CN114653914B - Crystallizer steel leakage early warning method based on morphological reconstruction and electronic device - Google Patents
Crystallizer steel leakage early warning method based on morphological reconstruction and electronic device Download PDFInfo
- Publication number
- CN114653914B CN114653914B CN202111658843.7A CN202111658843A CN114653914B CN 114653914 B CN114653914 B CN 114653914B CN 202111658843 A CN202111658843 A CN 202111658843A CN 114653914 B CN114653914 B CN 114653914B
- Authority
- CN
- China
- Prior art keywords
- connected domain
- temperature
- sequence
- direct
- bonding
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 43
- 239000010959 steel Substances 0.000 title claims abstract description 43
- 230000000877 morphologic effect Effects 0.000 title claims abstract description 22
- 238000001914 filtration Methods 0.000 claims abstract description 10
- 238000013507 mapping Methods 0.000 claims abstract description 9
- 238000004891 communication Methods 0.000 claims description 12
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 claims description 11
- 229910052802 copper Inorganic materials 0.000 claims description 11
- 239000010949 copper Substances 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 8
- 238000005259 measurement Methods 0.000 claims description 3
- 239000013307 optical fiber Substances 0.000 claims description 3
- 238000007711 solidification Methods 0.000 abstract description 2
- 230000008023 solidification Effects 0.000 abstract description 2
- 125000004122 cyclic group Chemical group 0.000 abstract 1
- 230000002159 abnormal effect Effects 0.000 description 9
- 238000004519 manufacturing process Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 239000000853 adhesive Substances 0.000 description 3
- 230000001070 adhesive effect Effects 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 230000033001 locomotion Effects 0.000 description 3
- 238000009529 body temperature measurement Methods 0.000 description 2
- 238000009749 continuous casting Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000004484 Briquette Substances 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000005266 casting Methods 0.000 description 1
- 238000002507 cathodic stripping potentiometry Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000001066 destructive effect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000012717 electrostatic precipitator Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000005499 meniscus Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000002893 slag Substances 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 230000001360 synchronised effect Effects 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mechanical Engineering (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Radiation Pyrometers (AREA)
Abstract
The invention discloses a crystallizer steel leakage early warning method and an electronic device based on morphological reconstruction, wherein the method comprises the following steps: and (5) acquiring temperature data of the thermocouple, and constructing a temperature sequence. Constructing a direct index sequence, dividing the data of the two-dimensional temperature determinant into a plurality of temperature segments after sorting according to the size, sequentially and respectively establishing mapping between each temperature segment and at least two direct index numbers in the direct index sequence K, constructing an equivalent direct set of the two-dimensional temperature determinant by utilizing the thermocouple position and the direct index numbers to form a time sequence direct sequence equivalent to the temperature sequence, visualizing the time sequence direct sequence into a direct image, acquiring a connected domain of each layer, forming a connected domain set, performing cyclic traversal on the connected domain, and obtaining potential bonding billets according to the shape, the position and the direct index numbers of the connected domain. According to the method, the temperature of the crystallizer is subjected to a squaring image, the filtering condition and the judging condition are built by combining the growth characteristics of the bonding solidification blank shell, and the accuracy of bonding early warning is improved.
Description
Technical Field
The invention relates to the technical field of continuous casting, in particular to a crystallizer steel leakage early warning method based on morphological reconstruction and an electronic device.
Background
In metallurgical continuous casting production, crystallizer steel leakage is a catastrophic production accident with higher frequency and higher destructive, the reasons for steel leakage are more, the production events of steel leakage are also same, and there are pouring steel leakage, bonding steel leakage, crack steel leakage and the like, the bonding steel leakage is adhesion between a primary blank shell and a copper plate of the crystallizer in the solidification process of molten steel, bonding steel leakage is the most frequent accident type in all steel leakage accidents, and the bonding steel leakage accounts for about 80% of the total number of steel leakage, so the main contradiction of steel leakage prevention is to solve the identification and early warning of bonding steel leakage.
Bonding is a defect of surface quality of a casting blank, and occurs in a zone where a green shell is formed at a meniscus in a crystallizer, and the movement change of a high-temperature hot spot and a bonding block cold spot is shown along with the production, wherein the high-temperature hot spot moves downwards along with the development of bonding, and the bonding block is increased and thickened continuously to form the bonding cold spot.
After bonding, the pulling speed is reduced (generally reduced to a lower pulling speed rather than waiting for shutdown), the occurrence of steel leakage accidents can be avoided after the bonding blocks are ensured to completely fall off, and the accurate judgment of the falling of the bonding blocks is an important measure for ensuring that bonding steel leakage does not occur.
The metallurgical science and technology workers carry out a great deal of researches on bonding steel leakage, and the judgment of bonding early warning is mainly based on a neural network method and a logic judgment method, and also based on data mining and a related method based on a temperature matrix image.
