CN111666710A - Method for predicting continuous casting billet longitudinal cracks by logistic regression classification - Google Patents
Method for predicting continuous casting billet longitudinal cracks by logistic regression classification Download PDFInfo
- Publication number
- CN111666710A CN111666710A CN202010385010.7A CN202010385010A CN111666710A CN 111666710 A CN111666710 A CN 111666710A CN 202010385010 A CN202010385010 A CN 202010385010A CN 111666710 A CN111666710 A CN 111666710A
- Authority
- CN
- China
- Prior art keywords
- temperature
- sample
- logistic regression
- time
- classification
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B22—CASTING; POWDER METALLURGY
- B22D—CASTING OF METALS; CASTING OF OTHER SUBSTANCES BY THE SAME PROCESSES OR DEVICES
- B22D2/00—Arrangement of indicating or measuring devices, e.g. for temperature or viscosity of the fused mass
- B22D2/006—Arrangement of indicating or measuring devices, e.g. for temperature or viscosity of the fused mass for the temperature of the molten metal
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Mechanical Engineering (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Investigating Or Analyzing Materials Using Thermal Means (AREA)
Abstract
A method for predicting continuous casting billet longitudinal cracks by logistic regression classification belongs to the technical field of ferrous metallurgy continuous casting detection. The method comprises the steps of carrying out time sequence processing on temperature data under longitudinal cracks and normal working conditions to obtain a temperature time sequence sample library; training and testing the temperature time sequence sample base by using a logistic regression classification algorithm to obtain an optimal classification decision function corresponding to logistic regression classification; and predicting the real-time temperature data measured on line by using the decision function, and judging whether the real-time temperature data belongs to the longitudinal cracks or not. The method extracts and fuses the temperature and the change rate characteristics of the original temperature data from the dual angles of time and space, trains and tests the temperature time sequence sample base by using the logistic regression classification model, finally obtains the optimal logistic regression classification decision function to predict the online real-time temperature, has the advantages of strong real-time performance and high detection efficiency, and can greatly improve the longitudinal crack identification efficiency and accuracy of field operators.
Description
Technical Field
The invention belongs to the technical field of ferrous metallurgy continuous casting detection, and relates to a method for predicting continuous casting billet longitudinal cracks by logistic regression classification.
Background
The main reason for generating the longitudinal cracks on the surface of the casting blank is that the thickness of a primary blank shell is uneven, the stress is concentrated at the weak part of the blank shell, when the stress exceeds the tensile strength of the blank shell, a crack source is generated, and then the longitudinal cracks are further expanded in a crystallizer and a secondary cooling area to form the longitudinal cracks. In order to prevent longitudinal cracking and improve the quality of casting blanks, proper factors such as crystallizer taper, casting powder, cooling strength, pulling speed, molten steel superheat degree, narrow-surface-to-wide-surface heat flow ratio and the like are selected to ensure stable and uniform heat transfer between a blank shell and a crystallizer as much as possible.
From the practical online detection result of the temperature of the crystallizer copper plate, the longitudinal crack has a spatial one-dimensional propagation characteristic, rarely expands along the transverse direction in the propagation process, and only shows the fluctuation process of descending, stabilizing and ascending in sequence of the temperature of the thermocouples in the same row. In addition, because the blank shell near the longitudinal crack is sunken to form a thicker air gap, the heat transfer between the blank shell and the crystallizer is greatly influenced, and the temperature of the blank shell and the temperature of the copper plate thermocouple are obviously lower than the temperature of the blank shell and the temperature of the copper plate thermocouple under normal working conditions. Therefore, the characteristics can be used as an important basis for detecting and identifying the longitudinal crack temperature abnormality of the continuous casting billet.
Aiming at the typical characteristics of the temperature of the copper plate when the surface longitudinal cracks appear on the continuous casting blank, the invention provides that the time sequence change trend and the characteristics of the longitudinal crack temperature are respectively extracted and fused from the time and space angles, the longitudinal crack temperature time sequence sample is modeled by means of a logistic regression classification algorithm, an optimal classification model and a classification decision function corresponding to the optimal classification model are obtained, and the longitudinal cracks of the casting blank are predicted on line on the basis.
