CN117744503A - Prediction method for tapping temperature of converter - Google Patents
Prediction method for tapping temperature of converter Download PDFInfo
- Publication number
- CN117744503A CN117744503A CN202410186311.5A CN202410186311A CN117744503A CN 117744503 A CN117744503 A CN 117744503A CN 202410186311 A CN202410186311 A CN 202410186311A CN 117744503 A CN117744503 A CN 117744503A
- Authority
- CN
- China
- Prior art keywords
- data
- converter
- tapping temperature
- temperature prediction
- prediction method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000010079 rubber tapping Methods 0.000 title claims abstract description 68
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000004519 manufacturing process Methods 0.000 claims abstract description 31
- 238000000513 principal component analysis Methods 0.000 claims abstract description 12
- 238000003723 Smelting Methods 0.000 claims abstract description 8
- 238000005457 optimization Methods 0.000 claims abstract description 6
- 238000013527 convolutional neural network Methods 0.000 claims description 20
- 238000013528 artificial neural network Methods 0.000 claims description 19
- 241000282421 Canidae Species 0.000 claims description 16
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 claims description 14
- 239000011159 matrix material Substances 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 10
- 230000002159 abnormal effect Effects 0.000 claims description 9
- 238000012847 principal component analysis method Methods 0.000 claims description 8
- 239000002893 slag Substances 0.000 claims description 8
- 229910052742 iron Inorganic materials 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 7
- 241000282461 Canis lupus Species 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 5
- 229910052799 carbon Inorganic materials 0.000 claims description 4
- 238000004091 panning Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical group [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 claims description 3
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical group [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 3
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 3
- 238000009776 industrial production Methods 0.000 claims description 3
- 239000000463 material Substances 0.000 claims description 3
- CSJDCSCTVDEHRN-UHFFFAOYSA-N methane;molecular oxygen Chemical compound C.O=O CSJDCSCTVDEHRN-UHFFFAOYSA-N 0.000 claims description 3
- 229910052760 oxygen Inorganic materials 0.000 claims description 3
- 239000001301 oxygen Substances 0.000 claims description 3
- 229910052698 phosphorus Inorganic materials 0.000 claims description 3
- 239000011574 phosphorus Substances 0.000 claims description 3
- 238000003556 assay Methods 0.000 claims description 2
- 230000007423 decrease Effects 0.000 claims description 2
- 238000012804 iterative process Methods 0.000 claims description 2
- 238000013178 mathematical model Methods 0.000 claims description 2
- 238000010606 normalization Methods 0.000 claims description 2
- 238000011176 pooling Methods 0.000 claims description 2
- 238000010200 validation analysis Methods 0.000 claims description 2
- 238000009628 steelmaking Methods 0.000 abstract description 12
- 238000009851 ferrous metallurgy Methods 0.000 abstract description 6
- 229910000831 Steel Inorganic materials 0.000 description 20
- 239000010959 steel Substances 0.000 description 20
- 238000001514 detection method Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000009825 accumulation Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000007664 blowing Methods 0.000 description 1
- 238000002485 combustion reaction Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 239000011819 refractory material Substances 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Landscapes
- Carbon Steel Or Casting Steel Manufacturing (AREA)
Abstract
The invention belongs to the technical field of ferrous metallurgy, in particular to a prediction method of tapping temperature of a converter, and provides a data driving prediction model based on converter production data, wherein main characteristic components of initial data are extracted by using a Principal Component Analysis (PCA) method, mathematical dimension reduction is performed, accurate optimization of different smelting data in a complex production environment is realized by combining GWO-CNN, rapid and accurate prediction of tapping temperature of the converter is realized, component hit rate and product stability in the tapping process of the converter are improved, steelmaking cost is saved, and the method has good application prospect in the ferrous metallurgy field.
Description
Technical Field
The invention belongs to the technical field of ferrous metallurgy, and particularly relates to a method for predicting tapping temperature of a converter.