The above methods are all based on the technology of measuring the temperature of the crystallizer copper plate, and the method of measuring the temperature by using a plurality of rows of thermocouples is adopted at home and abroad at present, and the problems that the maintenance of the temperature thermocouples and the production process are more abnormal and the methods are difficult to be well adapted are all caused.
CN111496211B discloses a steel leakage early warning method based on a temperature and temperature rise rate matrix, which adopts preset temperature and temperature rise rate thresholds to carry out plane segmentation and identify bonding hot spot areas, and because hot spot high temperature areas frequently occur in production, the hot spot high temperature areas have quite close relation with the distribution of a crystallizer flow field and the slagging of protective slag, and therefore, the occurrence of bonding is judged by insufficient existence of hot spots.
CN107096899a discloses a method for early warning of missing steel based on time sequence characteristics of local temperature measurement data, which is sensitive to abnormalities of the thermocouple due to the dependence on the temperature measurement result of the local temperature thermocouple, and in which the motion characteristics are inevitably thresholded.
Disclosure of Invention
In order to solve the problems, the invention discloses a crystallizer steel leakage early warning method based on morphological reconstruction, which comprises the following steps:
the temperature data acquisition is carried out on thermocouples buried on a crystallizer copper plate for k times, the temperature data acquired at the time t is combined with the position of the thermocouples to form a two-dimensional temperature determinant M (t) at the time t, and the two-dimensional temperature determinant extracted for k times in succession is collected to form a temperature sequence
U={M(0),M(1),M(2),…M(k-2),M(k-1)};
Constructing a direct index sequence M= { k 1 ,k 2 ,…,k b-1 ,k b The direct index sequence K comprises b direct index numbers, all data of each two-dimensional temperature determinant are divided into a plurality of temperature segments after being ordered according to the size, mapping is sequentially and respectively built between each temperature segment and at least two direct index numbers in the direct index sequence K, and an equivalent direct set q of the two-dimensional temperature determinant M is constructed by utilizing thermocouple positions and the direct index numbers to form a time sequence direct sequence equivalent to the temperature sequence U
Q={q(0),q(1),q(2),…q(k-2),q(k-1)};
Visualizing the time sequence straightening sequence Q into a straightening image, dividing the straightening image into different layers according to a straightening index number, and obtaining a connected domain of each layer to form a connected domain set
B={b(0),b(1),b(2),…b(n-2),b(n-1)},
n-1 represents an nth connected domain, each connected domain containing a plurality of independent element objects e j The independent element object e j Namely, representing the position and the straight index number of each thermocouple;
and (3) carrying out circulation traversal on all the connected domains, and obtaining the potential bonding compact according to the shape and the position of the connected domains and the corresponding straight index numbers.
Optionally, the obtaining the potential bonding compact according to the shape and the position of the connected domain and the corresponding low-straight index number refers to that the following filtering rule is satisfied, and then the potential bonding compact is determined:
a: independent element objects in the connected domain corresponding to the low-square index number are reserved, and other connected domains corresponding to the low-square index number are removed, wherein the low-square index number refers to at least two square index numbers corresponding to the temperature range of the low-temperature bonding compact under the bonding condition;
b: only the connected domain containing the upper boundary element is reserved;
c: removing connected domains containing both upper and lower boundary elements;
d: only the lower triangle communication domain is reserved;
the lower triangle communication domain means that the communication domain is triangular in shape and one corner of the triangle is downward.
Optionally, the potentially cohesive compact satisfying the following cohesive determination conditions is determined to be a cohesive compact:
e: judging whether the connected domain meets
Wherein i, j are corresponding direct index numbers, and the signs represent the connected domain b i The independent element objects also comprise all connected domains inside the element objects;
f: judging whether the connected domain meets the continuous k * In each cycle, b i (t).Width>b i (t-1).Width,b i (t).A>b i (t-1).A,
Wherein b i Width represents the Width of the connected domain at time t;
b i (t). A represents the area of the connected domain at time t;
k * less than the period interval k of the temperature acquisition.
Optionally, before constructing the direct index sequence K, the method further includes: and carrying out subdivision interpolation on the whole thermocouple measurement area by using normal thermocouple data and adopting a Lagrange interpolation method on each two-dimensional temperature determinant, and forming a new two-dimensional temperature determinant M (t) after interpolation.
Optionally, the thermocouple is a K-type temperature sensor or an optical fiber sensor, and the thermocouple is directly connected with the acquisition card for data acquisition, or the thermocouple is connected with the PLC automation system, the PLC automation system acquires temperature data at high frequency, and the acquisition card reads the temperature data from the PLC automation system through network communication.
Optionally, the direct index sequence K is a monotonically increasing integer number column.
Optionally, the number of the direct index numbers in the direct index sequence K is odd and greater than or equal to 3.
Optionally, the method includes the steps of:
calculating the statistical parameters of each connected domain, wherein the statistical parameters comprise the number of independent element objects, the shape of the connected domain, the width of the connected domain, the height of the connected domain and an index weighted average value, the index weighted average value is the sum of products of straight index numbers corresponding to each independent element object and the connected domain in the connected domain, and dividing the sum by the total number of independent element objects, and each connected domain contains the objects in the internal connected domain when calculating the statistical parameters.
