CN116441502A - Method and system for identifying longitudinal crack defects of continuous casting slab based on crystallizer temperature - Google Patents
Method and system for identifying longitudinal crack defects of continuous casting slab based on crystallizer temperature Download PDFInfo
- Publication number
- CN116441502A CN116441502A CN202210015464.4A CN202210015464A CN116441502A CN 116441502 A CN116441502 A CN 116441502A CN 202210015464 A CN202210015464 A CN 202210015464A CN 116441502 A CN116441502 A CN 116441502A
- Authority
- CN
- China
- Prior art keywords
- continuous casting
- longitudinal crack
- crystallizer
- slab
- heat removal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000009749 continuous casting Methods 0.000 title claims abstract description 86
- 230000007547 defect Effects 0.000 title claims abstract description 67
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000001514 detection method Methods 0.000 claims abstract description 74
- 230000002159 abnormal effect Effects 0.000 claims abstract description 45
- 230000005856 abnormality Effects 0.000 claims abstract description 19
- 229910000831 Steel Inorganic materials 0.000 claims abstract description 9
- 239000010959 steel Substances 0.000 claims abstract description 9
- 239000007788 liquid Substances 0.000 claims abstract description 7
- 238000005266 casting Methods 0.000 description 11
- 238000010801 machine learning Methods 0.000 description 8
- 238000004519 manufacturing process Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 238000007689 inspection Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 230000002950 deficient Effects 0.000 description 4
- 238000001816 cooling Methods 0.000 description 3
- 230000004907 flux Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000005096 rolling process Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000005499 meniscus Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 238000005336 cracking Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 239000002893 slag Substances 0.000 description 1
- 230000008646 thermal stress Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B22—CASTING; POWDER METALLURGY
- B22D—CASTING OF METALS; CASTING OF OTHER SUBSTANCES BY THE SAME PROCESSES OR DEVICES
- B22D11/00—Continuous casting of metals, i.e. casting in indefinite lengths
- B22D11/16—Controlling or regulating processes or operations
- B22D11/18—Controlling or regulating processes or operations for pouring
- B22D11/181—Controlling or regulating processes or operations for pouring responsive to molten metal level or slag level
- B22D11/182—Controlling or regulating processes or operations for pouring responsive to molten metal level or slag level by measuring temperature
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B22—CASTING; POWDER METALLURGY
- B22D—CASTING OF METALS; CASTING OF OTHER SUBSTANCES BY THE SAME PROCESSES OR DEVICES
- B22D2/00—Arrangement of indicating or measuring devices, e.g. for temperature or viscosity of the fused mass
- B22D2/006—Arrangement of indicating or measuring devices, e.g. for temperature or viscosity of the fused mass for the temperature of the molten metal
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Continuous Casting (AREA)
Abstract
The invention discloses a method for identifying longitudinal crack defects of a continuous casting slab based on crystallizer temperature, which comprises the following steps: arranging a plurality of heat removal couples below the liquid level of the molten steel of the crystallizer along the continuous casting drawing speed direction, wherein each heat removal couple comprises a plurality of thermocouples; collecting temperature data of thermocouples of m time point windows in real time; calculating to obtain the temperature data average value of each heat removal couple in each time point window, wherein each heat removal couple obtains m temperature data average values; abnormality detection is carried out on the average value of m temperature data of each thermocouple by adopting an abnormality detection algorithm, and the number n of abnormality detection points of each thermocouple is obtained i The method comprises the steps of carrying out a first treatment on the surface of the Comparing the number of the abnormal detection points of each heat removal thermocouple along the continuous casting drawing speed direction, if the number of the abnormal detection points of the next row is not less than that of the abnormal detection points of the previous rowAnd if the number of the measuring points is equal to the number, judging that the slab generates the longitudinal crack defect, otherwise, judging that the slab is normal.
Description
Technical Field
The invention relates to an online slab quality prediction method, in particular to an online slab longitudinal crack prediction method.