Disclosure of Invention
The invention aims to provide a method for predicting the longitudinal cracks of a continuous casting blank in a classified manner by adopting logistic regression, which is used for carrying out time sequence processing on the longitudinal cracks and temperature data under normal working conditions to obtain a temperature time sequence sample library; training and testing the temperature time sequence sample base by using a logistic regression classification algorithm to obtain an optimal classification decision function corresponding to logistic regression classification; and predicting the real-time temperature data measured on line by using the decision function, and judging whether the real-time temperature data belongs to the longitudinal cracks or not.
In order to achieve the purpose, the technical scheme of the invention is as follows:
a method for predicting continuous casting billet longitudinal cracks by logistic regression classification comprises the following steps:
first step, temperature data acquisition
(1) For the longitudinal crack historical temperature sample, the same row of thermocouple temperature data of L seconds including the temperature falling-stabilizing-rising fluctuation process is intercepted.
(2) And intercepting the temperature data of the thermocouples in the same row for continuous L seconds for the temperature data under the normal working condition.
Second, temperature pretreatment
(1) And (3) performing difference processing on the temperatures of the thermocouples in the r th row and the r +1 th row of the thermocouples in the same row:
in the formula, T(r)i、T(r+1)iThe values of the thermocouple temperature at the ith moment in the r th row and the r +1 th row respectively;
is the difference value of the values of the thermocouple temperature at the ith moment in the r and r +1 rows.
(2) Solving the r and r +1 rows of thermocouple temperature difference TminusRate of change at intervals of k seconds:
in the formula (I), the compound is shown in the specification,
is the difference value of the values of the thermocouple temperature at the i + k th time in the r th and r +1 th rows.
Taking the rate of change T _ M _ V obtained by the thermocouple temperatures in the same column as a sample, and forming a temperature time sequence sample library in such a way:
ST_M_V＝{(T_M_VC1,1),(T_M_VC2,1),…,(T_M_VCm,1),(T_M_VN1,0),(T_M_VN2,0),…,(T_M_VNn0), where T _ M _ VCThe class label of (1) represents a longitudinal crack sample; t _ M _ VNThe class label of (1) is 0, indicating a normal operating condition sample. And m and n are the number of longitudinal cracks and normal working condition samples in the sample library respectively.
The number of the longitudinal crack samples and the number of the normal working condition samples in the temperature time sequence sample library are not less than 40.
Thirdly, obtaining an optimal classification decision function
Library S of temperature time series samplesT_M_VRandomly dividing the test sample into H groups of sub-sample libraries, wherein each group of sub-sample libraries comprises a training sample set and a test sample set, training the training sample set in each group of sub-sample libraries by using a logistic regression classification method, and then testing the test sample set in the sub-sample libraries by using the obtained model. The H groups of sub-sample libraries are respectively trained and tested to obtain H test accuracy rates, and a classification prediction model with the optimal test accuracy rate and a corresponding classification decision function are selected, wherein the method comprises the following steps:
(1) model training: at ST_M_VRandomly extracting more than 30 longitudinal crack samples T _ M _ VCAnd more than 30 normal working condition samples T _ M _ VNForming a training set, training the training set by using a logistic regression classification method to obtain a classification decision function corresponding to the logistic regression classification method, namely:
wherein the content of the first and second substances,
representing a temperature time series sample vector; omega is a weight vector corresponding to the time sequence sample, omegaTIs the transpose of ω, and b is the displacement term.
(2) And (3) testing a model: the fraction obtained in (1)Class decision function pair ST_M_VAnd (5) testing a test set consisting of the residual samples, and recording the prediction accuracy.
(3) And (3) repeating the steps (1) to (2) for H times, randomly extracting the longitudinal crack samples and the normal working condition samples again each time to form a new training sample set, and taking the residual samples as the test sample set to finally obtain H classification decision functions and the corresponding prediction accuracy rates thereof.