Optionally, in the connected domain identification process, each layer independently performs connected domain identification, and uses conditions e and f in the bonding judgment conditions to perform bonding compact judgment;
or adopting all layers with low straight index numbers to conduct connected domain identification, wherein the layers with straight index numbers smaller than or equal to the straight index numbers are adopted as a whole to conduct layer connected domain identification of each straight index number, and if b j =(b i ∩b j ) I { j < i }, where i, j is the corresponding direct index number, then the bonding compact determination is made using the condition f in the bonding determination condition.
The invention also provides an electronic device which comprises a memory and at least one processor, wherein at least one instruction is stored in the memory, and the at least one instruction realizes the crystallizer steel leakage early warning method based on morphological reconstruction when being executed by the at least one processor.
According to the method, the crystallizer copper plate temperature is integrally distributed as a temperature data source, a straightened image is formed through temperature data, filtering conditions and judging conditions are built by combining the growth characteristics of the bonding solidified blank shell, the form of the bonding blank shell and the movement characteristics of the bonding blank shell are identified by the filtering conditions and the judging conditions, so that the bonding defect is identified, the accuracy of identifying the bonding quality defect is greatly improved, the accuracy of bonding early warning is greatly improved, and the steel leakage accident rate is reduced. The method is suitable for plate blanks, square billets, round billets, special-shaped billets, CSPs, ESPs and the like, reduces false alarms,
drawings
The above-mentioned features and technical advantages of the present invention will become more apparent and readily appreciated from the following description of the embodiments thereof, taken in conjunction with the accompanying drawings.
FIG. 1 is a flow chart showing a method for warning of a crystallizer bleed-out based on morphological reconstruction according to an embodiment of the present invention;
FIG. 2 is a layout diagram of a thermocouple showing an embodiment of the present invention;
FIG. 3 is a schematic diagram showing the mapping of the direct index sequence and the temperature segments according to an embodiment of the present invention;
FIG. 4 is a schematic diagram showing a squaring diagram of an embodiment of the invention;
FIG. 5 is a schematic diagram showing connected domains obtained by any layer in an embodiment of the present invention;
fig. 6 is a schematic diagram illustrating a potential cohesive compact obtained using the filtering rules in accordance with an embodiment of the present invention.
Detailed Description
Embodiments of the present invention will be described below with reference to the accompanying drawings. Those skilled in the art will recognize that the described embodiments may be modified in various different ways, or combinations thereof, without departing from the spirit and scope of the invention. Accordingly, the drawings and description are to be regarded as illustrative in nature and not as restrictive in scope. Furthermore, in the present specification, the drawings are not drawn to scale, and like reference numerals denote like parts.
The method for early warning of crystallizer steel leakage based on morphological reconstruction in this embodiment, as shown in fig. 1, includes:
step S1, performing synchronous temperature data acquisition on thermocouples 1 buried on a crystallizer copper plate 2 at a period interval k times, wherein the thermocouples are m+1 rows and n+1 columns, M > =1, n > =1, constructing a two-dimensional temperature determinant at the time t as M (t) according to the corresponding thermocouple arrangement positions, and collecting the two-dimensional temperature determinant at the time t into a temperature sequence U, wherein the two-dimensional temperature determinant at the time t is as follows:
where i is the ith row, j is the jth column, t i,j The thermocouple temperature at the (i, j) position is shown. The temperature sequence U is as follows:
U={M(0),M(1),M(2),…M(k-2),M(k-1)}
preferably k=5, i.e. preferably k takes a value of 5 cycles, constituting the temperature sequence U.
Fig. 2 is an expanded view of a crystallizer copper plate exemplified by a slab caster in a certain factory, wherein the total width of the crystallizer wide copper plate is 1775mm, the width of the narrow copper plate is 170mm, the height of the crystallizer is 900mm, the wide side is embedded with 3 rows and 11 columns of thermocouples, the narrow side is embedded with 3 rows and 2 columns of thermocouples, 78 thermocouples are embedded in total, the row spacing of the thermocouples is 115mm, and the column spacing is 150mm. Table one shows the measured temperature data of the thermocouple at a certain time.
List one
The thermocouple 1 may be a K-type temperature sensor, or may be an optical fiber sensor, the temperature obtained from the thermocouple may be obtained by directly connecting the thermocouple with the acquisition card, or may be obtained by connecting the thermocouple with a PLC automation system, and the PLC automation system acquires temperature data at high frequency and reads the temperature data from the PLC automation system through network communication.
