performed by adding an appropriate number of data equal to the temperature data at the front and rear ends in the time direction to enable moving average processing of the temperature data at the front and rear ends. Then, in step 104, moving average processing of the temperature data is performed with a window size w to remove noise.

[0020] Here, Figure 5(1) shows an example of a temperature data sequence Tn after moving average processing when no breakout occurs, and Figure 6(1) shows an example of a temperature data sequence Ta after moving average processing when a breakout occurs.

[0021] In step 105 of Fig. 4, the temperature data sequences Tn and Ta are converted into differential data sequences Tnd and Tad by differential processing that takes the difference between the preceding and following temperature data. Examples of the differential data sequences Tnd and Tad are shown in Fig. 5(2) and Fig. 6(2), respectively. In step 106 of Fig. 4, it is checked whether the differential data sequences Tnd and Tad contain values ​​that exceed the threshold value th. If not, the index idx is set to 0 (step 107). If present, the data position at which the threshold value th was first exceeded is set to index idx.

[0022] 5 and 6, if no breakout occurs (FIG. 5), there is no value in the differential data string Tnd that exceeds the threshold value th, so the thermocouple 4 that measured that temperature is assigned an index of 0. On the other hand, if a breakout occurs (FIG. 6), there is a value in the differential data string Tad that exceeds the threshold value th, so the data position when the threshold value th was first exceeded (260 in this embodiment) is assigned as the index idx of the thermocouple 4 that measured that temperature.

[0023] After the above process is performed on 28 thermocouples 4 (#1 to #28), a list is created in which they are arranged in ascending order of the index idx assigned to each thermocouple 4 (excluding those with an index of 0). An example of this list is shown in Table 1. Table 1 shows the numbers of the first six thermocouples 4 on the list, and according to this, when a breakout due to a corner occurs, the thermocouples 4 at two positions on either side of the corner (angle) of the mold 3, #23 and #20 or #15 and #13 (see Figure 3), appear at the beginning of the list and in the following positions (within the dashed lines in Table 1), as shown in (1) and (2) of Table 1.

[0024] [Table 1]

[0025] This indicates that the thermocouple 4 with the smallest index idx value and the thermocouple 4 with the next smallest index idx value, i.e., the thermocouple 4 whose difference data value exceeded the threshold value th earliest and the thermocouple 4 with the next smallest index idx value, were the thermocouples 4 numbered #23, #20 or #15, #13.

[0026] On the other hand, if a breakout other than that caused by the corner occurs, as shown in Table 1 (3), thermocouples 4 numbered #14 and #11, other than the two positions sandwiching the corner of mold 3, will appear at the beginning and subsequent positions of the list.

[0027] In this way, if two thermocouples 4 appear at two positions on either side of a corner of the mold 3 at the first and subsequent positions in the list in which the thermocouples 4 are arranged in ascending order of the index idx value, it can be determined that a breakout has occurred due to the corner.

[0028] As shown in FIG. 7, even when no breakout has occurred, the value of the difference data Tnd of the temperature data Tn may exceed the threshold value th due to noise or the like. In this case, however, the value of the index idx of thermocouple 4 is large, so it does not appear at the beginning of the list. Furthermore, the probability that the thermocouples 4 at two positions on either side of a corner will appear at the beginning of the list is sufficiently low, so no erroneous determination will occur.

[0029] In this way, it is possible to distinguish breakouts caused by corners. By assigning a classification of corner-caused breakouts to the temperature changes of each thermocouple at this time and using this as learning data for the neural network, it becomes possible to accurately predict the occurrence of corner-caused breakouts. [Explanation of symbols]

[0030] 1... ladle, 2... tundish, 3... mold, 4... thermocouple (temperature detection means), 5... computer.

Claims

1. A breakout discrimination method for a continuous casting machine, characterized in that temperature detection means are arranged at multiple positions on the outer peripheral surface of the mold of a continuous casting machine, including two positions on either side of a corner, and a temperature differential value is calculated for each of the temperature detection means during casting, and if the temperature detection means at the two positions at which the temperature differential value exceeds a predetermined threshold value earliest and the following time are those arranged, the breakout that has occurred is discriminated as being caused by the corner.

2. a temperature detection means for detecting a breakout in a continuous casting machine, the temperature detection means being arranged at a plurality of positions on the outer peripheral surface of a mold in a continuous casting machine, including two positions on either side of a corner; temperature signals from each of the temperature detection means being acquired at regular intervals during casting to create a temperature data string; a difference between one piece of temperature data and the temperature data immediately preceding it being calculated to create a differential data string; a data position at which the differential value of the differential data string exceeds a predetermined threshold being assigned to the temperature detection means as an index; and when the temperature detection means are arranged in ascending order of the index, if the temperature detection means at the leading position and the temperature detection means at the following position are installed at the two positions, the breakout that has occurred is identified as being caused by the corner.

Images