Background
The converter is an important step in the steel industry, the tapping temperature is an important control parameter of the steelmaking process, and accurate molten steel temperature prediction has important guiding significance for converter end point control. Converter steelmaking is a complex physicochemical reaction process at temperatures up to 1550-1650 ℃. The steel tapping temperature is too high, so that the energy consumption and the cost in the steel making process are increased, molten steel is easy to oxidize, steel ladle refractory materials are corroded, foreign impurities are generated, and the quality of the molten steel is affected. The continuous and accurate measurement of the temperature of molten steel in a hearth is difficult to realize by the existing means, when most of iron and steel enterprises approach to a converting end point, the tapping time is judged by means of manual experience according to a sublance sampling result, the control precision is low and unstable, and therefore, the establishment of an accurate converter end point temperature prediction model has great significance for converter end point control.
The detection of the converter end temperature is divided into a sensor detection model, a flame image processing model and a data driving prediction model. The sensor detection is mainly that a sensor is arranged in a converter to directly measure the temperature of molten steel, and the method is matched with manual experience prediction to improve the hit rate of a terminal point, but only aims at temperature detection at a single moment, and has higher use and maintenance cost. The flame image processing can accurately measure the temperature in real time, replaces the traditional method relying on the experience of workers, effectively reduces subjective influence and labor cost, but has instability in the flame combustion process, and has certain characteristics of non-rigidity, multi-scale and randomness, and the flame image has certain difficulty in reflecting the blowing state only by means of a single image.
Disclosure of Invention
The invention aims to solve the problems in the prior art, and mainly aims to provide a method for predicting the tapping temperature of a converter end point, which aims to solve the problem that the tapping temperature is difficult to predict in the current converter smelting tapping process.
In order to solve the technical problems, according to one aspect of the present invention, the following technical solutions are provided:
the converter tapping temperature prediction method comprises the following steps:
s1, collecting an industrial production converter production data set, and establishing a prediction model database;
s2, carrying out data screening and panning on the collected production data set, and preprocessing the screened and panned data;
s3, extracting main characteristic components from the converter production process data by using a Principal Component Analysis (PCA) method, and then mathematically reducing the dimension;
s4, establishing a Convolutional Neural Network (CNN) neural network converter tapping temperature prediction model optimized by a wolf optimization algorithm (GWO);
s5, training and verifying a prediction model by adopting a preprocessed historical data set;
s6, collecting real-time data of a field smelting process;
s7, performing main component analysis method dimension reduction treatment on the acquired real-time data;
s8, substituting a GWO-CNN neural network converter tapping temperature prediction model to predict the converter tapping temperature.
The preferable scheme of the converter tapping temperature prediction method is as follows: s9, judging tapping time according to the real-time tapping temperature prediction result of the converter, storing data after tapping into a prediction model database, and updating the prediction model periodically.
The preferable scheme of the converter tapping temperature prediction method is as follows: in the step S1, the converter production dataset includes: furnace number, furnace age, operating shift, operating hands, tapping amount, molten iron weight, scrap type (heavy scrap, medium scrap, light scrap, self-produced slag steel), converting period, total oxygen amount, slag alkalinity, molten iron temperature, slag forming material addition amount, number of times of pouring, terminal temperature, terminal carbon content, terminal phosphorus content, carbon oxygen accumulation.
The preferable scheme of the converter tapping temperature prediction method is as follows: in the step S2, data screening and panning are performed on the collected converter production dataset, which specifically includes:
deleting the repeated data, deleting the abnormal point and the extreme abnormal point, and judging the abnormal data as shown in the formula (1):
（1）
wherein,indicating the first +.>Data of->Is the upper quartile, & lt & gt>For the lower quartile,/->Representing a quarter bit distance.
The preferable scheme of the converter tapping temperature prediction method is as follows: in the step S2, preprocessing the screened and panned data specifically includes:
normalizing the data set to [ -1,1] in the following manner:
（2）
wherein,is a characteristic variable of the input, wherein->And->Maximum and minimum values for each individual sample data.