Step S3, straightening the temperature data,
step S31, constructing a direct index sequence K= { K 1 ,k 2 ,…,k b-1 ,k b The sequence of direct indexes K is b elements, each element representing a direct index number, preferably defined as a monotonically increasing series of integers, for exampleFor example, k= {1 2 … 89 }, in this embodiment, the number of elements in the direct index sequence K is preferably odd and 3 or more, and b=7 is recommended;
step S32, for any two-dimensional temperature determinant, obtaining the data maximum value m in the two-dimensional temperature determinant max And a minimum value m min The data in the two-dimensional temperature determinant is sorted by size, and (m) min ,…,m max ) The temperature data in the index sequence K is divided into a plurality of temperature segments at a certain interval, and each temperature segment is mapped with at least two elements in the index sequence K in turn (that is, the temperature data does not need to be mapped to each element in the index sequence K in an all-coverage mode). As shown in fig. 3, the direct index sequence k= {1,2,3,4,5,6,7}, the two-dimensional temperature determinant M is divided into 7 segments, the first segment establishes a mapping with the number 1, the second segment establishes a mapping with the number 2, the third segment establishes a mapping with the number 3, the fourth segment establishes a mapping with the number 4, and so on.
S33, constructing an equivalent rectangularity set q of M by using the mapping, wherein M < = > q when M is equivalent to q information;
corresponding to the temperature sequence
U= { M (0), M (1), M (2), … M (k-2), M (k-1) } equivalent squaring thereof
The sequential orthonormal set sequence formed by the set q is
Q={q(0),q(1),q(2),…q(k-2),q(k-1)}
In step S34, the equivalent rectangulation set q is visualized to form a rectangulated image, and fig. 4 shows a rectangulated image when 7 elements of the rectangulated index sequence K are present.
Step S4, the rectangulated image information is decomposed,
in step S41, the straightened image is divided into different layers according to the elements in the sequence of the straightened indexes, and layer L (i) corresponds to the straight index number in the sequence of the straight indexes K. For example, in fig. 4, other layers are masked, only layer 7 is shown, and the resulting layer information is shown in fig. 5.
Step S42, for any layer, obtaining each independent area not communicated with each other by adopting recursive deduction, as shown in FIG. 5, 4 independent areas are arranged in layer 7The domains, defining independent regions as connected domains b, different connected domains being given different numbers b i . Each layer is circulated, and a connected domain of each layer is generated to form a connected domain set B;
B={b(0),b(1),b(2),…b(n-2),b(n-1)}。
each connected domain contains an indefinite number of independent element objects e j Each independent element object e j Constitutes the smallest counting unit, where the individual element object e j I.e. thermocouple temperature data representing each temperature measuring point, b i ={e j Straight side index numbers are the same and communicated;
in the communicating region identification process, each layer is preferably used for carrying out communicating region identification independently, but in the implementation process, particularly in the problem of adhesive block identification, all layers with low straight index numbers are also used as a whole for carrying out communicating region identification, wherein the low straight index numbers refer to one or more straight index numbers corresponding to the temperature range of the low-temperature adhesive compact in the case of adhesive, for example, straight index numbers 1 and 2 are low straight index numbers.
Wherein, each layer is adopted to identify the connected domain independently (the connected domain may have a containing relationship, that is, the connected domain of one straight index number may contain the connected domain of other straight index numbers), for example, for 1 to 7 layers, each layer identifies the connected domain respectively. The conditions e and f are used in the adhesion determination condition to make the adhesion briquette determination.
Wherein, the use of all layers of the low-direct index number for identifying the connected domain means that, in the low-direct index number, the connected domain is identified by adopting the layer of the direct index number less than or equal to the layer of the direct index number as a whole (the connected domain may have an overlapping relationship, i.e. the connected domain of one direct index number may overlap with the connected domain of the other direct index number (also less than or equal to the layer of the low-direct index number)), namely, if b j =(b i ∩b j ) I { j < i }, where i, j is the corresponding direct index number, then the bonding compact determination is made using the condition f in the bonding determination condition. For example, the index number of low straight direction is 1 to 3, then the layers of 1 to 3 are reserved, and the straight direction is to be formedAnd when the layers of index numbers 1 to 3 are used as the whole to identify the connected domain, and then the connected domain of the straight index number 2 is identified, the layers of the straight index numbers 1 and 2 are used as the whole to identify the connected domain, and then the connected domain of the straight index number 1 is identified. Identifying and then using the identified connected domain according to b j =(b i ∩b j ) I { j < i } to determine whether there is an overlapping relationship.
And (3) carrying out connected domain identification by adopting all layers with low straight index numbers, wherein the influence is that the step e of judging the set inclusion rule is not needed in the follow-up bonding judgment rule. The low-straight index number refers to a straight index number corresponding to the temperature of the low-temperature bonded compact where bonding is generally occurring. It should be noted that the above-mentioned inclusion and overlap phenomenon only occurs for the connected domain identification of the layer under different division methods, but the actual effect is the same.
Step S43, calculating the statistical parameters of each connected domain, wherein the statistical parameters comprise the number of independent element objects, the shape of the connected domain, the width of the connected domain, the height of the connected domain and an index weighted average; wherein the index weighted average is the sum of products of the direct index numbers of each independent element object (the object in the connected domain containing the internal low direct index number) in the connected domain and the corresponding direct index number of the connected domain, and divided by the total number of the independent element objects.