(4) And selecting the classification decision function with the highest prediction accuracy as the optimal logistic regression classification decision function.
Fourth step, longitudinal crack on-line prediction
(1) In the on-line detection process, the real-time temperature of the thermocouples in the same row at the current moment and in the previous continuous L-1 seconds is intercepted, and the real-time temperature is preprocessed in the same way as the second step to obtain a real-time temperature time sequence sample T _ M _ Vnew。
(2) Using the optimal logistic regression classification decision function obtained in the third step to perform real-time temperature time sequence sample T _ M _ VnewPredict the class label of (1):
(3) if the value of the decision function is greater than or equal to the threshold lambda, T _ M _ V is considerednewAnd (3) sending a longitudinal crack alarm for longitudinal cracks, otherwise, updating the time, and executing the fourth step (1), the fourth step (2) and the fourth step (3) on the real-time temperature of the same-row thermocouples continuously L-1 seconds before and at the next moment, namely the real-time temperature of the L seconds corresponding to the next moment, wherein the threshold lambda meets the condition lambda ∈ [0.5, 1).
The method is suitable for predicting the longitudinal cracks of the 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 for predicting the longitudinal cracks of the continuous casting billets by adopting the logistic regression classification extracts and fuses the temperature and the change rate characteristics of the original temperature data from the double angles of time and space, trains and tests a temperature time sequence sample library by utilizing the logistic regression classification model, finally obtains the optimal logistic regression classification decision function to predict the online real-time temperature, has the advantages of strong real-time performance and high detection efficiency, and can greatly improve the longitudinal crack recognition efficiency and accuracy of field operators.
Drawings
FIG. 1 is a schematic diagram of the distribution of four crystallizer copper plates and thermocouples;
FIG. 2 is a schematic representation of the longitudinal crack temperature and its pretreatment results; FIG. 2(a) is a graph showing the temperatures of the first and second thermocouples at the time of occurrence of a longitudinal crack, and FIG. 2(b) is a graph showing the results of pretreatment at the temperatures shown in FIG. 2 (a);
FIG. 3 is a schematic diagram of normal operating conditions and pretreatment results thereof; FIG. 3(a) is a graph of the temperatures of the first and second thermocouples under normal operating conditions, and FIG. 3(b) is a graph of the results of the pre-treatment of the temperatures shown in FIG. 3 (a);
FIG. 4 is a flow chart of on-line detection of longitudinal cracks;
FIG. 5 is a graph of longitudinal crack temperature as measured on line;
FIG. 6 is a normal operating temperature plot at online real time.
Detailed Description
The invention is further illustrated by the following specific examples in conjunction with the accompanying drawings.
Fig. 1 is a schematic diagram showing the distribution of the on-line crystallizer and its thermocouples in service in a certain steel mill. The crystallizer copper plate is 900mm in height, 800mm in effective height and formed by combining four copper plates. 3 rows of 19 rows of thermocouples are respectively arranged on the inner arc wide-surface copper plate and the outer arc wide-surface copper plate, 3 rows of 1 row of thermocouples are respectively arranged on the left side narrow-surface copper plate and the right side narrow-surface copper plate, and 120 thermocouples are arranged on the four copper plates in total. The distance between the first row of thermocouples and the upper opening of the crystallizer is 210mm, the distance between the first row of thermocouples and the second row of thermocouples is 115mm, the distance between the second row of thermocouples and the third row of thermocouples is 120mm, and the distance between two adjacent rows of thermocouples is 150 mm.
First step, temperature data acquisition
(1) For longitudinal crack temperature data, intercepting temperature data of thermocouples in the same row for 110 seconds, wherein the temperature data comprises a temperature descending-stabilizing-ascending fluctuation process;
(2) and intercepting the temperature data of the thermocouples in the same column for 110 seconds continuously according to the temperature data under the normal working condition.