The preferable scheme of the converter tapping temperature prediction method is as follows: in the step S2, the converter production dataset is represented by 8: the scale of 2 distinguishes between training and validation sets.
The preferable scheme of the converter tapping temperature prediction method is as follows: in the step S3, the principal component analysis method is very suitable for a high-dimensional dimension-reduction algorithm, and the processed data are converted into new data main characteristic components, and the variables are not related and ordered. The principal component analysis method comprises the following processing steps:
1) Forming an n-row m-column matrix X from the original data according to columns;
2) Zero-equalizing each row of X (representing an attribute field), i.e., subtracting the average of the row;
3) Obtaining a covariance matrix;
4) Obtaining eigenvalues and corresponding eigenvectors r of the covariance matrix;
5) Arranging the eigenvectors into a matrix according to the corresponding eigenvalue from top to bottom, and taking the first k rows to form a matrix P;
6) The matrix P is the data after the dimension is reduced to k dimension.
The preferable scheme of the converter tapping temperature prediction method is as follows: in the step S4, the GWO iterative algorithm is represented by formulas (3) - (13):
1) A hunting object:
（3）
（4）
（5）
（6）
t represents the current iteration number, a and C are coefficients,for the position of prey, < > for>Is of the wolf shapePosition of the t th generation of the body. r is (r) 1 、r 2 Is [0, 1]]Random values in (a) are provided. To simulate an approaching prey, A is the interval [ -a, a]Wherein a decreases from 2 to 0 in an iterative process.
2) Attack prey:
alpha, beta and delta are all individual wolves, and mathematical models of the individual wolves tracking the position of the hunting object are described as follows:
（7）
（8）
（9）
（10）
（11）
（12）
（13）
respectively representing the distances among alpha wolves, beta wolves and delta wolves and other individuals; />Respectively representing the current positions of alpha wolves, beta wolves and delta wolves; />Is a random number, and X is the current position of the individual gray wolf.
The preferable scheme of the converter tapping temperature prediction method is as follows: in the step S4, the CNN neural network module includes: six convolution layers, a pooling layer and a batch normalization layer, wherein each three convolution layers are grouped together, the first group consisting of 32 3 x 3 convolution kernels and the second group consisting of 64 3 x 3 convolution kernels.
The preferable scheme of the converter tapping temperature prediction method is as follows: in the step S6, the method for collecting real-time data in the on-site smelting process includes: obtained from the Oracle database of the assay system and the secondary system by establishing an ODBC connection.
The preferable scheme of the converter tapping temperature prediction method is as follows: the prediction method further comprises the steps of continuously optimizing and fine-tuning the temperature prediction model according to user feedback and model performance:
if the model performs poorly in some situations, the temperature prediction model is continually optimized by adding various data types in the user-specific needs database to the accurate temperature data to generate more training samples.
The beneficial effects of the invention are as follows:
the invention provides a converter tapping temperature prediction method, and provides a data-driven prediction model based on converter production data, wherein main characteristic components of initial data are extracted by using a Principal Component Analysis (PCA) method, then mathematical dimension reduction is performed, accurate optimization of different smelting data under a complex production environment is realized by combining GWO-CNN, the rapid and accurate prediction of the converter tapping temperature is realized, the component hit rate and the product stability in the converter tapping process are improved, the steelmaking cost is saved, and the method has a good application prospect in the field of ferrous metallurgy.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to the structures shown in these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a converter tapping temperature prediction method.
Fig. 2 is a graph showing predicted and actual values of the tapping temperature of the converter according to example 1 of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
The following description will be made clearly and fully with reference to the technical solutions in the embodiments, and it is apparent that the described embodiments are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The method for predicting the tapping temperature of the converter provided by the invention can rapidly and accurately predict the tapping temperature of the converter, improves the temperature hit rate in the tapping process of the converter, is beneficial to improving the quality of steel products, saves the steelmaking cost and has good application prospects in the field of ferrous metallurgy. In the converter steelmaking process, the reaction environment in the converter is complex, the measurement is difficult, a principal component analysis method is combined with a GWO-CNN neural network, a data driving prediction model based on converter production data is provided, principal characteristic components of initial data are extracted after mathematical analysis of the initial data by adopting the principal component analysis method, and the dimension is reduced after the principal and secondary relations among variables are found. GWO-CNN neural network overcomes the inherent defects that the traditional neural network is difficult to process high-dimensional data, has low training speed and is easy to fall into a local optimal solution, and GWO-CNN neural network has the advantages of high learning speed, automatic feature extraction, no pressure in processing the high-dimensional data, high precision and the like.