Step S5, the bonding form is identified,
and (3) performing circulation traversal on all the connected domains, and obtaining a potential bonding compact by adopting the following filtering rules, wherein after filtering the connected domain objects according to the following rules, as shown in fig. 6, the filtering rules are as follows:
a: and reserving independent element objects in the connected domain corresponding to the low straight-side index number, and removing the connected domain corresponding to the other straight-side index numbers. The low-straight index number refers to a straight index number corresponding to the temperature of the low-temperature bonded compact where bonding is generally occurring. For example, in this embodiment (the straight index numbers are 0 to 6), the independent element objects in the connected domain corresponding to the straight index numbers 0,1,2 are reserved (at least 2 straight index numbers are reserved);
b: only the connected domain containing the upper boundary element is reserved;
c: removing connected domains containing both upper and lower boundary elements;
d: only the lower triangle communication field is reserved.
The upper boundary element refers to an element positioned at the upper boundary of the straightened image;
the lower boundary element refers to an element positioned at the lower boundary of the rectangulated image;
the lower triangle communication domain means that the communication domain is triangular in shape and one corner of the triangle is downward.
Step S6, it is determined whether bonding occurs by the following bonding determination rule.
Whether the identified connected domain object is a low-temperature bonding compact or not is still insufficient only in form identification, whether bonding occurs or not needs to be further confirmed according to the following bonding judgment rule, and then the independent element object of the connected domain is considered as a bonding object, and the bonding event is judged to occur:
e: when the bonding compact occurs, the temperature of the bonding compact gradually rises from top to bottom, namely the connected domain element set is fully contained;
the connected domain object conforms to the collective operation:wherein i, j are the corresponding squaring indexes, and the sign indicates the connected domain b i But also contains the independent element objects of all connected domains inside.
f: along with the development of bonding, the bonding area is continuously and transversely expanded downwards, the area A of the connected domain object is continuously expanded, the direct index weighting index is continuously increased, and the continuous increasing period is k * ,k * The period interval k times smaller than the temperature acquisition, k in this embodiment * Taking the mixture to be more than or equal to 3;
i.e. continuously increasing the period k * In b i (t).Width>b i (t-1).Width,b i (t).A>b i (t-1).A,
Wherein b i Width represents the Width of the connected domain at time t;
b i and (t). A represents the area of the connected domain at time t.
After the bonding early warning is sent out, the bonding processing state is entered, the production system receives the early warning signal, starts the action of setting the low pulling speed for reducing the pulling speed, and carries out the speed reduction processing.
In an alternative embodiment, step S2 is further included before step S3, and the abnormal data correction is performed on each two-dimensional temperature determinant.
Specifically, in the use process of the thermocouple, the thermocouple cannot be replaced in time due to damage caused by failure or other reasons, and the phenomenon that the difference between the measured data of the thermocouple and the actual temperature value of the crystallizer copper plate is large is mainly shown in that the thermocouple is completely failed and is in a decoupling state, the measured temperature is seriously deviated from the actual temperature of the copper plate or is subjected to severe abnormal fluctuation, and the like, so that the abnormal measured data are required to be cleaned and corrected.
For the correction of the abnormal data, the normal data closest to the moment in the time sequence can be adopted to replace the abnormal data, or adjacent thermocouples can be adopted to replace the abnormal data through trend calculation, and in the embodiment, after the abnormal data points are marked, fitting interpolation is carried out through the temperature of the same row of thermocouples, so that the correction of the abnormal data is completed. Specifically, the thermocouple arrangement space is divided according to a certain interpolation interval, interpolation is preferably carried out by adopting the interpolation interval of 10mm, the whole thermocouple measurement area is subdivided and interpolated by adopting the Lagrange interpolation method by utilizing normal thermocouple data, and a new two-dimensional temperature determinant M (t) is formed after interpolation.
The invention also provides an electronic device, and in the embodiment, the crystallizer steel leakage early warning method based on morphological reconstruction can be applied to the electronic device to judge the detected real-time speed/real-time acceleration. Specifically, for an electronic device needing to perform the form-reconstruction-based crystallizer steel leakage early warning, the function for the form-reconstruction-based crystallizer steel leakage early warning provided by the method can be directly integrated on the electronic device, or the function can be operated on the electronic device in a form of a software development kit.
The following describes a hardware device architecture for implementing the crystallizer steel leakage early warning method based on morphological reconstruction, wherein the electronic device comprises a memory and at least one processor. The memory is used for storing program codes and various data, such as a crystallizer steel leakage early warning program based on morphological reconstruction and the like which are installed in the electronic device, and realizing high-speed and automatic program or data access in the running process of the electronic device. The memory includes read-only memory, programmable read-only memory, erasable programmable read-only memory, one-time programmable read-only memory, electronically erasable rewritable read-only memory, read-only or other optical disk memory, magnetic tape memory, or any other computer readable storage medium capable of being used to carry or store data.