Second, temperature pretreatment
(1) And (3) performing difference processing on the thermocouple temperatures of the 1 st row and the 2 nd row of thermocouples in the same row:
in the formula, T(1)i、T(2)iThe values of the thermocouple temperature at the ith moment in the rows 1 and 2 are respectively.
Is the difference of the values of the thermocouple temperature at the ith moment in rows 1 and 2
(2) Calculating the temperature difference T of the thermocouples in the 1 st and 2 nd rowsminusRate of change at intervals of 10 seconds:
in the formula (I), the compound is shown in the specification,
is the difference between the values of the thermocouple temperatures at the i +10 th time points in the 1 st and 2 nd rows.
Fig. 2(a) shows the temperature of the first and second rows of the galvanic couple array in which the longitudinal crack is located, and fig. 2(b) shows the rate of change in the temperature difference. Fig. 3(a) shows the temperatures of the first and second heat release couples in the normal operating condition, and fig. 3(b) shows the rate of change in the temperature difference.
Taking T _ M _ V obtained from thermocouple temperatures in the same column as a sample, and forming a temperature time sequence sample library in such a way that:
ST_M_V＝{(T_M_VC1,1),(T_M_VC2,1),…,(T_M_VC40,1),(T_M_VN1,0),(T_M_VN2,0),…,(T_M_VN400), where T _ M _ VCThe class label of (1) represents a longitudinal crack sample; t _ M _ VNThe class label of (1) is 0, indicating a normal operating condition sample. A total of 40 longitudinal crack samples and 40 normal working condition samples are selected, and are respectively shown in FIG. 2(b) and FIG. 3(b) As shown.
Thirdly, obtaining the classification decision function of the optimal logistic regression
FIG. 4 is a flow chart of the method for predicting longitudinal cracks of casting blanks by using a logistic regression classification method. Library S of temperature time series samplesT_M_VAnd randomly dividing the sample into 4 groups of sub-sample libraries, wherein each group of sub-sample libraries comprises a training sample set and a test sample set, training the training sample set in each group of sub-sample libraries by using a logistic regression classification method, and then testing the test sample set in the sub-sample libraries by using the obtained model. The 4 groups of sub-sample libraries are respectively trained and tested to obtain 4 testing accuracy rates, and the classification prediction model with the optimal testing accuracy rate and the corresponding classification decision function are selected, so that the method comprises the following steps:
(1) model training: at ST_M_VRandomly extracting more than 30 longitudinal crack samples T _ M _ VCAnd more than 30 normal working condition samples T _ M _ VNForming a training set, training the training set by using a logistic regression classification method to obtain a classification decision function corresponding to the logistic regression classification method, namely:
wherein the content of the first and second substances,
representing a vector of temperature time-series samples, ω being a weight vector corresponding to a time-series sample, ωTIs the transpose of ω, and b is the displacement term.
(2) Using the classification decision function pair S obtained in (1)T_M_VAnd (5) testing a test set consisting of the residual samples, and recording the prediction accuracy.
(3) And (3) repeating the steps (1) to (2) for 4 times, randomly extracting the longitudinal crack samples and the normal working condition samples again each time to form a new training sample set, and taking the residual samples as the test sample set to finally obtain 4 classification decision functions and the corresponding prediction accuracy rates thereof.
(4) Selecting the classification decision function with the highest prediction accuracy to obtain the best logicEdit regression classification decision function
Wherein:
ω＝[0.139,0.082,0.118,…,-0.431,-0.165,-0.214]|,|ω|＝100
b＝0.932
in the formula, | ω | represents the dimension of the weight vector ω.