Background
As is well known, the continuous casting billet hot charging and hot feeding and continuous rolling technology can greatly reduce equipment investment and production cost and improve product competitiveness, and the continuous casting billet hot charging and hot feeding technology requires a defect-free casting billet to be produced on a production line, namely, the surface quality and the internal quality of the casting billet are required to be basically cleaned, so that the requirement of direct rolling can be met.
However, the current continuous casting technology cannot thoroughly eliminate the generation of defective casting blanks, so that detection and judgment of the defective casting blanks are often required, and the quality of the casting blanks is timely and online forecasted and detected in production so as to timely sort the defective casting blanks off line or take measures. The method for detecting and judging the slab defects is a key for ensuring the production technology of the non-defective casting blanks, and has important significance for ensuring the production continuity, improving the product quality and reducing the production cost.
At present, in the existing continuous casting billet quality inspection method, the conventional cold continuous casting billet quality inspection method is difficult to meet the requirement, and the hot online inspection method is a necessary means for controlling the quality of the casting billet. The thermal state online inspection method can comprise two different detection means, namely online detection judgment based on physical means and continuous casting billet quality prediction judgment based on a model.
However, it should be noted that, since the continuous casting process is a high-temperature, rapid and totally enclosed production, in the hot online inspection method, the online detection and determination method (infrared method, eddy current method and flaw detection method) based on physical means cannot meet the requirements of continuous casting and rolling; the method needs to carry out on-line detection and judgment by means of a defect detection device working under a high-temperature severe environment state, has complex technology and expensive equipment, needs a great amount of daily maintenance work and mainly detects the surface quality of a casting blank.
Therefore, in general, the thermal state on-line inspection of the continuous casting slab can only rely on model-based continuous casting slab quality prediction determination systems.
Among the surface cracks of the continuous casting slab, the longitudinal crack is the most common crack defect, and the surface longitudinal crack generally occurs in the middle of the wide face of the continuous casting slab and is parallel to the casting direction. For slab crystallizers, a uniform heat flux density should be maintained over the entire width of the crystallizer, especially in the meniscus region, otherwise uneven heat transfer would cause very thin shells to irregularly separate from the walls of the crystallizer, and the heat recovery in the separation region would increase the thermal stress and could lead to surface longitudinal cracking of the continuous cast slab.
In order to predict the longitudinal cracks of the surface of the continuous casting slab, some researchers have conducted intensive studies and have selected to use monitoring the temperature change of the mold to predict the surface longitudinal cracks. However, the above prediction method still has a disadvantage that in the actual production process, factors affecting the temperature change of the crystallizer are not only split, and the performance of the mold flux, the uniformity of the inflow distribution of the mold flux, other defects on the surface of the casting blank and the like may affect the temperature value of the embedded thermocouple, but the split cannot be separated from other factors causing the temperature deviation in the same column only by generating the temperature deviation in the same column, so that false alarm of the split defect is caused, and the prediction accuracy is lower.
Thus, in order to overcome such a defect, chinese patent document with publication number CN1428216a, publication date of 2003 of 7/9, entitled "a continuous casting slab longitudinal crack prediction method", discloses a continuous casting slab longitudinal crack prediction method which requires not only that the temperatures of two adjacent thermocouples in the same column start to drop, but also that the continuous falling time of the next column is not less than the continuous falling time of the last column, i.e., that the temperature falling time of the next column is not less than the continuous falling time of the last column, meaning that cracks propagate in the longitudinal direction, thereby recognizing as longitudinal cracks.
In theory, compared with other patents, the technical scheme of the patent can more accurately distinguish the longitudinal crack from other factors causing temperature reduction, and false alarm is reduced. However, this patent has two disadvantages: (1) The fluctuation of the pull speed is also one of reasons for the abnormality of the temperature under the condition that the pull speed is stable; (2) The temperature drop rate is specified to be more than 3 ℃/s, and the temperature drop rate is stable to be the longitudinal crack threshold value according to different steel grades and different cooling conditions. Both the above factors easily lead to missing report of longitudinal cracks when an operator adopts the continuous casting slab longitudinal crack prediction method.