The at least one processor may be comprised of integrated circuits, for example, a single packaged integrated circuit, or may be comprised of multiple integrated circuits packaged with the same or different functionality, including one or more central processing units, microprocessors, digital processing chips, graphics processors, combinations of various control chips, and the like. The at least one processor is a control core of the electronic device, and is connected with various components of the whole electronic device by various interfaces and lines, and executes various functions of the electronic device and processes data, such as a crystallizer steel leakage early warning function based on morphological reconstruction, by running or executing programs or modules stored in the memory and calling data stored in the memory.
The crystallizer steel leakage early warning method based on morphological reconstruction can be divided into a plurality of functional modules consisting of program code segments, each functional module is different program codes corresponding to the division of the crystallizer steel leakage early warning method based on morphological reconstruction, and the program codes of each program segment can be stored in a memory and executed by at least one processor so as to realize the crystallizer steel leakage early warning method based on morphological reconstruction.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. A crystallizer steel leakage early warning method based on morphological reconstruction is characterized by comprising the following steps:
temperature data acquisition is carried out on thermocouples buried on a crystallizer copper plate for k times, the temperature data acquired at the time t is combined with the position of the thermocouples to form a two-dimensional temperature determinant M (t) at the time t, and the two-dimensional temperature determinant sets extracted for k times in succession are a temperature sequence U= { M (0), M (1), M (2), … M (k-2), M (k-1) };
constructing a direct index sequence K= { K 1 ,k 2 ,…,k b-1 ,k b The method comprises the steps that a direct index sequence K comprises b direct index numbers, all data of each two-dimensional temperature determinant are divided into a plurality of temperature sections after being ordered according to the size, mapping is sequentially and respectively built between each temperature section and at least two direct index numbers in the direct index sequence K, an equivalent direct set Q of the two-dimensional temperature determinant M is constructed by utilizing thermocouple positions and the direct index numbers, and a time sequence direct sequence Q= { Q (0), Q (1), Q (2), … Q (K-2), Q (K-1) } equivalent to the temperature sequence U is formed;
visualizing the time sequence squaring sequence Q into a squaring image, dividing the squaring image into different layers according to a squaring index number, obtaining a connected domain of each layer to form a connected domain set B= { B (0), B (1), B (2), … B (n-2), B (n-1) },
n-1 represents an nth connected domain, each connected domain containing a plurality of independent element objects e j The independent element object e j Namely, representing the position and the straight index number of each thermocouple;
performing circulation traversal on all the connected domains, obtaining potential bonding briquettes according to the shapes and positions of the connected domains and the corresponding straight index numbers,
the step of obtaining the potential bonding compact according to the shape and the position of the connected domain and the corresponding straight index number refers to that the following filtering rule is satisfied, and then the potential bonding compact is judged:
a: independent element objects in the connected domain corresponding to the low-square index number are reserved, and other connected domains corresponding to the low-square index number are removed, wherein the low-square index number refers to at least two square index numbers corresponding to the temperature range of the low-temperature bonding compact under the bonding condition;
b: only the connected domain containing the upper boundary element is reserved;
c: removing connected domains containing both upper and lower boundary elements;
d: only the lower triangle communication domain is reserved;
the lower triangle communication domain means that the communication domain is triangular in shape and one corner of the triangle is downward,
the potentially cohesive compact satisfying the following cohesive judgment conditions is judged as a cohesive compact:
e: judging whether the connected domain meets
Wherein i, j are corresponding direct index numbers, and the signs represent the connected domain b i The independent element objects also comprise all connected domains inside the element objects;
f: judging whether the connected domain meets the continuous k * In each cycle, b i (t).Width>b i (t-1).Width,b i (t).A>b i (t-1).A,
Wherein b i Width represents the Width of the connected domain at time t;
b i (t). A represents the area of the connected domain at time t;
k * less than k times the temperature was acquired.
2. The method for early warning of crystallizer steel leakage based on morphological reconstruction according to claim 1, further comprising, before constructing the direct index sequence K: and carrying out subdivision interpolation on the whole thermocouple measurement area by using normal thermocouple data and adopting a Lagrange interpolation method on each two-dimensional temperature determinant, and forming a new two-dimensional temperature determinant M (t) after interpolation.
3. The crystallizer steel leakage early warning method based on morphological reconstruction according to claim 1, wherein the thermocouple is a K-type temperature sensor or an optical fiber sensor, the thermocouple is directly connected with the acquisition card for data acquisition, or the thermocouple is connected with a PLC automation system, the PLC automation system acquires temperature data at high frequency, and the acquisition card reads the temperature data from the PLC automation system through network communication.
4. The method for early warning of crystallizer steel leakage based on morphological reconstruction according to claim 1, wherein the straight index sequence K is a monotonically increasing integer number sequence.
