CN112926622A - Crystallizer breakout prediction method for generating countermeasure network based on feature vector and SWGAN-GP - Google Patents

Crystallizer breakout prediction method for generating countermeasure network based on feature vector and SWGAN-GP Download PDF

Info

Publication number
CN112926622A
CN112926622A CN202110087095.5A CN202110087095A CN112926622A CN 112926622 A CN112926622 A CN 112926622A CN 202110087095 A CN202110087095 A CN 202110087095A CN 112926622 A CN112926622 A CN 112926622A
Authority
CN
China
Prior art keywords
model
sample
crystallizer
breakout
feature vector
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.)
Granted
Application number
CN202110087095.5A
Other languages
Chinese (zh)
Other versions
CN112926622B (en
Inventor
王旭东
王砚宇
段海洋
姚曼
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Dalian University of Technology
Original Assignee
Dalian University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Dalian University of Technology filed Critical Dalian University of Technology
Priority to CN202110087095.5A priority Critical patent/CN112926622B/en
Publication of CN112926622A publication Critical patent/CN112926622A/en
Application granted granted Critical
Publication of CN112926622B publication Critical patent/CN112926622B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • Molecular Biology (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Physics (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Mechanical Engineering (AREA)
  • Continuous Casting (AREA)

Abstract

The invention provides a crystallizer bleed-out forecasting method for generating a countermeasure network based on a feature vector and SWGAN-GP, belonging to the technical field of ferrous metallurgy continuous casting detection. According to the method, the characteristic vectors containing static and dynamic characteristics of the bonding area are constructed through the visualized thermograph of the temperature rate of the crystallizer copper plate, and the characteristic vectors are classified through the discrimination model of the countermeasure network generated by SWGAN-GP, so that the detection and the forecast of the breakout of the crystallizer are realized. The method is used for detecting and forecasting the breakout of the crystallizer in real time based on the SWGAN-GP model, so that the false alarm rate can be obviously reduced and the forecasting accuracy rate can be effectively improved on the premise of ensuring that the breakout is completely reported.