In view of the defects and shortcomings in the prior art, the inventor has conducted intensive studies to expect a new and easy-to-implement method and system for identifying longitudinal crack defects of continuous casting slabs.
In recent years, with rapid development of information technology, particularly development of big data and AI technology, and the like, successful application in industries such as the internet and medical treatment has been attempted to be industrially applied. At present, K-nearest neighbor, SVM and isolated forest are common machine learning anomaly detection algorithms, so that the algorithms can be adopted to detect abnormal points (compared with other points) in real time so as to acquire the anomaly characteristics of the characteristics to be detected. The machine learning algorithms are based on the mature frames of the python language, the environment of the frame is built, and the abnormal characteristics can be quickly identified by calling the abnormal detection algorithms, so that the machine learning algorithm is very convenient and quick.
Based on the above, the invention is expected to obtain a new and easy-to-implement method and system for identifying the longitudinal crack defects of the continuous casting slab, which can acquire the abnormal temperature characteristics of the crystallizer in real time based on a machine learning abnormal algorithm and identify the longitudinal crack of the slab according to the characteristics of the abnormal characteristics.
The method and the system can fully consider the uniqueness of the longitudinal cracks of the continuous casting slab, can effectively avoid false alarm and missing alarm, improve the accuracy of longitudinal crack prediction, are simple in calculation, are easy to realize and transplant and popularize, and have very good popularization prospect and application value.
Disclosure of Invention
The invention aims to provide a method for identifying the longitudinal crack defect of a continuous casting slab based on the temperature of a crystallizer, which fully considers the uniqueness of the longitudinal crack of the continuous casting slab, can effectively avoid false alarm and missing alarm, improves the accuracy of longitudinal crack prediction, has simple calculation, is easy to realize and transplant and popularize, and has very good popularization prospect and application value.
In order to achieve the above object, the present invention provides a method for identifying a longitudinal crack defect of a continuous casting slab based on a crystallizer temperature, comprising the steps of:
arranging a plurality of heat removal couples below the liquid level of the molten steel of the crystallizer along the continuous casting drawing speed direction, wherein each heat removal couple comprises a plurality of thermocouples;
collecting temperature data of thermocouples of m time point windows in real time;
calculating to obtain the temperature data average value of each heat removal couple in each time point window, wherein each heat removal couple obtains m temperature data average values;
abnormality detection is carried out on the average value of m temperature data of each thermocouple by adopting an abnormality detection algorithm, and the number n of abnormality detection points of each thermocouple is obtained i ;
And comparing the number of the abnormal detection points of each heat removal thermocouple along the continuous casting drawing speed direction, if the number of the abnormal detection points of the next row is not less than the number of the abnormal detection points of the previous row, judging that the slab generates the longitudinal crack defect, otherwise, judging that the slab is normal.
In the invention, the inventor finds that after a longitudinal crack source is formed on a meniscus primary green shell, an air gap is formed between a slag film and the green shell, and the air gap can cause the increase of interface thermal resistance, so that the abnormal temperature reduction rate of a corresponding region of a crystallizer is generated; as the draw is downward, the gap created by the longitudinal split is located in the shell and therefore moves downward gradually as the draw progresses, which necessarily affects the crystallizer area being traversed. Therefore, detecting the abnormal temperature of at least two rows of thermocouples distributed in the continuous casting drawing speed direction below the liquid level of the molten steel in the crystallizer is one of the necessary conditions for judging whether the longitudinal crack source is generated.
On the other hand, it should be noted that as time goes on, the longitudinal crack will propagate in the direction of pulling the blank, otherwise it will disappear, so the abnormal time of the next heat release thermocouple temperature in time sequence should not be smaller than the abnormal time of the last heat release thermocouple temperature in time sequence, which is the second necessary condition for the occurrence of the longitudinal crack and is an important feature that the thermal resistance is increased due to the distinction of the longitudinal crack from other factors.
It should be noted that, the "abnormal time of the next heat removal thermocouple temperature in time sequence should be not less than the abnormal time of the previous heat removal thermocouple temperature in time sequence" may be equal to that in the above technical scheme: the number of the next row of the abnormal detection points is not smaller than the number of the previous row of the abnormal detection points.
