CN115828772B - Rapid calculation method for billet temperature by combining forward mechanism and machine learning - Google Patents
Rapid calculation method for billet temperature by combining forward mechanism and machine learning Download PDFInfo
- Publication number
- CN115828772B CN115828772B CN202310110231.7A CN202310110231A CN115828772B CN 115828772 B CN115828772 B CN 115828772B CN 202310110231 A CN202310110231 A CN 202310110231A CN 115828772 B CN115828772 B CN 115828772B
- Authority
- CN
- China
- Prior art keywords
- billet
- temperature
- dimensional
- time
- heating section
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000010801 machine learning Methods 0.000 title claims abstract description 31
- 238000004364 calculation method Methods 0.000 title claims abstract description 16
- 238000010438 heat treatment Methods 0.000 claims abstract description 115
- 238000000034 method Methods 0.000 claims abstract description 47
- 229910000831 Steel Inorganic materials 0.000 claims description 83
- 239000010959 steel Substances 0.000 claims description 83
- 239000011159 matrix material Substances 0.000 claims description 49
- 238000004519 manufacturing process Methods 0.000 claims description 29
- 239000002436 steel type Substances 0.000 claims description 5
- 238000010521 absorption reaction Methods 0.000 claims description 3
- 238000010079 rubber tapping Methods 0.000 claims description 3
- 238000005098 hot rolling Methods 0.000 abstract description 3
- 238000005096 rolling process Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 239000002737 fuel gas Substances 0.000 description 1
- 238000009628 steelmaking Methods 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Control Of Heat Treatment Processes (AREA)
Abstract
The invention relates to the field of billet hot rolling control, and discloses a rapid calculation method for billet temperature by combining a forward mechanism and machine learning. The method solves the problem that the calculation efficiency and the solving precision can not be achieved when the mechanism model is used for calculating the internal temperature of the billet, and compared with a method for calculating by using the mechanism model alone, the method can give the result in a short time and can ensure the precision of the result; meanwhile, the state of the billet can be corrected according to the historical information of the billet in the furnace, and the corrected result is used for prediction, so that the influence of the temperature fluctuation of the heating furnace on the prediction result is reduced.
Description
Technical Field
The invention relates to the field of billet hot rolling control, in particular to a rapid calculation method for billet temperature by combining a forward mechanism and machine learning.
Background
Hot rolling of billets is an important feature of the modern steelmaking industry in which billets are heated to a suitable temperature in a steel rolling furnace and then fed into a rolling mill. Because a large amount of fuel gas is consumed in the heating process, the automatic control of the billet heating process is important to energy conservation and emission reduction. The temperature of the steel billet is one of key indexes affecting the subsequent process, and the temperature condition of the steel billet in a steel rolling heating furnace is one of variables which need to be concerned by automatic control. The temperature inside the steel billet is not uniform due to the influence of the environment of the steel rolling heating furnace and the thickness of the steel billet, so the calculation of the temperature inside the steel billet is always an important problem in the field of the steel rolling heating furnace. Because the internal temperature of the billet cannot be directly measured, the calculation of the internal temperature is mainly based on a physical mechanism model at present and is completed by solving a thermodynamic partial differential equation. Three-dimensional mechanism models, although highly accurate, do not provide feedback to the control system in real time because the models are complex and require a significant amount of computation time. In order to shorten the solving time, a one-dimensional model is often adopted in industry, but the accuracy of the one-dimensional model is often not high due to simplification, and the effect for feedback control is not good.
The invention provides a rapid calculation method for the temperature of a steel billet by combining a forward mechanism and machine learning, which can calculate the internal temperature condition of the steel billet after heating in a steel rolling heating furnace. According to the invention, through a machine learning technology, the relation between the one-dimensional model and the three-dimensional model result is searched, and the three-dimensional model result can be reversely deduced through the real-time result of the one-dimensional model.
Disclosure of Invention
In order to solve the technical problems, the invention provides a rapid calculation method for the temperature of a steel billet by combining a forward mechanism and machine learning.