Description

Crystallizer breakout prediction method for generating countermeasure network based on feature vector and SWGAN-GP
Technical Field
The invention belongs to the technical field of ferrous metallurgy continuous casting detection, and relates to a crystallizer bleed-out forecasting method for generating a countermeasure network based on a feature vector and SWGAN-GP.
Background
In the continuous casting process, the non-uniformly solidified primary blank shell in the crystallizer cannot bear the dual functions of the static pressure of molten steel and the withdrawal force, and the weak part is easy to break so as to form steel leakage. Breakout is a major safety accident in continuous casting production, which not only endangers personal safety and damages equipment, but also causes forced interruption of production and influences the yield and the product quality of a casting machine. With the continuous development and progress of the continuous casting technology, the occurrence probability of bleed-out can be effectively reduced by standardizing operation and maintaining the good running state of equipment. The breakout is closely related to a plurality of factors such as covering slag, drawing speed, liquid level fluctuation, heat flux density and the like, and although metallurgical workers and scholars at home and abroad carry out extensive research on the formation reason, the breakout is difficult to be thoroughly avoided. Therefore, the reduction and prevention of breakout accidents are always the key points of attention of metallurgical workers at home and abroad, and the detection of the breakout of the crystallizer is the core of the abnormal prediction in the continuous casting process, so that the method has important significance.
The invention patent 200710093907.7 discloses a continuous casting breakout prediction method, which predicts the occurrence of bonding breakout based on logic judgment and according to the temperature change condition of a thermocouple of a crystallizer. The method mainly comprises the following steps: capturing typical temperature characteristics, determining breakout probability and controlling casting speed. The method effectively solves the technical problems that interference factors in the existing breakout prediction method are not considered comprehensively, and small-range adhesion and slag entrapment cannot be reported timely. However, the breakout prediction model based on logic judgment has high dependence on equipment parameters, process conditions and physical parameters, and needs frequent adjustment of threshold values and parameters, resulting in poor robustness of the prediction algorithm.
The invention patent 201010207115.X discloses a continuous casting breakout prediction method, which uses a genetic algorithm to initialize a time sequence neural network breakout prediction model of a single thermocouple. The method mainly comprises the following steps: thermocouple temperature data are collected on line, data are preprocessed, and a model is forecasted to forecast breakout. However, the breakout prediction model based on the single neural network has strict limitations on the quality and quantity of training samples, and the process of making the samples is cumbersome, and the practicability and the applicability are low.
In view of the defects of the existing breakout prediction method and the complexity of the prediction algorithm and the complexity of the sample making process, the invention provides a method for detecting and predicting breakout of a crystallizer in real time by extracting visual characteristic vectors of a bonding area based on a temperature rate thermograph of a copper plate of the crystallizer, generating a countermeasure network model by combining SWGAN-GP, and training breakout and non-breakout samples of the crystallizer.
Disclosure of Invention
The invention aims to provide a crystallizer bleed-out forecasting method for generating a countermeasure network based on a characteristic vector and SWGAN-GP, which can timely and accurately detect and forecast the bonded bleed-out and provides a reliable means for monitoring the abnormity of the continuous casting process.
In order to achieve the purpose, the technical scheme of the invention is as follows:
a crystallizer breakout prediction method for generating a countermeasure network based on a feature vector and SWGAN-GP is characterized in that visual feature vectors are extracted from an abnormal region of a crystallizer copper plate temperature rate, and the countermeasure network is generated by the SWGAN-GP to classify the feature vectors, so that the crystallizer breakout is detected and predicted, and the method specifically comprises the following steps:
first, extracting the characteristics of the abnormal area of the temperature rate of the copper plate of the crystallizer
(1) 3 rows of 19 rows of thermocouples are arranged on the wide-surface copper plates of the inner arc and the outer arc of the crystallizer, and 3 rows of 1 row of thermocouples are arranged on the narrow-surface copper plates at the left side and the right side. And detecting the temperature of the thermocouple of the crystallizer copper plate on line, and calculating the temperature value of the crystallizer copper plate at the position of the non-thermocouple measuring point by an interpolation algorithm.