Therefore, the continuous casting slab longitudinal crack judgment method based on the two points can carry out longitudinal crack judgment on the continuous casting slab, and has the advantage of high prediction accuracy. After the system implementing the method generates longitudinal crack prediction and feeds back to an L1 system for collecting equipment information such as instruments, counter measures such as reducing the pulling speed and the specific water quantity of secondary cooling can be timely adopted in the subsequent process, so that the expansion of longitudinal cracks of the continuous casting slab is slowed down, and the aim of reducing the longitudinal crack defects of the continuous casting slab is fulfilled.
Further, in the method for identifying the longitudinal crack defect of the continuous casting slab based on the temperature of the crystallizer, the method further comprises the following steps: and outputting an alarm signal when the slab is considered to generate a longitudinal crack defect.
Further, in the method for identifying the longitudinal crack defect of the continuous casting slab based on the temperature of the crystallizer, the anomaly detection algorithm is a K-nearest neighbor algorithm.
Further, in the method for identifying the longitudinal crack defect of the continuous casting slab based on the temperature of the crystallizer, the anomaly detection algorithm is an SVM algorithm.
Further, in the method for identifying the longitudinal crack defect of the continuous casting slab based on the temperature of the crystallizer, the anomaly detection algorithm is an isolated forest algorithm.
In the method for identifying the longitudinal crack defect of the continuous casting slab based on the temperature of the crystallizer, the anomaly detection algorithm can be a K-nearest neighbor algorithm, an SVM algorithm or an isolated forest algorithm. These algorithms belong to the anomaly detection algorithms commonly used in the prior art, and are not described in detail here.
Accordingly, another object of the present invention is to provide a system for identifying a longitudinal crack defect of a continuous casting slab based on a crystallizer temperature, which can be used for implementing the method of the present invention, and has wide applicability and very wide application prospects.
In order to achieve the above object, the present invention provides a system for identifying a longitudinal crack defect of a continuous casting slab based on a crystallizer temperature, comprising the steps of:
the plurality of heat removal thermocouples are arranged below the liquid level position of the molten steel of the crystallizer and are arranged along the continuous casting drawing speed direction, wherein each heat removal thermocouple comprises a plurality of thermocouples;
the data acquisition device is used for actually collecting temperature data of each thermocouple of m time point windows;
a control device arranged to perform the steps of:
calculating and obtaining a temperature data average value of each heat removal couple in each time point window based on the temperature data of each thermocouple in the m time point windows, so that each heat removal couple obtains m temperature data average values;
abnormality detection is carried out on the average value of m temperature data of each thermocouple by adopting an abnormality detection algorithm, and the number n of abnormality detection points of each thermocouple is obtained i
And comparing the number of the abnormal detection points of each heat removal thermocouple along the continuous casting drawing speed direction, outputting the judgment that the slab generates the longitudinal crack defect if the number of the abnormal detection points of the next row is not less than the number of the abnormal detection points of the previous row, otherwise outputting the judgment that the slab is normal.
Further, in the system for identifying the longitudinal crack defect of the continuous casting slab based on the temperature of the crystallizer, the system further comprises an alarm device which is connected with the control device, wherein the alarm device outputs an alarm signal when the control device outputs the judgment that the longitudinal crack defect is generated by the slab.
Further, in the system for identifying the longitudinal crack defect of the continuous casting slab based on the temperature of the crystallizer, the anomaly detection algorithm is a K-nearest neighbor algorithm.
Further, in the system for identifying the longitudinal crack defect of the continuous casting slab based on the temperature of the crystallizer, the anomaly detection algorithm is an SVM algorithm.
Further, in the system for identifying the longitudinal crack defect of the continuous casting slab based on the temperature of the crystallizer, the anomaly detection algorithm is an isolated forest algorithm.