In order to solve the technical problems, the invention adopts the following technical scheme:
a rapid calculation method for the temperature of a steel billet by combining a forward mechanism and machine learning comprises the following steps:
step one: acquisition steelBlank
Production data during operation in a heating furnace, and establishing a discretized billet three-dimensional heat conduction model based on a finite difference method;
step two: solving the steel billet three-dimensional heat conduction model by a finite difference numerical value solving method to obtain a steel billet
Internal temperature at the outlet of the heating section +.>
；
Step three: repeating the first and second steps to obtain multiple billets
Internal temperature at the outlet of the heating section +.>
The billet is->
Internal temperature at the outlet of the heating section +.>
Steel billet->
The corresponding production data is used as a training sample and recorded in a training data set, and the corresponding production data is +.>
The number of training samples in the training dataset;
step four: establishing a one-dimensional model by using a finite element method, wherein the one-dimensional model assumes that a billet has infinite length and width, and only considers the condition that heat is transferred along the height direction of the billet;
step five: method for calculating steel billet by utilizing one-dimensional model and finite difference method
Internal temperature at the outlet of the heating section +.>
And the internal temperature is +.>
Added into steel billet->
Corresponding training samples;
step six: training by adopting an extreme gradient lifting method based on a training data set to obtain a regression tree machine learning model, wherein the input of the regression tree machine learning model is production data, and the output of the regression tree machine learning model is internal temperature
And internal temperature->
Is a difference in (2);
step seven: billet steel
In actual production, the steel billet is calculated>
A one-dimensional temperature matrix at the current moment; calculating the residual heating time of the billet according to the inlet distance of the billet from the heating section and the billet travelling speed; the residual heating time of the steel billet and a one-dimensional temperature matrix at the current moment of the steel billet are taken as initial temperature matrices and are input into a one-dimensional model, and the steel billet +_ in actual production is predicted>
Internal temperature at the outlet of the heating section +.>
；
Step eight: billet is made
Is produced by (a) a process for producingInputting the data into a regression tree machine learning model, and outputting the result of the regression tree machine learning model +.>
And->
Adding as billet->
And outputting the prediction result of the internal temperature in the heating section.
Further, the method comprises the step nine of: in actual production, for each billet in the heating furnace
And (d) repeating the step (eight) until the prediction results of the internal temperature of all billets are obtained when the billets go out of the heating section.
Further, if a plurality of heating sections exist in the heating furnace, repeating the steps one to nine for each heating section to obtain a prediction result of the internal temperature when the billet is discharged from each heating section.
Specifically, in the first step, the production data includes the positions of the billets, the temperatures of the temperature sensors in the top area, the bottom area, the left area and the right area of the hearth of the heating section, and the charging time of the billets
The tapping time of the billet is->
The size of the steel billet, the steel type of the steel billet, the travelling speed of the steel billet, and the temperature of the steel billet when entering the heating section; wherein the position of the billet is the distance between the billet and the inlet of the heating section; wherein the billet dimensions include the length, width and height of the billet;
the production data used in the third and sixth steps include temperatures of temperature sensors in the top, bottom, left and right areas of the heating section furnace, billet size, billet steel type, billet travel speed, and temperature of the billet as it enters the heating section.
Specifically, the billet three-dimensional heat conduction model selects the length direction of a billet as an x-axis, the width direction as a y-axis and the height direction as a z-axis; i represents an ith point in the billet longitudinal direction, j represents a jth point in the billet width direction, and k represents a kth point in the billet height direction; the equation for the billet three-dimensional heat conduction model is expressed as follows:
wherein ,
all are intermediate variables:
t is a three-dimensional temperature matrix,
representing the positions (i, j, k) in the three-dimensional temperature matrixThe temperature value of the point at the time t;
For the discretized time point, +.>
Is->
Time step of>
For the number of time steps>
The three-dimensional temperature matrix is the initial temperature matrix of the billet three-dimensional heat conduction model;
obtained by a billet three-dimensional heat conduction model
Time three-dimensional temperature matrix T, elements in three-dimensional temperature matrix T
I.e. the internal temperature of the billet when the billet is out of the heating section +.>
；
Specifically, the equation of the one-dimensional model is expressed as follows:
wherein 
Is a one-dimensional temperature matrix @>
The one-dimensional temperature matrix is the initial temperature matrix of the one-dimensional model,
representing the temperature of the kth point in the one-dimensional temperature matrix at time t, < >>
I.e. the internal temperature of the billet when the billet is out of the heating section +.>
。
Specifically, in step seven, the billet is calculated
When the one-dimensional temperature matrix at the current moment is used, the current moment and the charging time of the billet are added>
Dividing the difference by the time step +.>
The obtained result is used as a time step number N in a one-dimensional model; then using a finite difference method to take the furnace charging temperature of the billet as an initial temperature matrix of a one-dimensional model, and obtaining a result which is the billet
A one-dimensional temperature matrix at the current time.