(2) And calculating the temperature change rate of each point of the copper plate through an interframe difference algorithm, and mapping the temperature rate of the copper plate to a two-dimensional plane by using computer graphics and OpenGL technology to obtain a two-dimensional temperature rate thermal image corresponding to the temperature of the copper plate.
(3) After counting and summarizing temperature rate data of a plurality of breakout samples, setting a temperature rate threshold value as TzUsing threshold segmentationThe method comprises the steps of removing a normal temperature fluctuation area with the temperature rate smaller than a threshold value from a two-dimensional temperature rate thermal image, and performing connectivity search on temperature rate abnormal points by using a run-length recursive algorithm to obtain a temperature rate abnormal area.
(4) Extracting the height H, width W, area S and transverse moving speed V of the abnormal temperature speed regionyLongitudinal moving speed VxAnd (6) visualizing the features.
Second step, abnormal region feature vector construction and processing
(1) Combining the abnormal region features extracted in the first step into a feature vector XBSimultaneously combining the normal working condition region features into a feature vector XN：
XB＝[HB,WB,SB,VBx,VBy]
XN＝[HN,WN,SN,VNx,VNy]
(2) Pulling speed V under combined abnormal and normal working conditionscAnd Vc', for feature vector XBAnd XNPerforming continuous processing to construct continuous feature vector ZBAnd ZN：
Figure BDA0002911153410000021
Figure BDA0002911153410000022
(3) For continuous feature vector ZBAnd ZNAnd (3) carrying out normalization treatment:
Figure BDA0002911153410000023
in the formula, Zmin、ZmaxRespectively representing the minimum and maximum values of the continuous-type eigenvector Z, FiNormalized pair of ith dimension feature representing continuous feature vector ZThe corresponding values.
(4) Respectively obtaining m abnormal area bleed-out feature vector samples F according to the feature vector construction and processing modes of the steps (1), (2) and (3)BAnd n normal working condition area non-breakout eigenvector samples FNAnd establishing a sample set Q:
Q＝{(FB1,1),(FB2,1),…,(FBm,1),(FN1,0),(FN2,0),…,(FNn,0)}
in the formula, m and n respectively represent the number of the breakout sample and the non-breakout sample. 1 and 0 represent class labels for the breakout and non-breakout specimens, respectively, as (1, 0, 0) and (0, 1, 0) in the form of one-hot codes.
Thirdly, constructing an SWGAN-GP generation confrontation network model
(1) Constructing a generative model G comprising 1 noise input layer and 2 full-connection layer neural networks, wherein the concrete structure of the generative model G sequentially comprises the following steps: noise input layer → first fully connected layer → second fully connected layer. The input noise generates pseudo samples with the same dimension as the original feature vector after G.
(2) Constructing a discrimination model D containing 1 sample input layer and 3 full-connection layer neural networks, wherein the specific structure of the discrimination model D is as follows in sequence: sample input layer → first fully connected layer → parallel branch second fully connected layer → parallel branch third fully connected layer. And finally D, judging the authenticity of the input sample and classifying the sample.
(3) And combining the generation model G with the discrimination model D to construct a SWGAN-GP generation confrontation network model.
Fourthly, training SWGAN-GP to generate a confrontation network model
(1) And (3) training the discrimination model D constructed in the step (2) in the third step. And freezing the generated model when the discriminant model is trained, namely setting the parameters of the generated model to be not updatable.
1.1) randomly obtaining a breakout sample F from a sample set QBAnd non-breakout specimen FNEach train _ samples has a single sample denoted x, and x belongs to the real sample set PrI.e. x to Pr(ii) a Simultaneous acquisition of Gaussian distributed noiseSet of samples z, i.e. z-Pz。
1.2) inputting the noise sample z into the generation model G constructed in the step (1) in the third step to generate a pseudo sample
Figure BDA00029111534100000311
The label of (a) is represented in a one-hot coded form as (0, 0, 1):
Figure BDA0002911153410000031
in the formula, a pseudo sample
Figure BDA0002911153410000032
Belonging to a set of dummy samples PGI.e. by
Figure BDA0002911153410000033
1.3) obtaining PrSample x and P in (1)GSample of (1)
Figure BDA0002911153410000034
And interpolation is carried out to obtain a new sample
Figure BDA0002911153410000035
Figure BDA0002911153410000036
In which ε follows a uniform distribution over [0,1 ].
1.4) calculating the Total loss L of the discriminant model DD：
Figure BDA0002911153410000037
In the formula, m is the batch sample number batch _ size, i is 1, and 2 … m is the sample number index.
Figure BDA0002911153410000038
D (x) are respectively
Figure BDA0002911153410000039