BP neural networks are generally determined by repeated forward and backward propagation modes, and changing the number of hidden nodes changes the corresponding initial weights and thresholds, affecting the convergence and learning efficiency of the network. The CNN neural network reduces the number of weights, so that the network is easy to optimize, meanwhile, the complexity of a model is reduced, and the risk of overfitting is reduced. Meanwhile, in order to reduce the influence, GWO provides a global search in a decision space from the initial iteration to the final iteration, when the algorithm falls into a local optimum and is not easy to jump out, the randomness of the component plays a very important role in avoiding the local optimum, and a global optimum solution is obtained in the final iteration. And GWO-CNN neural network obtains optimal solution every iteration, and improves the learning efficiency and model accuracy of the network. The learning speed is high in the industrial production process, the large-scale data fusion can be performed, the data can be processed in parallel at high speed, and the method is suitable for solving the problems of large data volume and large data variable in converter steelmaking production.
The technical scheme of the invention is further described below by combining specific embodiments.
Example 1
Taking Q345B steel as an example, the average end temperature of the top-bottom combined blown converter of an 80t steel mill is 1615.2 ℃, the end temperatures are distributed within the range of 1606.9-1649.3 ℃, and the end temperature fluctuation is large, so that the tapping time after the converter is inaccurate. Meanwhile, molten steel is easy to oxidize and bring oxide inclusions at a higher tapping temperature, and the quality of the molten steel is affected. The most steel mills in China mainly utilize manual experience to manually control the converter steelmaking end point according to field operators, and the manual operation experience is obtained only through summarizing and analyzing field production and production control data, so that the stable and accurate effect is difficult to achieve. Therefore, the accurate prediction of the tapping temperature of the converter is significant for the smooth production of the converter, the production cost is saved and the stability of the tapping temperature is improved.
The production data of the 6207 group of the 80t converter of the factory is collected, repeated data are deleted, abnormal points and extreme abnormal points are deleted, and the evaluation formula of the abnormal data is as follows:
（1）
wherein,indicating the first +.>Data of->Is the upper quartile, & lt & gt>For the lower quartile,/->Representing a quarter bit distance.
4978 groups of effective data are obtained after screening. To make different variables have the same metric scale, the data are normalized and all mapped into the range of [ -1,1 ];
（2）
wherein,is a characteristic variable of the input, wherein->And->Maximum and minimum values for each individual sample data.
And performing principal component analysis dimension reduction processing on all the acquired data variables. PCA principal component analysis was performed on the normalized data for 21 parameters: furnace number, furnace age, operating shift, operating hands, tapping amount, molten iron weight, scrap type (heavy scrap, medium scrap, light scrap, self-produced slag steel), converting period, total oxygen amount, slag alkalinity, molten iron temperature, slag forming material addition amount, number of times of pouring, terminal temperature, terminal carbon content, terminal phosphorus content, carbon oxygen accumulation.
11 main characteristic parameters in all independent parameters are extracted, the extracted main characteristic parameters are used as input variables of a GWO-CNN neural network, the structure of the network and the values of various super parameters of the network are determined through an orthogonal experiment method, 4978 groups of screened standardized independent components are input into a model, wherein the front 3982 groups of data are selected as a training set of the model, the later 996 groups of data are selected as a test set, and the prediction capability of the model is checked.