Specifically, in the seventh step, the one-dimensional model is used for predicting the billet in actual production
Internal temperature at the time of exiting the heating section
At this time, the remaining heating time is related to the time step +.>
Ratio of (2) as the number of time steps of the one-dimensional model +.>
The method comprises the steps of carrying out a first treatment on the surface of the Step of time +.>
Steel billet->
Inputting a one-dimensional temperature matrix (initial temperature matrix as one-dimensional model) at the current moment into the one-dimensional model, and obtaining +.>
Namely steel billet in actual production>
Internal temperature at the outlet of the heating section +.>
。
Compared with the prior art, the invention has the beneficial technical effects that:
according to the invention, the relation between the one-dimensional model and the three-dimensional model result is searched through the machine learning technology, the three-dimensional model result is reversely deduced through the real-time result of the one-dimensional model, and the internal temperature condition of the billet after heating in the heating furnace can be calculated. The method solves the problem that the calculation efficiency and the solving precision can not be achieved when the mechanism model is used for calculating the internal temperature of the billet, and compared with a method for calculating by using the mechanism model alone, the method can give the result in a short time and can ensure the precision of the result; meanwhile, the state of the billet can be corrected according to the historical information of the billet in the furnace, and the corrected result is used for prediction, so that the influence of the temperature fluctuation of the heating furnace on the prediction result is reduced.
Drawings
FIG. 1 is a flow chart of the fast computing method of the present invention.
Detailed Description
A preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.
As shown in fig. 1, a method for quickly calculating a billet temperature by combining a forward mechanism and machine learning comprises the following steps:
s1: collecting steel billet
And (3) production data during operation in the heating furnace, and establishing a discretized billet three-dimensional heat conduction model based on a finite difference method.
Specifically, the production data comprises the positions of steel billets, the temperatures of temperature sensors in the top area, the bottom area, the left area and the right area of a heating section hearth, and the charging time of the steel billets
The tapping time of the billet is->
The size of the steel billet, the steel type of the steel billet, the travelling speed of the steel billet, and the temperature of the steel billet when entering the heating section; wherein the position of the billet is the distance between the billet and the inlet of the heating section; wherein the billet dimensions include the length, width and height of the billet. The billet position data and the temperature data are in the order of minutes, namely, are collected once per minute.
Finite difference method: a point is taken at intervals and the temperature around this point is considered to be the same as this point. By which the billet is formedThe temperature profile of the body is replaced by the temperature of the individual points, which are then described by means of a relevant model equation (e.g. a heat conduction equation). Substituting the initial temperature matrix into the equation to calculate, so as to obtain the point positions
The temperature after the time is then used as the initial temperature matrix to obtain the new temperature>
The temperature after the time, and so on, the temperature at the predetermined time can be predicted.
S2: solving the steel billet three-dimensional heat conduction model by a finite difference numerical value solving method to obtain a steel billet
Internal temperature at the outlet of the heating section +.>
。
In the invention, the billet three-dimensional heat conduction model selects the length direction of the billet as an x-axis, the width direction as a y-axis and the height direction as a z-axis; i represents an i-th point in the billet longitudinal direction (x-axis), j represents a j-th point in the billet width direction (y-axis), and k represents a k-th point in the billet height direction (z-axis); i. j and k are counted from 0, and the equation of the billet three-dimensional heat conduction model is expressed as follows:
wherein ,
all are intermediate variables, which are calculated by the following modes and then substituted into the corresponding equation: />
T is a three-dimensional temperature matrix,
a temperature value at time t representing a point of the three-dimensional temperature matrix at a position (i, j, k);
For the discretized time point, each increase of t by 1 indicates +.>
Time of (2)>
For the time step in the finite difference method, 1-3 seconds is typically taken.