Compared with the prior art, the method and the system for identifying the longitudinal crack defects of the continuous casting slab based on the crystallizer temperature have the following advantages and beneficial effects:
according to the method for identifying the longitudinal crack defects of the continuous casting slab based on the crystallizer temperature, disclosed by the invention, the abnormal characteristics of the crystallizer temperature can be obtained in real time based on a machine learning abnormal algorithm, and the longitudinal crack identification of the slab is carried out according to the characteristics of the abnormal characteristics.
The method for identifying the longitudinal crack defect of the continuous casting slab based on the crystallizer temperature can improve the accuracy of longitudinal crack prediction: on one hand, by identifying and comparing the abnormal probabilities of the temperatures of different rows of crystallizers, further judging longitudinal cracks, distinguishing the longitudinal cracks from other factors generating abnormal temperatures, and reducing the false alarm rate; on the other hand, the judgment of abnormal characteristics is not limited to a constant rate, and is not limited to a mode of adopting a conventional variance or a temperature drop rate, so that the missing report rate of the longitudinal crack forecast can be effectively reduced.
The method fully considers the uniqueness of the longitudinal cracks of the continuous casting slab, can effectively avoid false alarm and missing alarm, improves the accuracy of longitudinal crack prediction, and has the advantages of simple calculation, easy realization and transplanting popularization by means of a python machine learning library, and very good popularization prospect and application value.
Accordingly, the system for identifying the longitudinal crack defects of the continuous casting slab based on the temperature of the crystallizer can be used for implementing the method, and has the advantages and the beneficial effects.
Drawings
Fig. 1 schematically shows the thermocouple position of a mold in an embodiment of the system for identifying a longitudinal crack defect of a continuous casting slab based on the temperature of the mold according to the present invention.
Fig. 2 schematically shows a schematic flow chart of an embodiment of the method for identifying longitudinal crack defects of a continuous casting slab based on the temperature of a crystallizer according to the invention.
Detailed Description
The method and system for identifying the longitudinal crack defects of the continuous casting slab based on the temperature of the crystallizer according to the invention will be further explained and illustrated with reference to specific examples and the accompanying drawings, but the explanation and illustration do not unduly limit the technical scheme of the invention.
In the invention, the system for identifying the longitudinal crack defect of the continuous casting slab based on the crystallizer temperature can be used for implementing the method for identifying the longitudinal crack defect of the continuous casting slab based on the crystallizer temperature, and has wide applicability and very wide application prospect.
Fig. 1 schematically shows the thermocouple position of a mold in an embodiment of the system for identifying a longitudinal crack defect of a continuous casting slab based on the temperature of the mold according to the present invention.
In the present invention, the system for identifying the longitudinal crack defect of the continuous casting slab based on the temperature of the crystallizer according to the present invention may comprise: a plurality of heat removal thermocouples, a data acquisition device and a control device. The data acquisition device can collect temperature data of thermocouples of m time point windows in real time; the control device can be used for implementing the method step for identifying the longitudinal crack defect of the continuous casting slab based on the temperature of the crystallizer.
As shown in fig. 1, in the present embodiment, the system according to the present invention may include two heat rejection couples, namely, a first heat rejection couple 3 and a second heat rejection couple 2, wherein the first heat rejection couple 3 and the second heat rejection couple 2 are disposed below the level of molten steel in the mold 1, and the first heat rejection couple 3 and the second heat rejection couple 2 are disposed along the drawing speed direction of continuous casting (i.e., the direction indicated by the arrow in the drawing), wherein each heat rejection couple includes 8 thermocouples.
Fig. 2 schematically shows a schematic flow chart of an embodiment of the method for identifying longitudinal crack defects of a continuous casting slab based on the temperature of a crystallizer according to the invention.
In the invention, the method for identifying the longitudinal crack defect of the continuous casting slab based on the crystallizer temperature can be implemented by adopting the system, and the data acquisition device in the system can collect the temperature data of each thermocouple of 200 (m) time point windows in real time and analyze the data.
Correspondingly, the control device in the system can implement the method step for identifying the longitudinal crack defect of the continuous casting slab based on the temperature of the crystallizer, which comprises the following steps:
step 1: and reading in two-heat-rejection thermocouple temperature information of 200 time point windows in real time.