5. The method for warning of crystallizer steel leakage based on morphological reconstruction according to claim 4, wherein the number of the direct index numbers in the direct index sequence K is an odd number and is equal to or greater than 3.
6. The method for warning of mold steel leakage based on morphological reconstruction according to claim 1, wherein before circulating all connected domains and obtaining the potential bonding compact by adopting the filtering rule, further comprising:
calculating the statistical parameters of each connected domain, wherein the statistical parameters comprise the number of independent element objects, the shape of the connected domain, the width of the connected domain, the height of the connected domain and an index weighted average value, the index weighted average value is the sum of products of straight index numbers corresponding to each independent element object and the connected domain in the connected domain, and dividing the sum by the total number of independent element objects, and each connected domain contains the objects in the internal connected domain when calculating the statistical parameters.
7. The method for early warning of crystallizer steel leakage based on morphological reconstruction according to claim 1, wherein,
in the connected domain identification process:
each layer independently carries out connected domain identification, and uses conditions e and f in the bonding judgment conditions to judge the bonding compact;
or adopting all layers with low straight index numbers to conduct connected domain identification, wherein the layers with straight index numbers smaller than or equal to the straight index numbers are adopted as a whole to conduct layer connected domain identification of each straight index number, and if b j =(b i ∩b j )|{j<i, where i, j is the corresponding direct index number, the bonding compact determination is made using the condition f in the bonding determination condition.
8. An electronic device, characterized in that it comprises a memory and at least one processor, said memory having stored therein at least one instruction which, when executed by said at least one processor, implements the morphology-based reconstruction crystallizer breakout warning method according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111658843.7A CN114653914B (en) | 2021-12-30 | 2021-12-30 | Crystallizer steel leakage early warning method based on morphological reconstruction and electronic device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111658843.7A CN114653914B (en) | 2021-12-30 | 2021-12-30 | Crystallizer steel leakage early warning method based on morphological reconstruction and electronic device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114653914A CN114653914A (en) | 2022-06-24 |
CN114653914B true CN114653914B (en) | 2023-12-29 |
Family
ID=82026360
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111658843.7A Active CN114653914B (en) | 2021-12-30 | 2021-12-30 | Crystallizer steel leakage early warning method based on morphological reconstruction and electronic device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114653914B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114850431B (en) * | 2022-07-05 | 2022-10-21 | 北京科技大学 | Method for forecasting bleed-out of continuous casting crystallizer |
Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS63104766A (en) * | 1986-10-21 | 1988-05-10 | Sumitomo Metal Ind Ltd | Predicting method for breakout in continuous casting |
JP2001025851A (en) * | 1999-07-12 | 2001-01-30 | Kawasaki Steel Corp | Method and instrument for automatically detecting abnormal state in molten metal treating equipment |
US6885907B1 (en) * | 2004-05-27 | 2005-04-26 | Dofasco Inc. | Real-time system and method of monitoring transient operations in continuous casting process for breakout prevention |
JP2013221659A (en) * | 2012-04-16 | 2013-10-28 | Jfe Steel Corp | Prediction method for steel leakage position |
CN103639385A (en) * | 2013-12-05 | 2014-03-19 | 中冶连铸技术工程股份有限公司 | Bleed-out forecasting method and system based on least squares |
KR101905832B1 (en) * | 2017-06-01 | 2018-11-30 | 주식회사 포스코 | Method and apparatus for detecting breackout in continuous casting process |
CN109954854A (en) * | 2019-04-10 | 2019-07-02 | 中冶赛迪工程技术股份有限公司 | Conticaster crystallizer bleed-out monitoring method, device, storage medium and electric terminal |
CN111761040A (en) * | 2020-07-14 | 2020-10-13 | 五矿营口中板有限责任公司 | Method and device for online identifying reverse sequence of thermocouple installation of crystallizer breakout prediction system |
CN112033493A (en) * | 2020-08-11 | 2020-12-04 | 东北大学 | Continuous casting molten pool liquid level detection system and continuous casting molten pool liquid level control method |
CN112296297A (en) * | 2020-09-30 | 2021-02-02 | 首钢集团有限公司 | Method for controlling water gap blockage and electronic equipment |
CN113319259A (en) * | 2021-06-07 | 2021-08-31 | 东北电力大学 | Bonding breakout logic judgment method based on space-time sequence characteristics |
JP6950860B1 (en) * | 2020-06-18 | 2021-10-13 | Jfeスチール株式会社 | Breakout prediction method, continuous casting machine operation method, and breakout prediction device |
-
2021
- 2021-12-30 CN CN202111658843.7A patent/CN114653914B/en active Active