The three-dimensional temperature matrix is the initial temperature matrix of the billet three-dimensional heat conduction model. Except S7, the temperature of each point in the initial temperature matrix in other steps is the temperature when the billet enters the heating section; time step number +.>
The total duration of the billet (in-furnace time-out-of-furnace time) can be divided by +.>
The method comprises the following steps:
obtained by a billet three-dimensional heat conduction model
Time three-dimensional temperature matrix T, elements in three-dimensional temperature matrix T
I.e. the internal temperature of the billet when the billet is out of the heating section +.>
。
S3: repeating S1 and S2 to obtain a plurality of billets
Internal temperature at the outlet of the heating section +.>
The billet is->
Internal temperature at the outlet of the heating section +.>
Steel billet->
The corresponding production data is used as a training sample and recorded in a training data set, and the corresponding production data is +.>
The number of training samples in the training dataset; specifically, the steel billet type, the travelling speed of the steel billet, the temperature of the steel billet when entering the heating section, the temperature of the temperature sensors of the top area, the bottom area, the left area and the right area of the hearth of the heating section, the length, the width and the height of the steel billet are added into the training data set.
S4: a one-dimensional model is established by using a finite element method, wherein the one-dimensional model assumes that a billet has infinite length and width, and only considers the condition that heat is transferred along the height direction of the billet.
The equation for the one-dimensional model is expressed as follows:
wherein 
Is a one-dimensional temperature matrix @>
The one-dimensional temperature matrix is the initial temperature matrix of the one-dimensional model,
representing the temperature of the kth point in the one-dimensional temperature matrix at time t, < >>
I.e. the internal temperature of the billet when it goes out of the heating section
. The meaning of the remaining variables refers to the billet three-dimensional heat conduction model.
S5: method for calculating steel billet by utilizing one-dimensional model and finite difference method
Internal temperature at the outlet of the heating section +.>
And the internal temperature is +.>
Added into steel billet->
And in the corresponding training samples.
S6: training by adopting an extreme gradient lifting method (xgboost) based on a training data set to obtain a regression tree machine learning model, wherein the input of the regression tree machine learning model is the temperature of temperature sensors in the top area, the bottom area, the left area and the right area of a heating section hearth, the size of a billet, the steel grade of the billet, the travelling speed of the billet and the temperature when the billet enters the heating section; the output of the regression tree machine learning model is the internal temperature
And internal temperature->
Is a difference in (c).
S7: billet steel
In actual production, the steel billet is calculated>
A one-dimensional temperature matrix at the current moment; calculating the residual heating time of the steel billet according to the inlet distance of the steel billet from the heating section and the travelling speed of the steel billet: remaining heating time= (total length of heating section-billet position)/billet travel speed; predicting billet +_ in actual production by using one-dimensional model based on one-dimensional temperature matrix of current moment of billet and residual heating time of billet>
Internal temperature at the outlet of the heating section +.>
。
Specifically, the steel billet is calculated
When the one-dimensional temperature matrix at the current moment is used, the current moment and the charging time of the billet are calculated
Dividing the difference by the time step +.>
The obtained result is used as a time step number N in a one-dimensional model; then using the finite difference method to take the temperature of the billet entering the furnace as the initial temperature matrix of the one-dimensional model, and obtaining the result of the one-dimensional model, namely the billet +.>
A one-dimensional temperature matrix at the current time.
Specifically, predicting billet in actual production by using one-dimensional model
Internal temperature at the outlet of the heating section +.>
At this time, the remaining heating time is related to the time step +.>
Ratio of (2) as the number of time steps of the one-dimensional model +.>
The method comprises the steps of carrying out a first treatment on the surface of the Step of time +.>
Steel billet->
Inputting a one-dimensional temperature matrix at the current moment into a one-dimensional model to obtain +.>
Namely steel billet in actual production>
Internal temperature at the outlet of the heating section +.>
. Since the future temperature in the prediction process is not measurable, +.>
The temperature value of the sensor corresponding to the current moment is taken at different t time points.
S8: billet is made
Is input into the regression tree machine learning model, and the output result of the regression tree machine learning model is +.>
And->
Adding as billet->
And outputting the prediction result of the internal temperature in the heating section.
S9: in actual production, for each billet in the heating furnace
And S8, repeating until the prediction results of the internal temperature of all billets are obtained when the billets are discharged from the heating section.
And if a plurality of heating sections exist in the heating furnace, repeating S1 to S8 for each heating section to obtain a prediction result of the internal temperature when the billet is discharged from each heating section.