Step 2: the temperature data average value of each heat rejection couple in each time point window is calculated and obtained based on the temperature data of each thermocouple in the 200 time point windows, so that each heat rejection couple obtains 200 temperature data average values, and the calculation formula is as follows:
B 1 (i)=(A 11 +A 12 +A 13 +A 14 +A 15 +A 16 +A 17 +A 18 )/8
B 2 (i)=(A 21 +A 22 +A 23 +A 24 +A 25 +A 26 +A 27 +A 28 )/8
wherein B is 1 (i) Mean value of temperature of first heat rejection couple 3 at ith time node, A 11 、A 12 ……A 18 Values at the ith time node of the individual thermocouples of the first row are respectively represented; b2 (i) represents the average value of the temperature of the second heat rejection couple 2 at the ith time node, A 21 、A 22 ……A 28 Representing the values of the individual thermocouples of the second row at the ith time node, respectively.
Step 3: abnormality detection is carried out on the average value of 200 temperature data of each thermocouple by adopting an abnormality detection algorithm, and the number n of the abnormality detection points of the two thermocouples is obtained 1 ,n 2 。
In the step 3, the anomaly detection algorithm may be a K-nearest neighbor algorithm, an SVM algorithm or an isolated forest algorithm. These algorithms belong to the anomaly detection algorithms commonly used in the prior art, and are not described in detail here.
Step 4: comparing the number of the abnormal detection points of each heat extraction couple along the continuous casting drawing speed direction, and if the number of the abnormal detection points of the next row is not less than the number of the abnormal detection points of the previous row, namely the number n of the abnormal detection points of the second heat extraction couple 2 The number n of abnormal detection points of the first heat removal couple is not less than 1 Outputting judgment of the longitudinal crack defect of the slab, outputting an alarm signal, and otherwise outputting judgment of the normal slab.
Step 5: and updating the time, namely, moving the time window downwards by 20 time points to form a new time window slab, and executing the steps 1-4 again on the updated time window, so that whether the new time window slab has crack defects or not can be predicted.
Therefore, in this embodiment, the continuous casting slab longitudinal crack defect system based on crystallizer temperature identification according to the invention can detect real-time anomalies of the slab crystallizer longitudinal two-heat-removal couple by using the anomaly detection algorithm of machine learning, and obtain the quantity (n 1 ,n 2 ) When the number of the abnormal detection points of the second heat removal couple is n 2 The number n of abnormal detection points of the first heat removal couple is not less than 1 And predicting that the continuous casting slab has longitudinal cracks. The method fully considers the uniqueness of the longitudinal crack, avoids false alarm and missing alarm, can effectively improve the accuracy of longitudinal crack prediction, has simple calculation, and is easy to realize and transplant and popularize.
Of course, in some other embodiments, when the system is provided with a plurality of heat rejection couples below the liquid level position of the molten steel in the crystallizer and each heat rejection couple comprises a plurality of thermocouples, the control device in the system can calculate and obtain the average value of the temperature data of each heat rejection couple in each time point window based on the temperature data of each thermocouple of m time point windows collected by the data collecting device in real time, so that each heat rejection couple obtains m average values of the temperature data.
The control device can further adopt an anomaly detection algorithm to perform anomaly detection on the average value of m temperature data of each heat rejection couple so as to obtain the number n of anomaly detection points of each heat rejection couple i . And then, the number of the abnormal detection points of each heat removal couple can be compared along the continuous casting drawing speed direction, if the number of the abnormal detection points of the next row is not less than the number of the abnormal detection points of the previous row, the judgment of the longitudinal crack defect generated by the slab is output, and an alarm signal is output, otherwise, the judgment of the normal state of the slab is output.