x corresponds to the output value of the discriminator D. λ represents a gradient penalty coefficient.
Figure BDA00029111534100000310
Is a gradient penalty term. C denotes the number of sample classes, j is 1, and 2 … C is the class index. y isjFor the true label corresponding to the jth category, fj(x) In order for the arbiter to predict the value for that sample,
Figure BDA0002911153410000041
representing the loss of cross-entropy part of the multivariate classification.
(2) Training the generative model G constructed in the step (1) in the third step. And (4) freezing the discrimination model when the generation model is trained, namely setting the parameters of the discrimination model to be not updatable.
2.1) obtaining a set of noise samples z' -P obeying Gaussian distributionz′。
2.2) inputting the noise sample z' into the generation model G constructed in the step (1) in the third step to generate a pseudo sample
Figure BDA0002911153410000044
Then will be
Figure BDA0002911153410000045
Calculating the model loss L of the discriminant model DG：
Figure BDA0002911153410000042
(3) Training discriminant model n in each round of trainingcriticThen, generating a model ngenRepeating the training to judge model D and generate model G, and observing LGAnd LDAs a function of the number of training rounds, up to LGAnd LDThe loss curve gradually flattens and fluctuates steadily. At this time, the generative model G and the discriminant model D can be determinedNash balance is achieved, and the training is finished.
Fifthly, detecting and forecasting bleed-out on line based on SWGAN-GP model
(1) Extracting typical visual characteristics of the crystallizer copper plate temperature rate abnormal area in real time, preprocessing the typical visual characteristics, and constructing to obtain an abnormal area characteristic vector Ffv；
(2) Abnormal region feature vector FfvInputting the data into a discrimination model D of the SWGAN-GP generated countermeasure network to obtain a predicted value y of the model:
y＝D(Ffv)
(3) and forecasting the crystallizer bleed-out according to the output result y of the discrimination model.
Expressing y as (y) in a form of one-hot coding1，y2，y3): if y1＝max(y1,y2,y3) If the data label corresponding to y is (1, 0, 0), the alarm is given and the casting machine pulling speed is rapidly reduced; if y2＝max(y1,y2,y3) And if the detected steel leakage is normal, continuing to detect and forecast the steel leakage at the next moment corresponding to the data label with the value of y being (0, 1, 0). Feature vector x of abnormal regionfvAll derived from real samples detected on-line rather than pseudo-samples generated by generative model G
Figure BDA0002911153410000043
So that y cannot occur3The case of the maximum value.
The method for forecasting the breakout is suitable for forecasting the breakout of continuous casting billets such as plate blanks, square blanks, round blanks, special blanks and the like.
The invention has the beneficial effects that: the method constructs a characteristic vector containing static and dynamic characteristics of a bonding area through a visualized thermograph of the temperature rate of the crystallizer copper plate, and classifies the characteristic vector through a discrimination model of a countermeasure network generated by SWGAN-GP so as to realize detection and forecast of the breakout of the crystallizer. The method is used for detecting and forecasting the crystallizer bleed-out in real time based on the SWGAN-GP model, so that the false alarm rate can be obviously reduced and the forecasting accuracy rate can be effectively improved on the premise of ensuring that the bleed-out is completely reported.
Drawings
FIG. 1 is a flow of a crystallizer breakout prediction method.
Fig. 2 is a schematic diagram of arrangement of a thermocouple of a copper plate of the crystallizer.
Fig. 3 is a temperature rate abnormal region visualization characteristic diagram. FIG. 3(a) is a diagram of initial formation of bonding; FIG. 3(b) is a transverse propagation diagram of the bond region; FIG. 3(c) is a longitudinal propagation diagram of the bond region; FIG. 3(d) is a V-shaped characteristic diagram of the bonded breakout.
FIG. 4 is a normal operating condition region visualization characteristic diagram. FIG. 4(a) is a diagram of initial formation of bonding; FIG. 4(b) is a longitudinal contraction of the bonded area; FIG. 4(c) is an in-situ expansion diagram of the bonding region; FIG. 4(d) is a transverse contraction diagram of the bonded area.
FIG. 5 is SWGAN-GP generative model G.
FIG. 6 is a SWGAN-GP discriminant model D.
FIG. 7 is a visual characteristic diagram of an online detection temperature rate abnormal region. FIG. 7(a) is a V-shaped characteristic diagram of the bonded breakout; FIG. 7(b) is a normal condition bonding area diagram.
In the figure: 1, a thermocouple; 2 outer arc wide copper plate; 3, a left narrow-face copper plate; 4, a right narrow-face copper plate; 5 inner arc wide copper plate.
Detailed Description
The invention will be further elucidated by means of specific embodiments, in conjunction with the drawing
Fig. 1 shows a flow chart of a method for predicting breakout of a crystallizer. Firstly, extracting visual characteristics of a crystallizer copper plate temperature rate abnormal area and preprocessing the visual characteristics to construct and obtain a five-dimensional characteristic vector; secondly, constructing and training a SWGAN-GP model; and finally, classifying the feature vectors and forecasting the breakout through a discrimination model of SWGAN-GP.
First step, visualization of temperature rate of crystallizer copper plate and extraction of abnormal area characteristics
(1) Fig. 2 shows the distribution diagram of the copper plate of the mold and its thermocouple. The crystallizer is formed by combining four copper plates, the total height is 900mm, and the effective height during casting is 800 mm. 3 rows of 19 rows of thermocouples 1 are arranged on the inner and outer arc wide- surface copper plates 5 and 2 of the crystallizer, 3 rows of 1 row of thermocouples are arranged on the left and right narrow-surface copper plates 3 and 4, and the total number of the thermocouples is 120. The distance between the upper openings of the first row of thermocouple data crystallizers is 210mm, the distance between the first row of thermocouple data crystallizers and the second row of thermocouple data crystallizers is 115mm, the distance between the second row of thermocouple data crystallizers and the third row of thermocouple data crystallizers is 120mm, and the distance between two adjacent rows of thermocouples is 150 mm. And (3) detecting the temperature of all the thermocouples (1) of the crystallizer copper plate on line, and calculating the temperature value of the crystallizer copper plate at the position of the non-thermocouple measuring point by an interpolation algorithm.
(2) And calculating the temperature change rate of each point of the copper plate through an interframe difference algorithm, and mapping the temperature rate of the copper plate to a two-dimensional plane by using computer graphics and OpenGL technology to obtain a two-dimensional temperature rate thermal image corresponding to the temperature of the copper plate.
(3) Setting a temperature rate threshold value to be 0.3 ℃/s, removing a normal temperature fluctuation area with the temperature rate smaller than the threshold value from the two-dimensional temperature rate thermal image by using a threshold segmentation algorithm, and performing connectivity search on temperature rate abnormal points by using a run recursion algorithm to obtain a temperature rate abnormal area.
(4) Fig. 3 is a graph showing the visualization characteristic of the abnormal temperature rate region. The simulation interval from the first row to the third row along the casting direction is distributed to 100 pixel points; the horizontal simulation interval is from the first row to the nineteenth row of thermocouples distributed to 300 pixel points. T is1～T4Representing 4 moments in time corresponding to the bond region from initial formation to the appearance of a distinct "V" shaped feature, each moment being 3s apart. Extraction of T4Height H of time anomaly regionB20.45cm wide WB56.11cm, shaded area SB＝607.08cm2And calculating to obtain the transverse movement velocity V according to the change of the barycentric coordinates of the abnormal region along with the timeByLongitudinal moving speed V of-0.06 m/minBx＝0.25m/min。
Fig. 4 is a normal condition area visualization characteristic diagram. The visual characteristics of the region, height H, can be obtained by the same methodN5.64cm, width WN21.5cm, shaded area SN＝92.36cm2Transverse moving velocity VNy0m/min, longitudinal moving speed VNx＝-0.39m/min。
Second step, abnormal region feature vector construction and processing
(1) Combining the abnormal and normal working condition region characteristics extracted in the first step into a characteristic vector XBAnd XN：
XB＝[HB,WB,SB,VBx,VBy]＝[20.45,56.11,607.08,0.25,-0.06]
XN＝[HN,WN,SN,VNx,VNy]＝[5.64,21.5,92.36,-0.39,0]
(2) Pulling speed V under combined abnormal and normal working conditionsc0.9m/min and Vc' 0.65m/min, for feature vector XBAnd XNPerforming continuous processing to construct continuous feature vector ZBAnd ZN：
Figure BDA0002911153410000061
Figure BDA0002911153410000062
(3) For continuous feature vector ZBAnd ZNAnd (3) carrying out normalization treatment:
Figure BDA0002911153410000063
in the formula, Zmin、ZmaxRespectively representing the minimum and maximum values of the continuous-type eigenvector Z, FiAnd representing the corresponding numerical value after the ith dimension feature normalization of the continuous feature vector Z. Normalized feature vector FB＝[0.84,0.71,0.86,0.73,0.13]，FN＝[0.19,0.18,0.02,0,0.25]。
Respectively obtaining 50 bleed-out eigenvector samples F according to the above eigenvector construction and processing modeBAnd 50 non-breakout eigenvector samples FNAnd establishing a sample set Q:
Q＝{(FB1,1),(FB2,1),…,(FB50,1),(FN1,0),(FN2,0),…,(FN50,0)}
in the formula, 1 and 0 represent class labels of the breakout sample and the non-breakout sample, respectively, and are expressed as (1, 0, 0) and (0, 1, 0) in the form of one-hot codes.
Thirdly, constructing an SWGAN-GP generation confrontation network model
(1) As shown in fig. 5, a generative model including a noise input layer and a fully connected layer neural network is constructed for generating a pseudo sample, denoted as generative model G.
(2) As shown in fig. 6, a discrimination model including a sample input layer and a full connection layer neural network is constructed, and is used for discriminating an authentic sample and classifying the sample, and the discrimination model is recorded as a discrimination model D.
(3) And combining the generation model G with the discrimination model D to construct a SWGAN-GP generation confrontation network model.
Fourthly, training SWGAN-GP to generate a confrontation network model
(1) And (5) training a discrimination model. And freezing the generated model when the discriminant model is trained, namely setting the parameters of the generated model to be not updatable.
1.1) randomly obtaining a breakout sample F from a sample set QBAnd non-breakout specimen FNEach 30, a single sample is denoted x, and x belongs to the set of true samples PrI.e. x to Pr(ii) a Simultaneously obtaining a set of noise samples z, i.e. z-P, which obey a Gaussian distributionz。
1.2) input of noise samples z into the generative model G to generate pseudo samples
Figure BDA0002911153410000071
The label of (a) is represented in a one-hot coded form as (0, 0, 1):
Figure BDA0002911153410000072
in the formula, a pseudo sample
Figure BDA0002911153410000073
Belonging to a set of dummy samples PGI.e. by
Figure BDA0002911153410000074
1.3) obtaining PrSample x and P in (1)GSample of (1)
Figure BDA0002911153410000075
And interpolation is carried out to obtain a new sample set
Figure BDA0002911153410000076
Figure BDA0002911153410000077
In which ε follows a uniform distribution over [0,1 ].
1.4) calculating the Total loss L of the discriminant model DD：
Figure BDA0002911153410000078
In the formula, m is the batch sample number 32, i is 1, and 2 … m is the sample number index.
Figure BDA0002911153410000079
D (x) are respectively
Figure BDA00029111534100000710
x corresponds to the output value of the discriminator D. The gradient penalty factor lambda is 10,
Figure BDA00029111534100000711
is a gradient penalty term. The number of sample types C is 3, and represents the breakout sample FBNon-bleed-out sample FNAnd a dummy sample
Figure BDA00029111534100000712
j is 1,2, 3 is a category index. y isjFor the true label corresponding to the jth category, fj(x) In order for the arbiter to predict the value for that sample,
Figure BDA00029111534100000713
representing the loss of cross-entropy part of the multivariate classification.
(2) And training the generated model. And (4) freezing the discrimination model when the generation model is trained, namely setting the parameters of the discrimination model to be not updatable.
2.1) obtaining a set of noise samples z' -P obeying Gaussian distributionz′。
2.2) inputting the noise sample z' into the generative model G to generate a pseudo sample
Figure BDA00029111534100000714
Then will be
Figure BDA00029111534100000715
Calculating the model loss L of the discriminant model DG：
Figure BDA00029111534100000716
Training the discriminant model 5 times and generating the model 1 time in each round of training, repeating the training of discriminant model D and generating model G, and observing LGAnd LDWhen the number of training rounds reaches 20000 times as the number of training rounds changes, L is observedGAnd LDThe loss curve gradually flattens and fluctuates steadily. At this time, it can be judged that nash balance is achieved between the generated model G and the discriminant model D, and the training is completed.
Fifthly, detecting and forecasting bleed-out on line based on SWGAN-GP model
(1) Extracting typical visual characteristics of the crystallizer copper plate temperature rate abnormal area in real time, as shown in fig. 7, preprocessing the typical visual characteristics, and constructing to obtain an abnormal area characteristic vector:
Ffv1＝[0.78,0.30,0.46,0.84,0.73]；Ffv2＝[0.06,0.23,0.08,0.63,0.56]
(2) will be abnormalRegion feature vector Ffv1And Ffv2Inputting the predicted value y of the model into a discrimination model D of the SWGAN-GP generated countermeasure network1And y2：
y1＝D(Ffv1)＝D([0.78,0.30,0.46,0.84,0.73])＝(0.81,0.19,0)
y2＝D(Ffv2)＝D([0.06,0.23,0.08,0.63,0.56])＝(0.12,0.88,0)
(3) And forecasting the crystallizer bleed-out according to the output result y of the discrimination model. y is1If the data label corresponding to y is (1, 0, 0) is breakout, (0.81,0.19,0) gives an alarm and rapidly reduces the casting machine pulling speed; y is2And if the value is equal to (0.12,0.88 and 0), the data label corresponding to (0, 1 and 0) is a normal working condition, and the steel leakage detection and prediction at the next moment are continued. Feature vector x of abnormal regionfvAll derived from real samples detected on-line rather than pseudo-samples generated by generative model G
Figure BDA0002911153410000081
So that y cannot occur3The case of the maximum value.
The above-mentioned embodiments only express the embodiments of the present invention, but not should be understood as the limitation of the scope of the invention patent, it should be noted that, for those skilled in the art, many variations and modifications can be made without departing from the concept of the present invention, and these all fall into the protection scope of the present invention.

Claims (4)

1. A crystallizer breakout prediction method for generating a countermeasure network based on a feature vector and SWGAN-GP is characterized in that the method extracts a visual feature vector from an abnormal region of a crystallizer copper plate temperature rate and classifies the feature vector by utilizing the SWGAN-GP to generate the countermeasure network, so as to detect and predict the crystallizer breakout, and comprises the following steps:
first, extracting the characteristics of the abnormal area of the temperature rate of the copper plate of the crystallizer
(1) Thermocouples are arranged on the wide-surface copper plates of the inner arc and the outer arc of the crystallizer and the narrow-surface copper plates at the left side and the right side; detecting the temperature of the thermocouple of the crystallizer copper plate on line, and calculating the temperature value of the crystallizer copper plate at the position of a non-thermocouple measuring point by an interpolation algorithm;
(2) calculating the temperature change rate of each point of the copper plate through an interframe difference algorithm, and mapping the temperature rate of the copper plate to a two-dimensional plane to obtain a two-dimensional temperature rate thermograph corresponding to the temperature of the copper plate;
(3) after counting and summarizing temperature rate data of a plurality of breakout samples, setting a temperature rate threshold value as TzRemoving a normal temperature fluctuation region with the temperature rate smaller than a threshold value from the two-dimensional temperature rate thermal image by using a threshold segmentation algorithm, and performing connectivity search on temperature rate abnormal points by using a run-length recursion algorithm to obtain a temperature rate abnormal region;
(4) extracting the height H, width W, area S and transverse moving speed V of the abnormal temperature speed regionyLongitudinal moving speed VxVisualization features;
second step, abnormal region feature vector construction and processing
(1) Combining the abnormal region features extracted in the first step into a feature vector XBSimultaneously combining the normal working condition region features into a feature vector XN：
XB＝[HB,WB,SB,VBx,VBy]
XN＝[HN,WN,SN,VNx,VNy]
(2) Pulling speed V under combined abnormal and normal working conditionscAnd Vc', for feature vector XBAnd XNPerforming continuous processing to construct continuous feature vector ZBAnd ZN：
Figure FDA0002911153400000011
Figure FDA0002911153400000012
(3) For continuous feature vector ZBAnd ZNAnd (3) carrying out normalization treatment:
Figure FDA0002911153400000013
in the formula, Zmin、ZmaxRespectively representing the minimum and maximum values of the continuous-type eigenvector Z, FiRepresenting a corresponding numerical value after the ith dimension characteristic normalization of the continuous characteristic vector Z;
(4) respectively obtaining m abnormal region bleed-out feature vector samples F according to the feature vector structures and the processing modes of (1), (2) and (3)BAnd n normal working condition area non-breakout eigenvector samples FNAnd establishing a sample set Q:
Q＝{(FB1,1),(FB2,1),…,(FBm,1),(FN1,0),(FN2,0),…,(FNn,0)}
in the formula, m and n respectively represent the number of bleed-out samples and non-bleed-out samples; 1 and 0 represent class labels of the breakout sample and the non-breakout sample, respectively, and are expressed as (1, 0, 0) and (0, 1, 0) in the form of one-hot codes;
thirdly, constructing an SWGAN-GP generation confrontation network model
(1) Constructing a generation model G containing 1 noise input layer and 2 full-connection layer neural networks, wherein the input noise generates a pseudo sample with the same dimension as the original characteristic vector after passing through G;
(2) constructing a discrimination model D comprising 1 sample input layer and 3 full-connection layer neural networks, discriminating the authenticity of the input sample through D and classifying the sample;
(3) combining the generation model G with the discrimination model D to construct an SWGAN-GP generation confrontation network model;
fourthly, training SWGAN-GP to generate a confrontation network model
(1) Training a discrimination model D constructed in the step (2) in the third step; freezing the generated model when training the discrimination model, namely setting the parameters of the generated model to be not updatable;
1.1) sampling fromRandomly obtaining breakout sample F in Q setBAnd non-breakout specimen FNEach train _ samples has a single sample denoted x, and x belongs to the real sample set PrI.e. x to Pr(ii) a Simultaneously obtaining a set of noise samples z, i.e. z-P, which obey a Gaussian distributionz；
1.2) inputting the noise sample z into the generation model G constructed in the step (1) in the third step to generate a pseudo sample
Figure FDA0002911153400000021
Figure FDA0002911153400000022
The label of (a) is represented in a one-hot coded form as (0, 0, 1):
Figure FDA0002911153400000023
in the formula, a pseudo sample
Figure FDA0002911153400000024
Belonging to a set of dummy samples PGI.e. by
Figure FDA0002911153400000025
1.3) obtaining PrSample x and P in (1)GSample of (1)
Figure FDA0002911153400000026
And interpolation is carried out to obtain a new sample
Figure FDA0002911153400000027
Figure FDA0002911153400000028
Wherein ε follows a uniform distribution over [0,1 ];
1.4) calculating the Total loss L of the discriminant model DD：
Figure FDA0002911153400000029
In the formula, m is batch sample number batch _ size, i is 1, and 2 … m is a sample number index;
Figure FDA00029111534000000210
d (x) are respectively
Figure FDA00029111534000000211
x is the output value of the discriminator D; λ represents a gradient penalty coefficient;
Figure FDA00029111534000000212
is a gradient penalty term; c represents the number of sample categories, j is 1, and 2 … C is a category index; y isjFor the true label corresponding to the jth category, fj(x) In order for the arbiter to predict the value for that sample,
Figure FDA00029111534000000213
represents the loss of cross-entropy part of the multivariate classification;
(2) training a generating model G constructed in the step (1) in the third step; freezing the discrimination model when training the generated model, namely setting the parameters of the discrimination model to be not updatable;
2.1) obtaining a set of noise samples z' -P obeying Gaussian distributionz′；
2.2) inputting the noise sample z' into the generation model G constructed in the step (1) in the third step to generate a pseudo sample
Figure FDA0002911153400000031
Then will be
Figure FDA0002911153400000032
Calculating the model loss L of the discriminant model DG：
Figure FDA0002911153400000033
(3) Training discriminant model n in each round of trainingcriticThen, generating a model ngenRepeating the training to judge model D and generate model G, and observing LGAnd LDAs a function of the number of training rounds, up to LGAnd LDThe loss curve gradually becomes gentle and fluctuates in a steady state; at this time, it can be judged that Nash balance is achieved between the generated model G and the discrimination model D, and the training is finished;
fifthly, detecting and forecasting bleed-out on line based on SWGAN-GP model
(1) Extracting typical visual characteristics of the crystallizer copper plate temperature rate abnormal area in real time, preprocessing the typical visual characteristics, and constructing to obtain an abnormal area characteristic vector Ffv；
(2) Abnormal region feature vector FfvInputting the data into a discrimination model D of the SWGAN-GP generated countermeasure network to obtain a predicted value y of the model:
y＝D(Ffv)
(3) forecasting crystallizer steel leakage according to the output result y of the discrimination model;
expressing y as (y) in a form of one-hot coding1，y2，y3): if y1＝max(y1,y2,y3) If the data label corresponding to y is (1, 0, 0), the alarm is given and the casting machine pulling speed is rapidly reduced; if y2＝max(y1,y2,y3) If the detected steel leakage is normal, corresponding to the data label with y being (0, 1, 0), continuing to detect and forecast the steel leakage at the next moment; feature vector x of abnormal regionfvAll derived from real samples detected on-line rather than pseudo-samples generated by generative model G
Figure FDA0002911153400000034
Therefore, y cannot appear3The case of the maximum value.
2. The method for predicting the breakout of a crystallizer according to claim 1, wherein said method is suitable for the breakout prediction of slabs, billets, round billets, beam billets or other continuous slabs.
3. The crystallizer breakout prediction method according to claim 1, wherein the concrete structure of the model G generated in the third step (1) is as follows: noise input layer → first fully connected layer → second fully connected layer.
4. The crystallizer breakout prediction method according to claim 1, wherein the specific structure of the discrimination model D in the third step (2) is as follows: sample input layer → first fully connected layer → parallel branch second fully connected layer → parallel branch third fully connected layer.
CN202110087095.5A 2021-01-22 2021-01-22 Crystallizer breakout prediction method for generating countermeasure network based on feature vector and SWGAN-GP Active CN112926622B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110087095.5A CN112926622B (en) 2021-01-22 2021-01-22 Crystallizer breakout prediction method for generating countermeasure network based on feature vector and SWGAN-GP

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110087095.5A CN112926622B (en) 2021-01-22 2021-01-22 Crystallizer breakout prediction method for generating countermeasure network based on feature vector and SWGAN-GP

Publications (2)

Publication Number Publication Date
CN112926622A true CN112926622A (en) 2021-06-08
CN112926622B CN112926622B (en) 2022-10-18

Family

ID=76164701

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110087095.5A Active CN112926622B (en) 2021-01-22 2021-01-22 Crystallizer breakout prediction method for generating countermeasure network based on feature vector and SWGAN-GP

Country Status (1)

Country Link
CN (1) CN112926622B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113814369A (en) * 2021-09-28 2021-12-21 大连理工大学 Crystallizer bleed-out forecasting method based on eigenvector and Adaboost integrated model
CN114850431A (en) * 2022-07-05 2022-08-05 北京科技大学 Method for forecasting bleed-out of continuous casting crystallizer

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105911095A (en) * 2016-05-04 2016-08-31 东北电力大学 Visual recognition method of continuous casting billet surface longitudinal cracks
CN106980729A (en) * 2015-07-24 2017-07-25 安徽工业大学 A kind of continuous casting breakout prediction method based on mixed model

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106980729A (en) * 2015-07-24 2017-07-25 安徽工业大学 A kind of continuous casting breakout prediction method based on mixed model
CN105911095A (en) * 2016-05-04 2016-08-31 东北电力大学 Visual recognition method of continuous casting billet surface longitudinal cracks

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
刘宇等: "基于图像处理的黏结漏钢可视化检测方法", 《中南大学学报(自然科学版)》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113814369A (en) * 2021-09-28 2021-12-21 大连理工大学 Crystallizer bleed-out forecasting method based on eigenvector and Adaboost integrated model
CN113814369B (en) * 2021-09-28 2022-11-18 大连理工大学 Crystallizer bleed-out forecasting method based on eigenvector and Adaboost integrated model
CN114850431A (en) * 2022-07-05 2022-08-05 北京科技大学 Method for forecasting bleed-out of continuous casting crystallizer
CN114850431B (en) * 2022-07-05 2022-10-21 北京科技大学 Method for forecasting bleed-out of continuous casting crystallizer

Also Published As

Publication number Publication date
CN112926622B (en) 2022-10-18

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
CN105328155B (en) A kind of bleed-out visual feature forecasting procedure based on improvement neutral net
CN109396375B (en) A kind of crystallizer bleedout prediction electric thermo method based on feature vector and hierarchical clustering
CN112926622B (en) Crystallizer breakout prediction method for generating countermeasure network based on feature vector and SWGAN-GP
CN102886504B (en) Crystallizer bleed-out visual forecasting method based on machine vision
CN105911095A (en) Visual recognition method of continuous casting billet surface longitudinal cracks
CN104181196B (en) A kind of continuous casting billet surface longitudinal crack online test method
CN108436050B (en) A method of continuous cast mold bleed-out is forecast using space density clustering DBSCAN
CN108580827B (en) A method of Crystallizer bleed-out is forecast based on Agglomerative Hierarchical Clustering
Liu et al. Computer vision detection of mold breakout in slab continuous casting using an optimized neural network
Zhou et al. Application of time series data anomaly detection based on deep learning in continuous casting process
CN108705058B (en) A method of forecast Crystallizer bleed-out is clustered based on K-Means
CN113340444A (en) Continuous casting billet surface longitudinal crack prediction method based on thermocouple temperature and logic judgment
Kumar et al. Development of intelligent mould for online detection of defects in steel billets
CN113814369B (en) Crystallizer bleed-out forecasting method based on eigenvector and Adaboost integrated model
CN111666710B (en) Method for predicting continuous casting billet longitudinal cracks by logistic regression classification
Zhang et al. Breakout prediction for continuous casting using genetic algorithm-based back propagation neural network model
CN108469313A (en) Copper plate of crystallizer temperature anomaly method for detecting area based on cellular automata
CN113935416B (en) Continuous casting breakout prediction method of multi-state crystallizer
CN117036797A (en) Continuous casting billet longitudinal crack prediction method based on feature extraction and random forest classification
CN115446276A (en) Continuous casting breakout early warning method for recognizing V-shaped bonding characteristics of crystallizer copper plate based on convolutional neural network
CN111680448B (en) Continuous casting billet longitudinal crack prediction method based on SVM classification
CN113505785A (en) Crystallizer steel leakage forecasting method based on eigenvector and logistic regression model
JPH09248661A (en) Method for predicting breakout caused by foreign matter biten in continuous casting
JPH0722811B2 (en) Constrained breakout prediction method for continuous casting

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