The four performance indicators are used for evaluating the performance of the tapping temperature prediction model, including Mean Absolute Error (MAE), root Mean Square Error (RMSE), mean Absolute Percentage Error (MAPE) and decision coefficient (R 2 ) The detailed results are shown in the following table.
From the evaluation result and the fitting performance of the model on the training data set, the PCA-GWO-CNN neural network can well fit the original data and has good prediction performance on the test set.
In order to verify the application effect of the prediction model in the actual production process, the PCA-GWO-CNN converter tapping temperature prediction model established by the invention is applied to actual tracking production calculation, and the converter tapping temperature model prediction error rate is researched. 25 groups of production data are randomly selected from 4978 groups of production data, and the qualification rate of the tapping temperature of the converter reaches 100%. After the model predicts the tapping temperature of the converter and then the converter goes out, the temperature of the finished product is stably controlled between 1610.7 ℃ and 1619.3 ℃, and the error between the predicted value and the actual value is not more than +/-2 ℃, as shown in figure 2. The practical application effect shows that the temperature control of the finished product after the tapping temperature of the converter is predicted by the model and then the tapping temperature of the converter is more stable than that of the traditional mode, the control range is smaller, and the accurate control of the molten steel temperature in the tapping process of the converter is realized.
According to the invention, a principal component analysis method is combined with a GWO-CNN neural network, the complex reaction environment in a converter in the steelmaking process of the converter is considered, the measurement is difficult, the principal component analysis method is combined with the GWO-CNN neural network, a data driving prediction model based on converter production data is provided, the principal characteristic components of initial data are extracted by using a principal component analysis method (PCA), then mathematical dimension reduction is performed, and the accurate optimization of different smelting data in the complex production environment is realized by combining GWO-CNN, so that the rapid and accurate prediction of converter tapping temperature is realized. PCA in the data preprocessing method is widely applied to a data prediction model, so that a data set is easier to use, the calculation cost of an algorithm is reduced, and the result is easy to understand and has no parameter limitation at all. Meanwhile, the GWO algorithm overcomes the inherent defects that the traditional neural network is difficult to process high-dimensional data, has low training speed and is easy to fall into a local optimal solution, and the GWO-CNN neural network has the advantages of high learning speed, automatic feature extraction, no pressure in processing the high-dimensional data, high precision and the like. The method can rapidly and accurately realize the prediction of the tapping temperature of the converter, improves the hit rate of the tapping temperature of the converter and the stability of steel products, saves the steelmaking cost, improves the quality of the steel products, and has good application prospect in the field of ferrous metallurgy. The method is verified by the actual production data on site, and the result shows that the method has better accuracy and applicability, and can provide beneficial guidance for the production process of the converter steelmaking site.
The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the invention, and all equivalent structural changes made by the content of the present invention or direct/indirect application in other related technical fields are included in the scope of the present invention.
Claims (10)
1. The converter tapping temperature prediction method is characterized by comprising the following steps of:
s1, collecting an industrial production converter production data set, and establishing a prediction model database;
s2, carrying out data screening and panning on the collected production data set, and preprocessing the screened and panned data;
s3, extracting main characteristic components from the converter production process data by using a principal component analysis method, and then mathematically reducing the dimension;
s4, establishing a CNN neural network converter tapping temperature prediction model optimized by a GWO optimization algorithm;
s5, training and verifying a prediction model by adopting a preprocessed historical data set;
s6, collecting real-time data of a field smelting process;
s7, performing main component analysis method dimension reduction treatment on the acquired real-time data;
s8, substituting a GWO-CNN neural network converter tapping temperature prediction model to predict the converter tapping temperature.
2. The converter tapping temperature prediction method according to claim 1, wherein in the step S1, the converter production data set includes: furnace number, furnace age, operating shift, operating hands, tapping amount, molten iron weight, scrap type, converting period, total oxygen amount, slag basicity, molten iron temperature, slag forming material addition amount, furnace pouring times, terminal temperature, terminal carbon content, terminal phosphorus content and carbon oxygen product.