Fourth, longitudinal crack real-time temperature on-line prediction
(1) In the process of on-line detection, the real-time temperature of the thermocouples in the same row at the current moment and for 109 seconds before the current moment is intercepted, and the real-time temperature is subjected to time sequence preprocessing in the same way as the second step to obtain a real-time temperature time sequence sample T _ M _ Vnew；
(2) Using the optimal logistic regression classification decision function obtained in the third step to perform real-time temperature time sequence sample T _ M _ VnewAnd (3) predicting a category label:
(3) setting the decision threshold lambda to 0.8, and if the value of the decision function is greater than or equal to 0.8, considering T _ M _ VnewAnd (3) sending a longitudinal crack alarm for longitudinal cracks, otherwise, updating the time, and executing a fourth step (1), a fourth step (2) and a fourth step (3) on the real-time temperature of the same row of thermocouples which are continuously 109 seconds before and at the next moment, namely the real-time temperature of 110 seconds corresponding to the next moment, wherein the threshold lambda meets the condition lambda ∈ [0.5, 1).
FIG. 5 is a graph of longitudinal crack temperature as measured on line; the time sequence preprocessing is performed on the 110-second temperature corresponding to the current time in fig. 5, so that: t _ M _ Vnew1Will T _ M _ Vnew1Substituting the logistic regression classification decision function to obtain:
and (4) sending a longitudinal crack alarm when the value of the decision function is greater than the decision threshold value of 0.8, namely the longitudinal crack sample is obtained.
FIG. 6 is a normal operating temperature plot at online real time. The time sequence preprocessing is performed on the 110-second temperature corresponding to the current time in fig. 6, so that: t _ M _ Vnew2Will T _ M _ Vnew2Substituting the logistic regression classification decision function to obtain:
and (3) updating the time because the value of the decision function is far smaller than the decision threshold value of 0.8, and executing the fourth step (1), the fourth step (2) and the fourth step (3) on the thermocouple temperature data of 110 seconds corresponding to the next moment.
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 (5)
1. A method for predicting continuous casting billet longitudinal cracks by logistic regression classification is characterized by comprising the following steps:
first, acquiring temperature data
(1) For a longitudinal crack historical temperature sample, intercepting temperature data of thermocouples in the same row for L seconds, wherein the temperature data comprises a temperature descending-stabilizing-ascending fluctuation process;
(2) intercepting temperature data of thermocouples in the same row for continuous L seconds for the temperature data under the normal working condition;
second, temperature pretreatment
(1) And (3) performing difference processing on the temperatures of the thermocouples in the r th row and the r +1 th row of the thermocouples in the same row:
in the formula, T(r)i、T(r+1)iThe values of the thermocouple temperature at the ith moment in the r th row and the r +1 th row respectively;
the difference value of the values of the thermocouple temperature at the ith moment in the r and r +1 rows;
(2) solving the r and r +1 rows of thermocouple temperature difference TminusRate of change at intervals of k seconds:
in the formula (I), the compound is shown in the specification,
the difference value of the values of the thermocouple temperature at the (i + k) th time in the (r) th and (r + 1) th rows;
taking T _ M _ V obtained by the temperature of the thermocouples in the same column as a sample to form a temperature time sequence sample library:
ST_M_V＝{(T_M_VC1,1),(T_M_VC2,1),…,(T_M_VCm,1),(T_M_VN1,0),(T_M_VN2,0),…,(T_M_VNn0), where T _ M _ VCThe class label of (1) represents a longitudinal crack sample; t _ M _ VNThe class label of (1) is 0, which represents a normal working condition sample; m and n are the number of longitudinal cracks and normal working condition samples in the sample library respectively;
thirdly, obtaining an optimal classification decision function
Library S of temperature time series samplesT_M_VRandomly dividing the test sample into H groups of sub-sample libraries, wherein each group of sub-sample libraries comprises a training sample set and a test sample set, training the training sample set in each group of sub-sample libraries by using a logistic regression classification method, and then testing the test sample set in the group of sub-sample libraries by using the obtained model; the H groups of sub-sample libraries are respectively trained and tested to obtain H test accuracy rates, and a classification prediction model with the optimal test accuracy rate and a corresponding classification decision function are selected, wherein the method comprises the following steps:
(1) model training: at ST_M_VRandomly extracting more than 30 longitudinal crack samples T _ M _ VCAnd more than 30 normal working condition samples T _ M _ VNThe training set is formed by the training data,training the classification function by using a logistic regression classification method to obtain a classification decision function corresponding to the logistic regression classification method, namely:
wherein the content of the first and second substances,
representing a temperature time sequence sample vector, wherein omega is a weight vector corresponding to the time sequence sample, and b is a displacement term;
(2) and (3) testing a model: using the classification decision function pair S obtained in (1)T_M_VTesting a test set consisting of the middle and residual samples, and recording the prediction accuracy;
(3) repeating the steps (1) to (2) for H times, randomly extracting longitudinal crack samples and normal working condition samples again each time to form a new training sample set, and taking the residual samples as a test sample set to finally obtain H classification decision functions and corresponding prediction accuracy rates thereof;
(4) selecting a classification decision function with the highest prediction accuracy as an optimal logistic regression classification decision function;
fourth step, longitudinal crack on-line prediction
(1) In the on-line detection process, the real-time temperature of the thermocouples in the same row at the current moment and in the previous continuous L-1 seconds is intercepted, and the real-time temperature is preprocessed in the same way as the second step to obtain a real-time temperature time sequence sample T _ M _ Vnew；
(2) Using the optimal logistic regression classification decision function obtained in the third step to perform real-time temperature time sequence sample T _ M _ VnewPredict the class label of (1):
(3) if the value of the decision function is greater than or equal to the threshold lambda, T _ M _ V is considerednewSending a longitudinal crack alarm for the longitudinal crack; otherwise, the time is updated for the next time and the last continuous L-1 secondAnd (3) executing the fourth step (1), the fourth step (2) and the fourth step (3) according to the real-time temperature of the thermocouples in the same column, namely the real-time temperature of L seconds corresponding to the next moment.
2. The method for predicting the longitudinal cracks of the continuous casting billet by using the logistic regression classification as claimed in claim 1, wherein the number of the longitudinal crack samples and the number of the normal working condition samples in the temperature time sequence sample library in the second step (2) are not less than 40.
3. The method for predicting the longitudinal crack of the continuous casting billet by using the logistic regression classification as claimed in claim 1 or 2, wherein the threshold value λ in the fourth step (3) is satisfied with a condition λ ∈ [0.5,1 ].
4. The method for predicting the longitudinal cracks of the continuous casting billet by using the logistic regression classification is characterized by being suitable for predicting the longitudinal cracks of the continuous casting billet of a slab billet, a square billet, a round billet and a beam blank.
5. The method for predicting the longitudinal cracks of the continuous casting billet by using the logistic regression classification is characterized by being suitable for predicting the longitudinal cracks of the continuous casting billet of a slab billet, a square billet, a round billet and a beam blank.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010385010.7A CN111666710B (en) | 2020-05-06 | 2020-05-06 | Method for predicting continuous casting billet longitudinal cracks by logistic regression classification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010385010.7A CN111666710B (en) | 2020-05-06 | 2020-05-06 | Method for predicting continuous casting billet longitudinal cracks by logistic regression classification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111666710A true CN111666710A (en) | 2020-09-15 |
| CN111666710B CN111666710B (en) | 2022-10-21 |
Family
ID=72383187
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010385010.7A Active CN111666710B (en) | 2020-05-06 | 2020-05-06 | Method for predicting continuous casting billet longitudinal cracks by logistic regression classification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111666710B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115049260A (en) * | 2022-06-15 | 2022-09-13 | 华院计算技术(上海)股份有限公司 | Application method and system of cognitive intelligent continuous casting ladle quality pre-judgment model |