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS63104766A (en) * | 1986-10-21 | 1988-05-10 | Sumitomo Metal Ind Ltd | Predicting method for breakout in continuous casting |
JP2001025851A (en) * | 1999-07-12 | 2001-01-30 | Kawasaki Steel Corp | Method and instrument for automatically detecting abnormal state in molten metal treating equipment |
US6885907B1 (en) * | 2004-05-27 | 2005-04-26 | Dofasco Inc. | Real-time system and method of monitoring transient operations in continuous casting process for breakout prevention |
JP2013221659A (en) * | 2012-04-16 | 2013-10-28 | Jfe Steel Corp | Prediction method for steel leakage position |
CN103639385A (en) * | 2013-12-05 | 2014-03-19 | 中冶连铸技术工程股份有限公司 | Bleed-out forecasting method and system based on least squares |
KR101905832B1 (en) * | 2017-06-01 | 2018-11-30 | 주식회사 포스코 | Method and apparatus for detecting breackout in continuous casting process |
CN109954854A (en) * | 2019-04-10 | 2019-07-02 | 中冶赛迪工程技术股份有限公司 | Conticaster crystallizer bleed-out monitoring method, device, storage medium and electric terminal |
JP6950860B1 (en) * | 2020-06-18 | 2021-10-13 | Jfeスチール株式会社 | Breakout prediction method, continuous casting machine operation method, and breakout prediction device |
CN111761040A (en) * | 2020-07-14 | 2020-10-13 | 五矿营口中板有限责任公司 | Method and device for online identifying reverse sequence of thermocouple installation of crystallizer breakout prediction system |
CN112033493A (en) * | 2020-08-11 | 2020-12-04 | 东北大学 | Continuous casting molten pool liquid level detection system and continuous casting molten pool liquid level control method |
CN112296297A (en) * | 2020-09-30 | 2021-02-02 | 首钢集团有限公司 | Method for controlling water gap blockage and electronic equipment |
CN113319259A (en) * | 2021-06-07 | 2021-08-31 | 东北电力大学 | Bonding breakout logic judgment method based on space-time sequence characteristics |
Non-Patent Citations (2)
Title |
---|
基于图像处理的黏结漏钢可视化检测方法;刘宇等;中南大学学报(自然科学版);第47卷(第02期);第380-385页 * |
漏钢预报新技术研发及应用;蔡娥等;钢铁研究;第37卷(第4期);第29-32页 * |
Also Published As
Publication number | Publication date |
---|---|
CN114653914A (en) | 2022-06-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR101189995B1 (en) | Bottleneck device extracting method and bottleneck device extracting assistance device | |
CN114653914B (en) | Crystallizer steel leakage early warning method based on morphological reconstruction and electronic device | |
JP6274226B2 (en) | Method, apparatus and program for determining casting state in continuous casting | |
CN102886504B (en) | Crystallizer bleed-out visual forecasting method based on machine vision | |
CN107159862B (en) | A kind of crystallizer bleedout prediction electric thermo method that logic-based judges | |
CN103639385B (en) | Based on breakout prediction method and the system of least square | |
CN104181196B (en) | A kind of continuous casting billet surface longitudinal crack online test method | |
CN102151814B (en) | Bonding alarm method and system during continuous casting production | |
CA2727558A1 (en) | Method for predicting the occurrence of longitudinal cracks in continuous casting | |
CN107096899B (en) | A kind of crystallizer bleedout prediction electric thermo system that logic-based judges | |
CN108469313B (en) | Copper plate of crystallizer temperature anomaly method for detecting area based on cellular automata | |
CN102019393A (en) | Casting body quality judging method and device for continuous casting machine | |
CN114664000B (en) | Method for identifying bonding and demolding of crystallizer and electronic device | |
KR20210018493A (en) | Lane property detection | |
CN111761040B (en) | Method and device for online identifying reverse sequence of thermocouple installation of crystallizer breakout prediction system | |
CN110523968B (en) | Early warning judgment method and early warning judgment device for bonding breakout | |
JPWO2017164368A1 (en) | Monitoring device, monitoring method, program | |
CN113814369B (en) | Crystallizer bleed-out forecasting method based on eigenvector and Adaboost integrated model | |
JPH09248661A (en) | Method for predicting breakout caused by foreign matter biten in continuous casting | |
Liu et al. | A Visual Detection Method of Longitudinal Crack Based on Computer Image Processing During Slab Continuous Casting | |
CN113065219A (en) | Method and device for detecting abnormality of train traction transformer cooling system | |
JP6358199B2 (en) | Method and apparatus for determining surface defects of continuous cast slab, and method for producing steel slab using the surface defect determination method | |
CN117034176B (en) | Casting abnormality prediction system in directional solidification process | |
JPH04178252A (en) | Method for foreseeing constrained break out of continuous casting | |
JP2002194405A (en) | Method and device for monitoring operation of blast furnace, and computer-readable recording medium |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20240222 Address after: Room 4093, 4th Floor, Building 26, No. 6 Haiying Road, Fengtai District, Beijing, 100071 Patentee after: Beijing Shuyu Technology Development Co.,Ltd. Country or region after: China Address before: 909, building 26, Fuxingmenwai street, Xicheng District, Beijing 100038 Patentee before: Bai Jubing Country or region before: China |