3. The converter tapping temperature prediction method according to claim 1, wherein in the step S2, the collected converter production dataset is subjected to data screening and panning, and specifically comprises:
deleting the repeated data, deleting the abnormal point and the extreme abnormal point, and judging the abnormal data as shown in the formula (1):
（1）
wherein,indicating the first +.>Data of->To get up toQuartile,/->For the lower quartile,/->Representing a quarter bit distance.
4. The converter tapping temperature prediction method according to claim 1, wherein in the step S2, the screened and panned data is preprocessed, specifically comprising:
normalizing the data set to [ -1,1] in the following manner:
（2）
wherein,is a characteristic variable of the input, wherein->And->Maximum and minimum values for each individual sample data.
5. The converter tapping temperature prediction method according to claim 1, wherein in the step S2, the converter production dataset is represented by 8: the scale of 2 distinguishes between training and validation sets.
6. The converter tapping temperature prediction method according to claim 1, wherein in the step S3, the principal component analysis processing step is as follows:
1) Forming an n-row m-column matrix X from the original data according to columns;
2) Zero-equalizing each row of X, namely subtracting the average value of the row;
3) Obtaining a covariance matrix;
4) Obtaining eigenvalues and corresponding eigenvectors r of the covariance matrix;
5) Arranging the eigenvectors into a matrix according to the corresponding eigenvalue from top to bottom, and taking the first k rows to form a matrix P;
6) The matrix P is the data after the dimension is reduced to k dimension.
7. The converter tapping temperature prediction method according to claim 1, wherein in the step S4, the GWO optimization algorithm is as shown in formulas (3) - (13):
1) A hunting object:
（3）
（4）
（5）
（6）
t represents the current iteration number, a and C are coefficients,for the position of prey, < > for>The position of the t generation of the gray wolf individuals; r is (r) 1 、r 2 Is [0, 1]]Random values of (a); to simulate an approaching prey, A is the interval [ -a, a]Wherein a decreases from 2 to 0 in an iterative process;
2) Attack prey:
alpha, beta and delta are all individual wolves, and mathematical models of the individual wolves tracking the position of the hunting object are described as follows:
（7）
（8）
（9）
（10）
（11）
（12）
（13）
respectively representing the distances among alpha wolves, beta wolves and delta wolves and other individuals; />Respectively representing the current positions of alpha wolves, beta wolves and delta wolves; />Is a random number, and X is the current position of the individual gray wolf.
8. The converter tapping temperature prediction method according to claim 1, wherein in the step S4, the CNN neural network module includes: six convolution layers, a pooling layer and a batch normalization layer, wherein each three convolution layers are grouped together, the first group consisting of 32 3 x 3 convolution kernels and the second group consisting of 64 3 x 3 convolution kernels.
9. The converter tapping temperature prediction method according to claim 1, wherein in the step S6, the method for collecting real-time data of the on-site smelting process comprises the following steps: obtained from the Oracle database of the assay system and the secondary system by establishing an ODBC connection.
10. The converter tapping temperature prediction method according to claim 1, further comprising the step of continuously optimizing and fine-tuning the temperature prediction model according to user feedback and model performance:
if the model performs poorly in some situations, the temperature prediction model is continually optimized by adding various data types in the user-specific needs database to the accurate temperature data to generate more training samples.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410186311.5A CN117744503A (en) | 2024-02-20 | 2024-02-20 | Prediction method for tapping temperature of converter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410186311.5A CN117744503A (en) | 2024-02-20 | 2024-02-20 | Prediction method for tapping temperature of converter |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117744503A true CN117744503A (en) | 2024-03-22 |
Family
ID=90281712
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410186311.5A Pending CN117744503A (en) | 2024-02-20 | 2024-02-20 | Prediction method for tapping temperature of converter |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117744503A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115392286A (en) * | 2022-07-20 | 2022-11-25 | 昆明理工大学 | Bearing fault diagnosis method based on 1dCNN-GWO-SVM model |
| CN115829938A (en) * | 2022-11-07 | 2023-03-21 | 中国钢研科技集团有限公司 | Method for predicting gray wolf and NGboost for surface defects of casting blank genetic hot-rolled strip steel |
| CN117034767A (en) * | 2023-08-11 | 2023-11-10 | 广东工业大学 | Ceramic roller kiln temperature prediction method based on KPCA-GWO-GRU |
| US20230368021A1 (en) * | 2022-05-11 | 2023-11-16 | University Of Science And Technology Beijing | Multi-mode smelting method based on the classification system of molten iron |
-
2024
- 2024-02-20 CN CN202410186311.5A patent/CN117744503A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20230368021A1 (en) * | 2022-05-11 | 2023-11-16 | University Of Science And Technology Beijing | Multi-mode smelting method based on the classification system of molten iron |
| CN115392286A (en) * | 2022-07-20 | 2022-11-25 | 昆明理工大学 | Bearing fault diagnosis method based on 1dCNN-GWO-SVM model |
| CN115829938A (en) * | 2022-11-07 | 2023-03-21 | 中国钢研科技集团有限公司 | Method for predicting gray wolf and NGboost for surface defects of casting blank genetic hot-rolled strip steel |
| CN117034767A (en) * | 2023-08-11 | 2023-11-10 | 广东工业大学 | Ceramic roller kiln temperature prediction method based on KPCA-GWO-GRU |
Non-Patent Citations (1)
| Title |
|---|
| 房荣波, 魏元, 杨海峰: "转炉钢水出钢温度的预测", 鞍钢技术, no. 04, 30 August 2002 (2002-08-30), pages 24 - 29 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111353656B (en) | Steel enterprise oxygen load prediction method based on production plan | |
| CN109447346B (en) | Converter oxygen consumption prediction method based on gray prediction and neural network combined model | |
| CN109583585B (en) | Construction method of power station boiler wall temperature prediction neural network model | |
| CN109935280B (en) | Blast furnace molten iron quality prediction system and method based on ensemble learning | |
| CN104630410A (en) | Real-time dynamic converter steelmaking quality prediction method based on data analysis | |
| CN111444942B (en) | Intelligent forecasting method and system for silicon content of blast furnace molten iron | |
| CN114678080B (en) | Converter end point phosphorus content prediction model, construction method and phosphorus content prediction method | |
| CN109359723A (en) | Based on the converter terminal manganese content prediction technique for improving regularization extreme learning machine | |
| CN111915080B (en) | Raw fuel cost optimal proportioning method based on molten iron quality constraint | |
| CN104899425A (en) | Variable selection and forecast method of silicon content in molten iron of blast furnace | |
| CN111310348A (en) | Material constitutive model prediction method based on PSO-LSSVM | |
| CN110849149A (en) | Energy perception-based sintering batching scheme cascade optimization obtaining method and device | |
| CN115860211A (en) | Casting blank quality prediction method based on local online modeling | |
| CN109934421B (en) | Blast furnace molten iron silicon content prediction and compensation method for fluctuating furnace condition | |
| CN116469481A (en) | LF refined molten steel composition forecasting method based on XGBoost algorithm | |
| CN112329269B (en) | Sintering ignition temperature modeling prediction method based on working condition identification | |
| CN111126550B (en) | Neural network molten steel temperature forecasting method based on Monte Carlo method | |
| US20240002964A1 (en) | Method and system for determining converter tapping quantity | |
| CN102968644A (en) | Method for predicting smelting finishing point of argon-oxygen refined iron alloy | |
| CN117744503A (en) | Prediction method for tapping temperature of converter | |
| Liu et al. | XGBoost-based model for predicting hydrogen content in electroslag remelting | |
| Wang et al. | Application research based on GA-FWA in prediction of sintering burning through point | |
| CN116127345B (en) | Converter steelmaking process mode design method based on deep clustering generation countermeasure network | |
| CN117711516A (en) | Method for predicting molten steel composition at converter end point | |
| Liu et al. | Predicting Alloying Element Yield in Converter Steelmaking Using t-SNE-WOA-LSTM |
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 |