In some cases, if the sensor temperature is unstable, the average of the time from the billet entering the heating section to the billet exiting the heating section is taken to replace the unstable sensor temperature.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
Furthermore, it should be understood that although the present disclosure describes embodiments, not every embodiment is provided with a single embodiment, and that this description is provided for clarity only, and that the disclosure is not limited to specific embodiments, and that the embodiments may be combined appropriately to form other embodiments that will be understood by those skilled in the art.
Claims (8)
1. A rapid calculation method for the temperature of a steel billet by combining a forward mechanism and machine learning comprises the following steps:
step one: collecting steel billet
Production data during operation in a heating furnace, and establishing a discretized billet three-dimensional heat conduction model based on a finite difference method;
step two: solving the steel billet three-dimensional heat conduction model by a finite difference numerical value solving method to obtain a steel billet
Internal temperature at the outlet of the heating section +.>
；
Step three: repeating the first and second steps to obtain multiple billets
Internal temperature at the time of exiting the heating section
The billet is->
Internal temperature at the outlet of the heating section +.>
Steel billet->
The corresponding production data is used as a training sample and recorded in a training data set;
The number of training samples in the training dataset;
step four: establishing a one-dimensional model by using a finite element method, wherein the one-dimensional model assumes that a billet has infinite length and width, and only considers the condition that heat is transferred along the height direction of the billet;
step five: method for calculating steel billet by utilizing one-dimensional model and finite difference method
Internal temperature at the outlet of the heating section +.>
And the internal temperature is +.>
Added into steel billet->
Corresponding training samples;
step six: training by adopting an extreme gradient lifting method based on a training data set to obtain a regression tree machine learning model, wherein the input of the regression tree machine learning model is production data, and the output of the regression tree machine learning model is internal temperature
And internal temperature->
Is a difference in (2);
step seven: billet steel
In actual production, the steel billet is calculated>
A one-dimensional temperature matrix at the current moment; calculating the residual heating time of the billet according to the inlet distance of the billet from the heating section and the billet travelling speed; the residual heating time of the billet is taken as an initial temperature matrix, and a one-dimensional temperature matrix at the current moment of the billet is input into a one-dimensional model for pre-treatmentMeasuring steel billet->
Internal temperature at the outlet of the heating section +.>
；
Step eight: billet is made
Is input into the regression tree machine learning model, and the output result of the regression tree machine learning model is +.>
And->
Adding as billet->
And outputting the prediction result of the internal temperature in the heating section.
2. The method of claim 1, comprising the step of: in actual production, for each billet in the heating furnace
And (d) repeating the step (eight) until the prediction results of the internal temperature of all billets are obtained when the billets go out of the heating section.
3. The method according to claim 2, wherein if there are a plurality of heating sections in the heating furnace, repeating steps one to nine for each heating section to obtain a predicted result of the internal temperature of the billet when it exits each heating section.
4. According to claim 1The rapid calculation method for the temperature of the steel billet by fusing the forward mechanism and the machine learning is characterized by comprising the following steps of: in the first step, the production data comprise the positions of steel billets, the temperatures of temperature sensors in the top area, the bottom area, the left area and the right area of a heating section hearth, and the charging time of the steel billets
The tapping time of the billet is->
The size of the steel billet, the steel type of the steel billet, the travelling speed of the steel billet, and the temperature of the steel billet when entering the heating section; wherein the position of the billet is the distance between the billet and the inlet of the heating section; wherein the billet dimensions include the length, width and height of the billet;
the production data used in the third and sixth steps include temperatures of temperature sensors in the top, bottom, left and right areas of the heating section furnace, billet size, billet steel type, billet travel speed, and temperature of the billet as it enters the heating section.
5. The method for quickly calculating the temperature of the steel billet by combining the forward mechanism and the machine learning according to claim 4, wherein the method comprises the following steps of: the billet three-dimensional heat conduction model selects the length direction of a billet as an x-axis, the width direction as a y-axis and the height direction as a z-axis; i represents an ith point in the billet longitudinal direction, j represents a jth point in the billet width direction, and k represents a kth point in the billet height direction; the equation for the billet three-dimensional heat conduction model is expressed as follows:
wherein ,
all are intermediate variables:
t is a three-dimensional temperature matrix,
a temperature value at time t representing a point of the three-dimensional temperature matrix at a position (i, j, k);
for the discretized time point, +.>
Is->
Time step of>
For the number of time steps>
The three-dimensional temperature matrix is the initial temperature matrix of the billet three-dimensional heat conduction model;
obtained by a billet three-dimensional heat conduction model
Time three-dimensional temperature matrix T, elements in three-dimensional temperature matrix T
I.e. the internal temperature of the billet when the billet is out of the heating section +.>
；
6. The method for quickly calculating the temperature of the steel billet by combining the forward mechanism and the machine learning according to claim 5, wherein the method comprises the following steps of: the equation for the one-dimensional model is expressed as follows:
wherein 
Is a one-dimensional temperature matrix @>
The one-dimensional temperature matrix is the initial temperature matrix of the one-dimensional model,>
representing the temperature of the kth point in the one-dimensional temperature matrix at time t, < >>
I.e. the internal temperature of the billet when the billet is out of the heating section +.>
。
7. The method for quickly calculating the temperature of the steel billet by combining the forward mechanism and the machine learning according to claim 6, wherein the method comprises the following steps of: step seven, calculating the billet
When the one-dimensional temperature matrix at the current moment is used, the current moment and the charging time of the billet are calculated
Dividing the difference by the time step +.>
The obtained result is used as a time step number N in a one-dimensional model; and then utilize finite differenceThe method takes the furnace-entering temperature of the billet as an initial temperature matrix of a one-dimensional model, and the obtained result is the billet +.>
A one-dimensional temperature matrix at the current time.
8. The method for rapid calculation of slab temperature by combining forward mechanism and machine learning as claimed in claim 6, wherein in step seven, slab temperature during actual production is predicted by using one-dimensional model
Internal temperature at the outlet of the heating section +.>
At this time, the remaining heating time is related to the time step +.>
Ratio of (2) as the number of time steps of the one-dimensional model +.>
The method comprises the steps of carrying out a first treatment on the surface of the Step of time +.>
Steel billet
Inputting a one-dimensional temperature matrix at the current moment into a one-dimensional model to obtain +.>
Namely steel billet in actual production>
Internal temperature at the outlet of the heating section +.>
。/>
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310110231.7A CN115828772B (en) | 2023-02-14 | 2023-02-14 | Rapid calculation method for billet temperature by combining forward mechanism and machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310110231.7A CN115828772B (en) | 2023-02-14 | 2023-02-14 | Rapid calculation method for billet temperature by combining forward mechanism and machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115828772A CN115828772A (en) | 2023-03-21 |
| CN115828772B true CN115828772B (en) | 2023-05-09 |
Family
ID=85521205
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310110231.7A Active CN115828772B (en) | 2023-02-14 | 2023-02-14 | Rapid calculation method for billet temperature by combining forward mechanism and machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115828772B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10533937B1 (en) * | 2018-08-30 | 2020-01-14 | Saudi Arabian Oil Company | Cloud-based machine learning system and data fusion for the prediction and detection of corrosion under insulation |
| CN112102890A (en) * | 2020-07-28 | 2020-12-18 | 浙江中控技术股份有限公司 | MCS synthesis device reaction temperature prediction method based on machine learning model |
| CN112417769A (en) * | 2020-12-08 | 2021-02-26 | 江苏科技大学 | Data-driven high-supercharged engine air flue structure optimization design method |
| CN113102516A (en) * | 2021-03-05 | 2021-07-13 | 东北大学 | Hot continuous rolling strip steel head width prediction method integrating rolling mechanism and deep learning |
| CN113821984A (en) * | 2021-10-18 | 2021-12-21 | 重庆赛迪热工环保工程技术有限公司 | Heating furnace steel billet temperature calculation method based on time domain convolution model |
| CN114898821A (en) * | 2022-05-13 | 2022-08-12 | 宁波博威合金材料股份有限公司 | Alloy performance prediction method based on machine learning model fusion |
| CN115141926A (en) * | 2022-06-23 | 2022-10-04 | 大连新瑞晨自动化科技有限公司 | Billet target temperature rise curve generation method for guiding billet heating process control |
| CN115293827A (en) * | 2022-10-09 | 2022-11-04 | 浩鲸云计算科技股份有限公司 | Novel model interpretability analysis method for assisting fine operation of enterprise |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US12032359B2 (en) * | 2021-02-25 | 2024-07-09 | Tata Consultancy Services Limited | Prediction of preheat refractory temperature profile of a ladle furnace |
-
2023
- 2023-02-14 CN CN202310110231.7A patent/CN115828772B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10533937B1 (en) * | 2018-08-30 | 2020-01-14 | Saudi Arabian Oil Company | Cloud-based machine learning system and data fusion for the prediction and detection of corrosion under insulation |
| CN112102890A (en) * | 2020-07-28 | 2020-12-18 | 浙江中控技术股份有限公司 | MCS synthesis device reaction temperature prediction method based on machine learning model |
| CN112417769A (en) * | 2020-12-08 | 2021-02-26 | 江苏科技大学 | Data-driven high-supercharged engine air flue structure optimization design method |
| CN113102516A (en) * | 2021-03-05 | 2021-07-13 | 东北大学 | Hot continuous rolling strip steel head width prediction method integrating rolling mechanism and deep learning |
| CN113821984A (en) * | 2021-10-18 | 2021-12-21 | 重庆赛迪热工环保工程技术有限公司 | Heating furnace steel billet temperature calculation method based on time domain convolution model |
| CN114898821A (en) * | 2022-05-13 | 2022-08-12 | 宁波博威合金材料股份有限公司 | Alloy performance prediction method based on machine learning model fusion |
| CN115141926A (en) * | 2022-06-23 | 2022-10-04 | 大连新瑞晨自动化科技有限公司 | Billet target temperature rise curve generation method for guiding billet heating process control |
| CN115293827A (en) * | 2022-10-09 | 2022-11-04 | 浩鲸云计算科技股份有限公司 | Novel model interpretability analysis method for assisting fine operation of enterprise |
Non-Patent Citations (2)
| Title |
|---|
| Runhao Zhang，etc..Comparison of the Prediction of BOF End-Point Phosphorus Content Among Machine Learning Models and Metallurgical Mechanism Model.《steel research int.》.2022,1-12. * |
| 肖立志.机器学习数据驱动与机理模型融合及可解释性问题.《石油物探》.2022,205-212. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115828772A (en) | 2023-03-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105045949B (en) | A kind of walking beam furnace steel billet temperature modeling and on-line correction method | |
| CN107042234B (en) | Intelligent technique production method based on the acquisition of bar whole process big data | |
| WO2023130666A1 (en) | Strip steel plate convexity prediction method based on data-driving and mechanism model fusion | |
| CN101509812A (en) | Soft measurement method for billet temperature distribution in smelting and heating-furnace | |
| CN103212585A (en) | A control device of a hot mill used for thin plates and a control method of the hot mill used for thin plates | |
| CN104894362A (en) | Method for setting temperature of heating furnace in cold and hot steel billet mixed loading | |
| CN101837373A (en) | Hot-rolling analogue device and rolling history recording analogue method | |
| TW200806984A (en) | Rolling line material quality prediction and control apparatus | |
| CN103447314B (en) | Rough rolling temperature-measuring feedback control method | |
| CN104881707B (en) | A kind of sintering energy consumption Forecasting Methodology based on integrated model | |
| CN114764547A (en) | Heating furnace steel billet temperature prediction method and device integrating mechanism and deep learning | |
| CN105677949A (en) | Method for predicting TC6 titanium alloy forging piece microstructural parameters | |
| CN109013717B (en) | A kind of hot continuous rolling centre base center portion temperature computation method | |
| CN107999547B (en) | Laminar cooling self-learning method and device | |
| CN115828772B (en) | Rapid calculation method for billet temperature by combining forward mechanism and machine learning | |
| JP4402502B2 (en) | Winding temperature controller | |
| CN1641356B (en) | Rigid line product microscopic structure and mechanical property forecasting system | |
| CN115470595A (en) | Hot-rolled strip steel convexity prediction method with data and mechanism fused | |
| CN108984943B (en) | Heating furnace billet temperature tracking model correction method | |
| CN111814402B (en) | Heating furnace temperature control method | |
| CN114091352A (en) | Hot continuous rolling outlet plate convexity dynamic prediction method based on Elman neural network | |
| CN117494531B (en) | Medium carbon steel decarburization depth prediction method based on finite element and XGBoost algorithm | |
| CN208162301U (en) | A kind of self study device of section cooling | |
| CN103593493B (en) | Reducer quality prediction method based on integrated gradient data ELM-PLS method | |
| Zanoli et al. | Analysis and modeling of steel industry reheating furnace billets temperature |
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 |