In summary, the method for identifying the longitudinal crack defect of the continuous casting slab based on the crystallizer temperature fully considers the uniqueness of the longitudinal crack defect relative to other factors causing the temperature rate to be reduced, and improves the accuracy of longitudinal crack prediction. After the system implementing the method generates longitudinal crack prediction and feeds back to an L1 system for collecting equipment information such as instruments, counter measures such as reducing the pulling speed and the specific water quantity of secondary cooling can be timely adopted in the subsequent process, so that the expansion of longitudinal cracks of the continuous casting slab is slowed down, and the aim of reducing the longitudinal crack defects of the continuous casting slab is fulfilled.
The method fully considers the uniqueness of the longitudinal cracks of the continuous casting slab, can effectively avoid false alarm and missing alarm, improves the accuracy of longitudinal crack prediction, is simple to calculate by means of a python machine learning library, is easy to realize and transplant and popularize, and has very good popularization prospect and application value.
It should be noted that the combination of the technical features in the present invention is not limited to the combination described in the claims or the combination described in the specific embodiments, and all the technical features described in the present invention may be freely combined or combined in any manner unless contradiction occurs between them.
It should also be noted that the above-recited embodiments are merely specific examples of the present invention. It is apparent that the present invention is not limited to the above embodiments, and similar changes or modifications will be apparent to those skilled in the art from the present disclosure, and it is intended to be within the scope of the present invention.
Claims (10)
1. The method for identifying the longitudinal crack defect of the continuous casting slab based on the temperature of the crystallizer is characterized by comprising the following steps:
arranging a plurality of heat removal couples below the liquid level of the molten steel of the crystallizer along the continuous casting drawing speed direction, wherein each heat removal couple comprises a plurality of thermocouples;
collecting temperature data of thermocouples of m time point windows in real time;
calculating to obtain the temperature data average value of each heat removal couple in each time point window, wherein each heat removal couple obtains m temperature data average values;
abnormality detection is carried out on the average value of m temperature data of each thermocouple by adopting an abnormality detection algorithm, and the number n of abnormality detection points of each thermocouple is obtained i ;
And comparing the number of the abnormal detection points of each heat removal thermocouple along the continuous casting drawing speed direction, if the number of the abnormal detection points of the next row is not less than the number of the abnormal detection points of the previous row, judging that the slab generates the longitudinal crack defect, otherwise, judging that the slab is normal.
2. The method for identifying a longitudinal crack defect of a continuous casting slab based on a crystallizer temperature according to claim 1, further comprising the steps of: and outputting an alarm signal when the slab is considered to generate a longitudinal crack defect.
3. The method for identifying continuous casting slab longitudinal crack defects based on the crystallizer temperature according to claim 1, wherein the anomaly detection algorithm is a K-nearest neighbor algorithm.
4. The method for identifying continuous casting slab longitudinal crack defects based on the temperature of a crystallizer according to claim 1, wherein the anomaly detection algorithm is an SVM algorithm.
5. The method for identifying continuous casting slab longitudinal crack defects based on the crystallizer temperature according to claim 1, wherein the anomaly detection algorithm is an isolated forest algorithm.
6. A system for identifying longitudinal crack defects of a continuous casting slab based on a crystallizer temperature, comprising:
the plurality of heat removal thermocouples are arranged below the liquid level position of the molten steel of the crystallizer and are arranged along the continuous casting drawing speed direction, wherein each heat removal thermocouple comprises a plurality of thermocouples;
the data acquisition device is used for actually collecting temperature data of each thermocouple of m time point windows; a control device arranged to perform the steps of:
calculating and obtaining a temperature data average value of each heat removal couple in each time point window based on the temperature data of each thermocouple in the m time point windows, so that each heat removal couple obtains m temperature data average values;
abnormality detection is carried out on the average value of m temperature data of each thermocouple by adopting an abnormality detection algorithm, and the number n of abnormality detection points of each thermocouple is obtained i ;
And comparing the number of the abnormal detection points of each heat removal thermocouple along the continuous casting drawing speed direction, outputting the judgment that the slab generates the longitudinal crack defect if the number of the abnormal detection points of the next row is not less than the number of the abnormal detection points of the previous row, otherwise outputting the judgment that the slab is normal.