| CN115446276A (en) * | 2022-10-05 | 2022-12-09 | 大连理工大学 | Continuous casting breakout early warning method for recognizing V-shaped bonding characteristics of crystallizer copper plate based on convolutional neural network |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108436050A (en) * | 2018-04-16 | 2018-08-24 | 大连理工大学 | A method of continuous cast mold bleed-out is forecast using space density clustering DBSCAN |
| CN108580827A (en) * | 2018-05-22 | 2018-09-28 | 大连理工大学 | A method of Crystallizer bleed-out is forecast based on Agglomerative Hierarchical Clustering |
-
2020
- 2020-05-06 CN CN202010385010.7A patent/CN111666710B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108436050A (en) * | 2018-04-16 | 2018-08-24 | 大连理工大学 | A method of continuous cast mold bleed-out is forecast using space density clustering DBSCAN |
| CN108580827A (en) * | 2018-05-22 | 2018-09-28 | 大连理工大学 | A method of Crystallizer bleed-out is forecast based on Agglomerative Hierarchical Clustering |
Non-Patent Citations (1)
| Title |
|---|
| 朱劲夫等: "正则化在逻辑回归与神经网络中的应用研究", 《信息技术》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115049260A (en) * | 2022-06-15 | 2022-09-13 | 华院计算技术(上海)股份有限公司 | Application method and system of cognitive intelligent continuous casting ladle quality pre-judgment model |
| CN115446276A (en) * | 2022-10-05 | 2022-12-09 | 大连理工大学 | Continuous casting breakout early warning method for recognizing V-shaped bonding characteristics of crystallizer copper plate based on convolutional neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111666710B (en) | 2022-10-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5505818B2 (en) | Defect determination device | |
| CN109396375A (en) | A kind of crystallizer bleedout prediction electric thermo method based on feature vector and hierarchical clustering | |
| CN111666710B (en) | Method for predicting continuous casting billet longitudinal cracks by logistic regression classification | |
| Lattanzi et al. | A fast methodology for the accurate characterization and simulation of laser heat treated blanks | |
| Xiao et al. | Modeling constitutive relationship of 6013 aluminum alloy during hot plane strain compression based on Kriging method | |
| Varfolomeev et al. | Statistical control of defects in a continuously cast billet based on machine learning and data analysis methods | |
| Duan et al. | Longitudinal crack detection approach based on principal component analysis and support vector machine for slab continuous casting | |
| CN117494531B (en) | Medium carbon steel decarburization depth prediction method based on finite element and XGBoost algorithm | |
| Kumar et al. | Development of intelligent mould for online detection of defects in steel billets | |
| CN111680448B (en) | Continuous casting billet longitudinal crack prediction method based on SVM classification | |
| CN117036797A (en) | Continuous casting billet longitudinal crack prediction method based on feature extraction and random forest classification | |
| Duan et al. | Modeling of breakout prediction approach integrating feature dimension reduction with K-means clustering for slab continuous casting | |
| Cheng et al. | Molten steel breakout prediction based on genetic algorithm and BP neural network in continuous casting process | |
| JP7287416B2 (en) | Thick steel plate manufacturing specification determination support device, manufacturing specification search method, computer program, computer-readable recording medium, and thick steel plate manufacturing method | |
| Gang et al. | Modeling constitutive relationship of 6013 aluminum alloy during hot plane strain compression based on Kriging method | |
| CN111618265B (en) | Continuous casting billet longitudinal crack online detection method based on K nearest neighbor classification | |
| CN115985424A (en) | BP neural network-based part heat treatment performance prediction method and system | |
| CN117171936A (en) | Slab quality prediction method for extracting real-time characteristic value of crystallizer based on defect mechanism | |
| CN115203903A (en) | Temperature state evaluation method based on billet target temperature rise process | |
| JP2002236119A (en) | Material estimating device for steel product | |
| Normanton et al. | Developments in online surface and internal quality forecasting of continuously cast semis | |
| Kwon et al. | Experimental and numerical prediction of austenite grain size distribution in round-oval shape rolling | |
| Kawalla et al. | An empirical examination of the thickness profile formation of twin-roll-cast magnesium strips | |
| CN115446276B (en) | Continuous casting steel leakage early warning method based on convolutional neural network recognition crystallizer copper plate V-shaped bonding characteristics | |
| Vlado et al. | The production of cracks evolution in continuously cast steel slab |
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 |