CN113063916A - Hot-dip galvanized strip steel coating aluminum content prediction method based on PSO-SVR model - Google Patents
Hot-dip galvanized strip steel coating aluminum content prediction method based on PSO-SVR model Download PDFInfo
- Publication number
- CN113063916A CN113063916A CN202110266302.3A CN202110266302A CN113063916A CN 113063916 A CN113063916 A CN 113063916A CN 202110266302 A CN202110266302 A CN 202110266302A CN 113063916 A CN113063916 A CN 113063916A
- Authority
- CN
- China
- Prior art keywords
- strip steel
- particle
- aluminum content
- svr
- coating
- 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
- 229910052782 aluminium Inorganic materials 0.000 title claims abstract description 76
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 title claims abstract description 76
- 239000010959 steel Substances 0.000 title claims abstract description 69
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 68
- 239000011248 coating agent Substances 0.000 title claims abstract description 58
- 238000000576 coating method Methods 0.000 title claims abstract description 58
- 238000000034 method Methods 0.000 title claims abstract description 26
- 239000002245 particle Substances 0.000 claims abstract description 60
- 238000012706 support-vector machine Methods 0.000 claims abstract description 17
- 238000005457 optimization Methods 0.000 claims abstract description 14
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 13
- HCHKCACWOHOZIP-UHFFFAOYSA-N Zinc Chemical compound [Zn] HCHKCACWOHOZIP-UHFFFAOYSA-N 0.000 claims description 47
- 239000011701 zinc Substances 0.000 claims description 46
- 229910052725 zinc Inorganic materials 0.000 claims description 46
- 238000005070 sampling Methods 0.000 claims description 30
- 238000001514 detection method Methods 0.000 claims description 19
- 238000012549 training Methods 0.000 claims description 19
- 239000007788 liquid Substances 0.000 claims description 13
- 238000012360 testing method Methods 0.000 claims description 8
- 238000012216 screening Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000005259 measurement Methods 0.000 claims description 4
- 239000000203 mixture Substances 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 239000002436 steel type Substances 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 abstract description 7
- 230000006870 function Effects 0.000 description 17
- 239000002893 slag Substances 0.000 description 7
- 239000004411 aluminium Substances 0.000 description 5
- 229910001335 Galvanized steel Inorganic materials 0.000 description 2
- 239000008397 galvanized steel Substances 0.000 description 2
- 238000005246 galvanizing Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 239000010410 layer Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000011247 coating layer Substances 0.000 description 1
- 238000005097 cold rolling Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/20—Metals
- G01N33/202—Constituents thereof
- G01N33/2028—Metallic constituents
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Engineering & Computer Science (AREA)
- Food Science & Technology (AREA)
- Medicinal Chemistry (AREA)
- Physics & Mathematics (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Coating With Molten Metal (AREA)
Abstract
The invention discloses a hot-dip galvanized strip steel coating aluminum content prediction method based on a PSO-SVR model, which belongs to the technical field of intelligent manufacturing, wherein a penalty function C and a bandwidth sigma of a Gaussian kernel function in a support vector machine regression are optimized through a particle swarm optimization algorithm, so that a more accurate support vector machine regression model is established, the accuracy of the hot-dip galvanized strip steel coating aluminum content prediction result is improved, the SVR model established based on the support vector machine regression can well act on a nonlinear system, and the method has good generalization capability.
Description
Technical Field
The invention belongs to the technical field of intelligent manufacturing, and particularly relates to prediction of aluminum content of a strip steel coating in a hot galvanizing process.
Background
Aluminium is hot-dip galvanized crude productThe very critical metal element in production, the aluminum content in the zinc pot determines the structure of the coating and the behavior of the zinc bath in the zinc pot. In actual industrial mass production, the effective aluminum content of the molten zinc is controlled by a target value, but factors influencing the effective aluminum in the molten zinc are more, the effective aluminum content is difficult to accurately control through an ingot adding strategy, but t is known through analysis1At the moment, the total aluminum (effective aluminum and slag aluminum) + (t) in the zinc liquid2-t1) Aluminum- (t) contained in zinc ingot melted at any time2-t1) Aluminium- (t) contained in zinc slag of instant slag-dragging2-t1) T is the aluminium taken away by the strip steel coating at the moment2Total aluminium in the zinc pot at time (effective aluminium + slag aluminium), where t2>t1The aluminum content in the zinc liquid can be measured by an online detection device, the slag dragging operation is regular, the aluminum taken away by the slag in a certain time can be known, and an ingot adding strategy can be obtained by predicting the aluminum content taken away by a strip steel coating, so that t is realized2The accurate control of the aluminum content at the moment, so that the prediction of the aluminum content of the strip steel coating becomes a key.
At present, the aluminum content of a coating is mainly measured by a spectrometer in a laboratory after field sampling, and the offline measurement mode cannot provide real-time data support for a production field ingot adding strategy.
Disclosure of Invention
Aiming at the defects or improvement requirements in the prior art, the invention provides a hot-dip galvanized strip steel coating aluminum content prediction method based on a PSO-SVR model, which can make full use of field acquired data and predict the strip steel coating aluminum content by means of a data analysis model.
In order to achieve the aim, the invention provides a hot-dip galvanized strip steel coating aluminum content prediction method based on a PSO-SVR model, which comprises the following steps:
at a sampling frequency f1Sampling the detection parameters of the zinc pot area with a sampling frequency f2Sampling the galvanized strip steel and detecting the aluminum content of the coating, taking the average value of the detection parameters of the zinc pot area between the current strip steel sampling and the last strip steel sampling as the value of the detection parameters of the zinc pot area during the current strip steel sampling, and selecting the strip steel within the sampling timeForming an initial sample set by using the strip steel coating aluminum content data and zinc pot area detection data corresponding to the coating aluminum content data for a period of time;
carrying out data preprocessing on an initial sample set, dividing the initial sample set into a training set and a testing set, and selecting strip steel coating aluminum content data as decision attribute data and condition attribute data related to the strip steel coating aluminum content in zinc pot area detection parameters;
training a support vector machine regression SVR model by adopting a training set, and testing the support vector machine regression SVR model by adopting a testing set to obtain an SVR prediction model, wherein the input of the SVR prediction model is condition attribute data related to the aluminum content of a strip steel coating, and the output of the SVR prediction model is strip steel coating aluminum content data;
and globally optimizing the penalty factor parameter C and the kernel width parameter sigma in the SVR prediction model by utilizing a particle swarm optimization algorithm to obtain a proper model parameter, so that the SVR prediction model predicts the aluminum content of the band steel coating through the optimized penalty factor parameter C and the kernel width parameter sigma.
In some optional embodiments, the globally optimizing the penalty factor parameter C and the kernel width parameter σ in the SVR prediction model by using a particle swarm optimization algorithm to obtain suitable model parameters includes:
(a) penalty factor parameter C and kernel function parameter sigma of initialized SVR model, particle position range, particle speed range, initialized particle number N', iteration number I, local parameter C1Global search capability parameter c2And randomly generating initial positions x of all particlespd 0And velocity vpd 0Initializing the best position p of the current particlepdAnd the best position p in the whole populationgd；
(b) Substituting the parameters of each particle into the SVR model, calculating the fitness of each particle, and combining the current position of each particle with ppdComparing the best position of each particle in the population with pgdComparing, if the current position of the particle is optimal, replacing p with the current positiongdElse pgdIs kept unchanged；
(c) Updating the position x of the particle of the current iterationpd iAnd velocity vpd iJudging whether the iteration times are reached, if so, outputting the optimal parameters of the penalty factor parameter C and the kernel function parameter sigma, establishing an SVR model, and if not, returning to the step (b) for iterative calculation.
In some alternative embodiments, the composition is prepared by
Determining fitness f of ith iteration of p-th particlep i，p＝1,2，…，N'，ypThe actual value of the aluminum content of the coating of the strip steel is shown,
and the predicted value of the content of the aluminum of the coating output according to the condition attribute data at the pth in the training set is shown.
In some alternative embodiments, the composition is prepared by
Updating the velocity v of the particle of the current iterationpd iFrom
Updating the position x of the particle of the current iterationpd iWhere i denotes the current number of iterations,
which represents the velocity of the particles to be updated,
indicating the position of the particle to be updated, ppdRepresenting the individual optimum value, p, of the particle to be updatedgdRepresenting the global optimum of the entire population, rand1 and rand2 are [0,1 ]]Random numbers within the range, w is the inertial weight,
i denotes the maximum number of iterations, wmaxRepresenting the maximum value of the inertial weight, wminRepresenting the inertial weight minimum.
In some alternative embodiments, the SVR prediction model is:
wherein, betaqAnd alphaqSolving for Lagrange multiplier by SMO algorithm, wherein sigma is bandwidth of Gaussian kernel function, b is offset, and solving according to KKT optimal condition, and xqIs the condition attribute data in the qth training set, N is the number of training sets, yqAnd (4) obtaining decision attribute data for prediction, namely the aluminum content of the coating output by the target.
In some optional embodiments, the pre-processing the data of the initial sample set comprises:
and manually screening values of the data with the missing values, manually screening unreasonable data, and normalizing the data.
In some alternative embodiments, the condition attribute data related to the aluminum content of the strip coating is reduced by using a roughness set, and the condition attribute data related to the aluminum content of the strip coating comprises: zinc liquid composition, zinc liquid temperature, strip steel zinc pot temperature, strip steel running speed, strip steel thickness, upper and lower surface zinc layer measurement thickness, strip steel width and strip steel type data.
In general, compared with the prior art, the above technical solution contemplated by the present invention can achieve the following beneficial effects:
according to the technical scheme, the penalty function C in the regression of the support vector machine and the bandwidth sigma of the Gaussian kernel function are optimized through the particle swarm optimization algorithm, so that a more accurate regression model of the support vector machine is established, and the accuracy of the prediction result of the aluminum content of the hot-dip galvanized strip steel coating is improved. The SVR model established based on the support vector machine regression can well act on a nonlinear system, and has good generalization capability.
Drawings
FIG. 1 is a schematic flow chart of a method for predicting the aluminum content of a coating of a hot-dip galvanized steel strip based on a PSO-SVR model according to an embodiment of the present invention;
fig. 2 is a schematic flow chart illustrating an optimization process of a particle swarm optimization algorithm on a target parameter according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
Fig. 1 is a schematic flow chart of a method for predicting the aluminum content of a coating layer of a hot-dip galvanized steel strip based on a PSO-SVR model according to an embodiment of the present invention, where the method shown in fig. 1 includes the following steps:
step 1: constructing an initial sample set at a sampling frequency f1Sampling the detection parameters of the zinc pot area with a sampling frequency f2Sampling the galvanized strip steel and detecting the aluminum content of the coating, taking the average value of the zinc pot area detection parameters between the strip steel sampling and the strip steel sampling last time as the value of the zinc pot area detection parameters during the strip steel sampling, and selecting strip steel coating aluminum content data within a period of time within the sampling time and zinc pot area detection data corresponding to the coating aluminum content data to form an initial sample set.
In the embodiment of the invention, a hot galvanizing production line of a certain cold rolling mill in China is taken as an application object, actual production data from 11/month 1 in 2020 to 11/month 30 in 2020 is selected, wherein the sampling frequency f of detection data of a zinc pot area is11 time in 30 seconds, sampling frequency f of the strip steel2Is not fixed. Because the accurate aluminum content of the strip steel coating can be obtained only by sampling the tail part of a coil of strip steel and measuring the aluminum content by using a spectrometer in a laboratory, the problem that the aluminum content of the strip steel coating is not matched with the data detected in a zinc pot area in time exists, and the strip steel coating is subjected to two timesAnd taking the average value of the detection parameters of the zinc pot area between the samplings as the value of the detection parameters of the zinc pot area during the current strip steel sampling so as to obtain an initial sample set.
In the embodiment of the invention, the detection parameters of the zinc pot area comprise 10 index data such as the effective aluminum content of zinc liquid, the slag aluminum content in the zinc liquid, the zinc liquid temperature, the strip steel zinc pot temperature, the strip steel running speed, the strip steel thickness, the strip steel width and the like, and the corresponding coating aluminum content. The method comprises the steps of reducing 10 index data by adopting a rough set, selecting 8 indexes related to the aluminum content of a coating, establishing a PSO-SVR prediction model, wherein the indexes comprise zinc liquid components (the effective aluminum content of the zinc liquid and the content of slag aluminum in the zinc liquid), zinc liquid temperature, strip steel zinc pot temperature, strip steel running speed, strip steel thickness, upper and lower surface zinc layer measurement thickness, strip steel width and steel type, the 8 index data are condition attribute data, the coating aluminum content data are decision attribute data, and 1328 initial samples are obtained in total.
Step 2: the method comprises the steps of preprocessing data of an initial sample set, dividing the initial sample set into a training set and a testing set according to the proportion of 8:2, manually screening and dereferencing data with missing values, manually screening and deleting unreasonable data, normalizing the data, and dividing 1328 sample data into the training set and the testing set according to the proportion of 8: 2.
And step 3: modeling by a Support Vector Regression (SVR), inputting the training set divided in the step2 into the SVR, and training the model:
assuming that the prediction data set of the aluminum content of the strip steel coating to be detected is D, the method comprises the following steps:
D＝{(x1,y1),(x2,y2),(x3,y3),……(xm,ym)}；
wherein x ismFor 8-dimensional conditional attribute data, i.e. input quantity, ymThe decision attribute data of the aluminum content of the coating is shown, m is the number of training set samples, and the regression function of the SVR support vector machine can be defined as: and f (x) ω · x + b, where ω and b are the corresponding weight coefficient and threshold of the SVR regression machine, respectively.
Using epsilon to represent a loss function to optimize the prediction model, searching a regression function with the best parameter value through the minimum value of the function,
the constraint conditions are as follows:
wherein C is a penalty factor, ξq,
N is the number of training sets.
Linear regression function can be obtained by operation
Wherein, betaqAnd alphaqIs Lagrange multiplier, K (x, x)q) For the kernel function of the support vector machine, the present embodiment uses the Gaussian kernel function
A predictive model of SVR can be derived:
βqand alphaqAnd b, solving by using an SMO algorithm, wherein the offset is solved according to the KKT optimal condition, and the regression performance of the support vector machine depends on a penalty parameter C and the bandwidth sigma of a Gaussian kernel function.
And 4, step 4: and (4) SVR parameter optimization, wherein a particle swarm optimization algorithm is utilized to perform global optimization on a penalty factor parameter C and a Gaussian kernel function bandwidth sigma in a regression model of the support vector machine so as to obtain a proper model parameter.
In the embodiment of the present invention, as shown in fig. 2, step4 may be implemented by:
step 1: penalty factor parameter for initializing SVR modelC and kernel function parameter σ, particle position range [ X ]min,Xmax]Particle velocity range [ V ]min,Vmax]The number of initialization particles N' is 20, the number of iterations I, and the local parameter c1Global search capability parameter c ═ 1.521.7, and randomly generating the position x of all particlespd 0And velocity vpd 0Initializing the best position p of the current particlepdAnd the best position p in the whole populationgd；
Step 2: substituting the parameters of each particle into the SVR model, and calculating the fitness f of each particlep iThe fitness is calculated according to the formula
Denotes the fitness of the ith iteration of the p-th particle, p ═ 1,2, …, N', ypThe actual value of the aluminum content of the coating of the strip steel is shown,
the predicted value of the aluminum content of the coating output according to the condition attribute data of the pth training set is represented;
step 3: the current position of each particle and the best position of the particle in the population are compared with ppdAnd pgdComparing, if the current position of the particle is optimal, replacing p with the current positiongdElse pgdKeeping the same;
step 4: updating the position x of a particlepd iAnd velocity vpd i：
Where i denotes the current number of iterations,
which represents the velocity of the particles to be updated,
indicating the position of the particle to be updated, ppdRepresenting the individual optimum value, p, of the particle to be updatedgdA global optimum representing the entire population of particles; rand1 and rand2 are [0,1 ]]Random number within the range, w is the inertial weight:
wherein, I represents the maximum iteration number, and w is takenmax＝0.9，wmin＝0.4。
Step 5: judging whether the iteration times are reached, if so, outputting the optimal parameters of the penalty factor parameter C and the kernel function parameter sigma, establishing an SVR model, otherwise, returning to step2 for iterative calculation.
The value range of the penalty factor C of the SVR network is 1-100000, the smaller the value is, the smaller the penalty of the empirical error is, the smaller the model complexity is, otherwise, the model complexity is high, and the data fitting degree is high. For the embodiment, after the particle swarm optimization, the SVR network obtains the best prediction performance when C is 5.3199 and σ is 0.01.
And 5: substituting the penalty factor parameter C and the kernel width parameter sigma in the step4 into the SVR prediction model in the step3 to predict the aluminum content of the strip steel coating, wherein the accuracy rate of the obtained prediction result within 0.1% is 89.782%;
the control method is used for programming on an MATLAB R2013 platform and carrying out algorithm simulation on sample data, wherein the SVR is realized by using a libsvm tool box, and the experimental environment of the MATLAB is Intel (R) core (TM) i5-4590 MCPU @3.3GHz and 4G internal memory.
In summary, by the aid of the PSO-SVR model-based hot-dip galvanized strip steel coating aluminum content prediction method, the penalty function C in the support vector machine regression and the bandwidth sigma of the Gaussian kernel function are optimized through the particle swarm optimization algorithm, so that a more accurate support vector machine regression model is established, accuracy of the hot-dip galvanized strip steel coating aluminum content prediction result is improved, the SVR model established based on the support vector machine regression can well act on a nonlinear system, and the method has good generalization capability.
It should be noted that, according to the implementation requirement, each step/component described in the present application can be divided into more steps/components, and two or more steps/components or partial operations of the steps/components can be combined into new steps/components to achieve the purpose of the present invention.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (7)
1. A hot-dip galvanized strip steel coating aluminum content prediction method based on a PSO-SVR model is characterized by comprising the following steps:
at a sampling frequency f1Sampling the detection parameters of the zinc pot area with a sampling frequency f2Sampling the galvanized strip steel and detecting the aluminum content of a coating, taking the average value of the detection parameters of the zinc pot area between the current strip steel sampling and the last strip steel sampling as the value of the detection parameters of the zinc pot area during the current strip steel sampling, and selecting strip steel coating aluminum content data within a period of time within the strip steel sampling time and zinc pot area detection data corresponding to the coating aluminum content data to form an initial sample set;
carrying out data preprocessing on an initial sample set, dividing the initial sample set into a training set and a testing set, and selecting strip steel coating aluminum content data as decision attribute data and condition attribute data related to the strip steel coating aluminum content in zinc pot area detection parameters;
training a support vector machine regression SVR model by adopting a training set, and testing the support vector machine regression SVR model by adopting a testing set to obtain an SVR prediction model, wherein the input of the SVR prediction model is condition attribute data related to the aluminum content of a strip steel coating, and the output of the SVR prediction model is strip steel coating aluminum content data;
and globally optimizing the penalty factor parameter C and the kernel width parameter sigma in the SVR prediction model by utilizing a particle swarm optimization algorithm to obtain a proper model parameter, so that the SVR prediction model predicts the aluminum content of the band steel coating through the optimized penalty factor parameter C and the kernel width parameter sigma.
2. The method according to claim 1, wherein the global optimization of the penalty factor parameter C and the kernel width parameter σ in the SVR prediction model by using the particle swarm optimization algorithm to obtain suitable model parameters comprises:
(a) penalty factor parameter C and kernel function parameter sigma of initialized SVR model, particle position range, particle speed range, initialized particle number N', iteration number I, local parameter C1Global search capability parameter c2And randomly generating initial positions x of all particlespd 0And velocity vpd 0Initializing the best position p of the current particlepdAnd the best position p in the whole populationgd；
(b) Substituting the parameters of each particle into the SVR model, calculating the fitness of each particle, and combining the current position of each particle with ppdComparing the best position of each particle in the population with pgdComparing, if the current position of the particle is optimal, replacing p with the current positiongdElse pgdKeeping the same;
(c) updating the position x of the particle of the current iterationpd iAnd velocity vpd iJudging whether the iteration times are reached, if so, outputting the optimal parameters of the penalty factor parameter C and the kernel function parameter sigma, establishing an SVR model, and if not, returning to the step (b) for iterative calculation.
3. The method of claim 2, wherein the method is performed by
Determining fitness f of ith iteration of p-th particlep i，p＝1,2，…，N'，ypThe actual value of the aluminum content of the coating of the strip steel is shown,
and the predicted value of the content of the aluminum of the coating output according to the condition attribute data at the pth in the training set is shown.
4. The method of claim 3, wherein the method is performed by
Updating the velocity v of the particle of the current iterationpd iFrom
Updating the position x of the particle of the current iterationpd iWhere i denotes the current number of iterations,
which represents the velocity of the particles to be updated,
indicating the position of the particle to be updated, ppdRepresenting the individual optimum value, p, of the particle to be updatedgdRepresenting the global optimum of the entire population, rand1 and rand2 are [0,1 ]]Random numbers within the range, w is the inertial weight,
i denotes the maximum number of iterations, wmaxRepresenting the maximum value of the inertial weight, wminRepresenting the inertial weight minimum.
5. The method of any of claims 1 to 4, wherein the SVR prediction model is:
wherein, betaqAnd alphaqSolving for Lagrange multiplier by SMO algorithm, wherein sigma is bandwidth of Gaussian kernel function, b is offset, and solving according to KKT optimal condition, and xqIs the condition attribute data in the qth training set, N is the number of training sets, yqAnd (4) obtaining decision attribute data for prediction, namely the aluminum content of the coating output by the target.
6. The method of claim 1, wherein the pre-processing the data of the initial sample set comprises:
and manually screening values of the data with the missing values, manually screening unreasonable data, and normalizing the data.
7. The method of claim 1, wherein the condition attribute data relating to the aluminum content of the strip coating is reduced using a roughness set, and wherein the condition attribute data relating to the aluminum content of the strip coating comprises: zinc liquid composition, zinc liquid temperature, strip steel zinc pot temperature, strip steel running speed, strip steel thickness, upper and lower surface zinc layer measurement thickness, strip steel width and strip steel type data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110266302.3A CN113063916A (en) | 2021-03-11 | 2021-03-11 | Hot-dip galvanized strip steel coating aluminum content prediction method based on PSO-SVR model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110266302.3A CN113063916A (en) | 2021-03-11 | 2021-03-11 | Hot-dip galvanized strip steel coating aluminum content prediction method based on PSO-SVR model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113063916A true CN113063916A (en) | 2021-07-02 |
Family
ID=76560056
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110266302.3A Pending CN113063916A (en) | 2021-03-11 | 2021-03-11 | Hot-dip galvanized strip steel coating aluminum content prediction method based on PSO-SVR model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113063916A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114325693A (en) * | 2021-12-19 | 2022-04-12 | 南京市测绘勘察研究院股份有限公司 | Goaf center deformation prediction method based on InSAR time sequence deformation result |
| TWI800054B (en) * | 2021-10-26 | 2023-04-21 | 中國鋼鐵股份有限公司 | Aluminum content virtual detector of and aluminum content prediction method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101717906A (en) * | 2009-12-23 | 2010-06-02 | 攀钢集团攀枝花钢钒有限公司 | Continuous hot dip galvanizing zinc liquid aluminium content adjusting method |
| CN102778538A (en) * | 2012-07-06 | 2012-11-14 | 广东电网公司电力科学研究院 | Soft measuring method based on improved SVM (Support Vector Machine) for measuring boiler unburned carbon content in fly ash |
| CN104459089A (en) * | 2014-12-12 | 2015-03-25 | 东北大学 | Soft measurement method of freeness of high-consistency pulp grinding system |
| CN104899425A (en) * | 2015-05-07 | 2015-09-09 | 浙江大学 | Variable selection and forecast method of silicon content in molten iron of blast furnace |
| CN107016455A (en) * | 2017-02-27 | 2017-08-04 | 浙江大学 | The forecasting system and method for circulating fluid bed domestic garbage burning boiler furnace outlet flue gas oxygen content |
| CN109596543A (en) * | 2018-11-25 | 2019-04-09 | 西安建筑科技大学 | The spectral reflectance recovery method of particle group optimizing multicore support vector regression |
-
2021
- 2021-03-11 CN CN202110266302.3A patent/CN113063916A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101717906A (en) * | 2009-12-23 | 2010-06-02 | 攀钢集团攀枝花钢钒有限公司 | Continuous hot dip galvanizing zinc liquid aluminium content adjusting method |
| CN102778538A (en) * | 2012-07-06 | 2012-11-14 | 广东电网公司电力科学研究院 | Soft measuring method based on improved SVM (Support Vector Machine) for measuring boiler unburned carbon content in fly ash |
| CN104459089A (en) * | 2014-12-12 | 2015-03-25 | 东北大学 | Soft measurement method of freeness of high-consistency pulp grinding system |
| CN104899425A (en) * | 2015-05-07 | 2015-09-09 | 浙江大学 | Variable selection and forecast method of silicon content in molten iron of blast furnace |
| CN107016455A (en) * | 2017-02-27 | 2017-08-04 | 浙江大学 | The forecasting system and method for circulating fluid bed domestic garbage burning boiler furnace outlet flue gas oxygen content |
| CN109596543A (en) * | 2018-11-25 | 2019-04-09 | 西安建筑科技大学 | The spectral reflectance recovery method of particle group optimizing multicore support vector regression |
Non-Patent Citations (2)
| Title |
|---|
| 冯国昌 等: "镀锌生产中锌合金镀层铁含量的在线检测", 《自动化仪表》 * |
| 唐贤伦等: "基于混沌粒子群支持向量回归的高炉铁水硅含量预测", 《四川大学学报(工程科学版)》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI800054B (en) * | 2021-10-26 | 2023-04-21 | 中國鋼鐵股份有限公司 | Aluminum content virtual detector of and aluminum content prediction method |
| CN114325693A (en) * | 2021-12-19 | 2022-04-12 | 南京市测绘勘察研究院股份有限公司 | Goaf center deformation prediction method based on InSAR time sequence deformation result |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Huang et al. | Air quality prediction using improved PSO-BP neural network | |
| CN113063916A (en) | Hot-dip galvanized strip steel coating aluminum content prediction method based on PSO-SVR model | |
| CN108241901B (en) | Transformer early warning evaluation method and device based on prediction data | |
| CN110378064A (en) | A method of the prediction dead stock column temperature of blast furnace furnace core | |
| WO2023226322A1 (en) | Mechanism- and data driving-based engine cylinder head milled surface quality prediction method | |
| CN110598958B (en) | Ladle hierarchical management analysis method and system | |
| Lee et al. | Spatial and sequential deep learning approach for predicting temperature distribution in a steel-making continuous casting process | |
| Cica et al. | Intelligent process modeling and optimization of porosity formation in high-pressure die casting | |
| CN116689503A (en) | Strip steel full-length thickness prediction method based on memory function network | |
| Tian et al. | Hybrid modeling of molten steel temperature prediction in LF | |
| Chintakindi et al. | Annealing of monel 400 alloy using principal component analysis, hyper-parameter optimization, machine learning techniques, and multi-objective particle swarm optimization | |
| Botnikov et al. | Development of the metal temperature prediction model for steel-pouring and tundish ladles used at the casting and rolling complex | |
| CN110851911B (en) | Terminal state calculation model training method, control sequence searching method and device | |
| CN113239562B (en) | Method for predicting mechanical property of whole plate surface of strip steel | |
| Guo et al. | Prediction of maximum carbon element content in continuous casting billets of 82B cord steel based on statistics of extreme values method | |
| CN108256239B (en) | Prediction method suitable for cold shut defect in casting and mold filling process | |
| Liu et al. | XGBoost-based model for predicting hydrogen content in electroslag remelting | |
| CN116324323A (en) | Method for generating learned prediction model for predicting energy efficiency of melting furnace, method for predicting energy efficiency of melting furnace, and computer program | |
| García Nieto et al. | Chrome layer thickness modelling in a hard chromium plating process using a hybrid PSO/RBF–SVM–based model | |
| Chen et al. | Anode effect prediction based on collaborative two-dimensional forecast model in aluminum electrolysis production. | |
| CN114861555A (en) | Regional comprehensive energy system short-term load prediction method based on Copula theory | |
| Qiang et al. | The combination prediction of BTP in sintering process based on Bayesian framework and LS-SVM | |
| Rannetbauer et al. | Enhancing Predictive Quality in HVOF Coating Technology: A Comparative Analysis of Machine Learning Techniques | |
| Gutiérrez et al. | Statistical study to evaluate the effect of processing variables on shrinkage incidence during solidification of nodular cast irons | |
| Zhang et al. | Demand forecasting of a fused magnesia smelting process based on LSTM and FRA |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210702 |