7. The system for identifying a longitudinal crack defect in a continuous casting slab based on a crystallizer temperature according to claim 6, further comprising an alarm device connected to the control device, wherein the alarm device outputs an alarm signal when the control device outputs a judgment that the slab has a longitudinal crack defect.
8. The system for identifying continuous casting slab longitudinal crack defects based on the crystallizer temperature according to claim 6, wherein the anomaly detection algorithm is a K-nearest neighbor algorithm.
9. The system for identifying continuous casting slab longitudinal crack defects based on the crystallizer temperature according to claim 6, wherein the anomaly detection algorithm is an SVM algorithm.
10. The system for identifying continuous casting slab longitudinal crack defects based on crystallizer temperature according to claim 6, wherein the anomaly detection algorithm is an isolated forest algorithm.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210015464.4A CN116441502A (en) | 2022-01-07 | 2022-01-07 | Method and system for identifying longitudinal crack defects of continuous casting slab based on crystallizer temperature |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210015464.4A CN116441502A (en) | 2022-01-07 | 2022-01-07 | Method and system for identifying longitudinal crack defects of continuous casting slab based on crystallizer temperature |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116441502A true CN116441502A (en) | 2023-07-18 |
Family
ID=87118873
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210015464.4A Pending CN116441502A (en) | 2022-01-07 | 2022-01-07 | Method and system for identifying longitudinal crack defects of continuous casting slab based on crystallizer temperature |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116441502A (en) |
-
2022
- 2022-01-07 CN CN202210015464.4A patent/CN116441502A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
He et al. | Mold breakout prediction in slab continuous casting based on combined method of GA-BP neural network and logic rules | |
US11105758B2 (en) | Prediction method for mold breakout based on feature vectors and hierarchical clustering | |
CN110568010B (en) | Method for online predicting and positioning internal cracks of slab continuous casting | |
CN105522133B (en) | A kind of continuous casting billet bonds the decision method of behavior with copper plate of crystallizer | |
CN104181196B (en) | A kind of continuous casting billet surface longitudinal crack online test method | |
CN102581244A (en) | Online control system and control method for surface quality of continuous casting billet | |
CN108705058B (en) | A method of forecast Crystallizer bleed-out is clustered based on K-Means | |
Zhou et al. | Application of time series data anomaly detection based on deep learning in continuous casting process | |
CN116307289A (en) | Textile processing procedure parameter detection and prediction method, system and storage medium | |
CN117036797A (en) | Continuous casting billet longitudinal crack prediction method based on feature extraction and random forest classification | |
Kumar et al. | Development of intelligent mould for online detection of defects in steel billets | |
CN102019393A (en) | Casting body quality judging method and device for continuous casting machine | |
CN116441502A (en) | Method and system for identifying longitudinal crack defects of continuous casting slab based on crystallizer temperature | |
CN113340444A (en) | Continuous casting billet surface longitudinal crack prediction method based on thermocouple temperature and logic judgment | |
CN112926622B (en) | Crystallizer breakout prediction method for generating countermeasure network based on feature vector and SWGAN-GP | |
CN111421119A (en) | Online prediction method for longitudinal cracks on surface of continuous casting slab | |
CN110918973A (en) | Crystallizer thermal image abnormal region marking method based on run | |
JPS5946703B2 (en) | Continuous casting method using a mold equipped with a mold temperature measuring element | |
CN117171936A (en) | Slab quality prediction method for extracting real-time characteristic value of crystallizer based on defect mechanism | |
CN111666710A (en) | Method for predicting continuous casting billet longitudinal cracks by logistic regression classification | |
CN111680448B (en) | Continuous casting billet longitudinal crack prediction method based on SVM classification | |
CN115446276A (en) | Continuous casting breakout early warning method for recognizing V-shaped bonding characteristics of crystallizer copper plate based on convolutional neural network | |
CN111618265B (en) | Continuous casting billet longitudinal crack online detection method based on K nearest neighbor classification | |
CN116451158A (en) | Online prediction method for transverse cracks of continuous casting slab | |
CN116451159A (en) | Method for online detection and identification of continuous casting slab crack defects |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |