CN112063782A - Blast furnace burden surface clustering method - Google Patents
Blast furnace burden surface clustering method Download PDFInfo
- Publication number
- CN112063782A CN112063782A CN202010898856.0A CN202010898856A CN112063782A CN 112063782 A CN112063782 A CN 112063782A CN 202010898856 A CN202010898856 A CN 202010898856A CN 112063782 A CN112063782 A CN 112063782A
- Authority
- CN
- China
- Prior art keywords
- blast furnace
- clustering
- furnace burden
- determining
- clustering 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.)
- Withdrawn
Links
Images
Classifications
-
- C—CHEMISTRY; METALLURGY
- C21—METALLURGY OF IRON
- C21B—MANUFACTURE OF IRON OR STEEL
- C21B5/00—Making pig-iron in the blast furnace
-
- C—CHEMISTRY; METALLURGY
- C21—METALLURGY OF IRON
- C21B—MANUFACTURE OF IRON OR STEEL
- C21B7/00—Blast furnaces
- C21B7/24—Test rods or other checking devices
-
- C—CHEMISTRY; METALLURGY
- C21—METALLURGY OF IRON
- C21B—MANUFACTURE OF IRON OR STEEL
- C21B2300/00—Process aspects
- C21B2300/04—Modeling of the process, e.g. for control purposes; CII
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Manufacturing & Machinery (AREA)
- Materials Engineering (AREA)
- Metallurgy (AREA)
- Organic Chemistry (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention provides a blast furnace charge level clustering method, which is used for solving the problem that the existing charge level reconstruction method can only obtain the approximate shape of the charge level, and comprises the steps of obtaining coordinate information of the blast furnace charge level; determining a cluster analysis strategy based on the coordinate information; and outputting clustering parameters according to the clustering analysis strategy. The invention relates to the technical fields of cluster analysis, spectral clustering, blast furnace charge level and the like.
Description
Technical Field
The invention relates to the technical fields of cluster analysis, spectral clustering, blast furnace charge level and the like, in particular to a blast furnace charge level clustering method.
Background
Blast furnace ironmaking production is an important link in the steel industry in China, and the final efficiency of blast furnace ironmaking production directly influences the final yield of the steel industry in China. Blast furnace ironmaking burden operation is an important component of blast furnace ironmaking operation and is also one of important factors directly influencing ironmaking quality. The overall shape distribution of the burden surface is the direct mutual reflection of the burden distribution and the smelting furnace conditions of the blast furnaces, and the reasonable dynamic distribution and the normal coal gas movement of the burden and the coal gas flow of the blast furnaces can be ensured only by fully knowing the overall shape of the surface of the burden distribution of the blast furnaces, so that the blast furnace smelting can be continuously, stably and efficiently carried out, and better economic benefit and technical indexes of the blast furnaces can be obtained.
When a blast furnace operator performs burden distribution adjustment, burden distribution decisions are often obtained by observing several important burden surface characteristics, and an ideal burden surface shape is realized. However, the existing charge level reconstruction method can only obtain the approximate shape of the charge level, cannot represent the specific characteristic information of the discharge level, and is not beneficial to guiding the cloth production.
Disclosure of Invention
The invention provides a blast furnace charge level clustering method.
Provided is a blast furnace burden surface clustering method, which comprises the following steps:
acquiring coordinate information of a blast furnace charge level;
determining a cluster analysis strategy based on the coordinate information;
and outputting clustering parameters according to the clustering analysis strategy.
According to the method for clustering the charge level of the blast furnace, the coordinate information of the charge level of the blast furnace is obtained through a mechanical swing arm radar.
The method for clustering the blast furnace burden surfaces comprises the step of acquiring the coordinate information through a mechanical swing arm radar, wherein the coordinate information comprises 10 point positions on a radial stockline which can be detected by the mechanical radar.
The method for clustering the blast furnace charge level is characterized in that the mechanical swing arm radar is used for imaging the radial charge shape of the blast furnace charge level, and only one position of the furnace top needs to be provided with a hole.
The blast furnace charge level clustering method is characterized in that the clustering analysis strategy is determined by adopting a spectral clustering algorithm based on Nystrom based on the coordinate information.
The blast furnace charge level clustering method comprises the following steps of:
converting the similarity matrix of all the calculated samples into a similarity matrix between sampling samples and a similarity matrix between the sampling samples and the residual samples;
and determining the scale parameters, the clustering number and the number of the characteristic vectors of the Gaussian kernel function in advance.
The blast furnace charge level clustering method comprises the following steps of:
solving Euclidean distance between a data concentration point and a point to determine a scale parameter;
determining the number of clusters by using the evaluation result of the effectiveness index;
and determining the number of the feature vectors by taking the same value according to the cluster number.
The blast furnace charge level clustering method is characterized in that the output clustering parameters comprise calculation time, iteration times and/or sum of squares of errors according to the clustering analysis strategy.
Drawings
Other features, objects, and advantages of the disclosure will become apparent from a reading of the following detailed description of non-limiting embodiments which proceeds with reference to the accompanying drawings.
FIG. 1 is a flow chart of one embodiment of a blast furnace burden level clustering method according to the present disclosure.
Detailed Description
The present disclosure is described in further detail below with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings.
It should be noted that, in the present disclosure, the embodiments and features of the embodiments may be combined with each other without conflict. The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
FIG. 1 illustrates a flow diagram of one embodiment of a blast furnace burden level clustering method according to the present disclosure. The voice adjusting method comprises the following steps:
in step 101, coordinate information of a blast furnace charge level is acquired.
In the step, the coordinate information of the blast furnace burden surface can be obtained through a mechanical swing arm radar. The coordinate information acquired by the mechanical swing arm radar comprises a 10-point position on a radial stockline which can be detected by the mechanical radar. The mechanical swing arm radar is characterized in that the radial material shape of the blast furnace burden surface is imaged, and only one position of the furnace top needs to be provided with a hole.
In some embodiments, the height of the opening of the radar is 16500mm from the lower edge of the furnace top, the included angle between the opening of the radar and the furnace shell is 135 degrees, and the radar is arranged at a position which is 4800mm from the zero stockline. In addition, the inclination angle of the top of the blast furnace is 52 degrees, and the radius of the blast furnace near the zero material line is 4050 mm.
At step 102, based on the coordinate information, a cluster analysis strategy is determined.
In this step, the clustering analysis strategy adopts a spectral clustering algorithm based on Nystrom. Wherein the Nystrom-based spectral clustering algorithm comprises: converting the similarity matrix of all the calculated samples into a similarity matrix between sampling samples and a similarity matrix between the sampling samples and the residual samples; and determining the scale parameters, the clustering number and the number of the characteristic vectors of the Gaussian kernel function in advance. The method for determining the scale parameters, the cluster number and the number of the feature vectors comprises the following steps: solving Euclidean distance between a data concentration point and a point to determine a scale parameter; determining the number of clusters by using the evaluation result of the effectiveness index; and determining the number of the feature vectors by taking the same value according to the cluster number.
In some embodiments, the data visualization technology can be used to obtain the distribution information of the data, without a large number of repeated numerical operations, so as to truly cluster the number. By a data visualization method, the charge level characteristic data is divided into a plurality of clusters according to the fact that the data in each cluster are densely distributed, and whether the boundary between the clusters is clear or not, so that the clustering number of the charge level characteristic data is determined.
In step 103, clustering parameters are output according to the clustering analysis strategy.
In this step, outputting the clustering parameters includes calculating time, iteration times and/or sum of squares of errors according to the clustering analysis strategy.
The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention in the embodiments of the present disclosure is not limited to the specific combination of the above-mentioned features, but also encompasses other embodiments in which any combination of the above-mentioned features or their equivalents is made without departing from the inventive concept as defined above. For example, the above features and (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure are mutually replaced to form the technical solution.
Claims (8)
1. A blast furnace burden surface clustering method, the method comprising:
acquiring coordinate information of a blast furnace charge level;
determining a cluster analysis strategy based on the coordinate information;
and outputting clustering parameters according to the clustering analysis strategy.
2. The blast furnace burden level clustering method according to claim 1, wherein the coordinate information of the blast furnace burden level is obtained by a mechanical swing arm radar.
3. The blast furnace burden level clustering method according to claim 1 or 2, wherein said obtaining said coordinate information by mechanical swing arm radar comprises locating 10 points on a radial stockline detectable by said mechanical radar.
4. The blast furnace burden level clustering method according to claim 2, wherein the mechanical swing arm radar is characterized in that a radial burden shape of a blast furnace burden level is imaged, and only one position of the blast furnace burden level needs to be drilled and installed on the furnace top.
5. The blast furnace burden level clustering method of claim 1, wherein said determining a cluster analysis strategy based on said coordinate information is using a Nystrom-based spectral clustering algorithm.
6. The blast furnace burden level clustering method of claim 1 or 5, wherein said Nystrom-based spectral clustering algorithm comprises:
converting the similarity matrix of all the calculated samples into a similarity matrix between sampling samples and a similarity matrix between the sampling samples and the residual samples;
and determining the scale parameters, the clustering number and the number of the characteristic vectors of the Gaussian kernel function in advance.
7. The blast furnace burden level clustering method of claim 6, wherein the method of determining the scale parameter, the number of clusters and the number of eigenvectors comprises:
solving Euclidean distance between a data concentration point and a point to determine a scale parameter;
determining the number of clusters by using the evaluation result of the effectiveness index;
and determining the number of the feature vectors by taking the same value according to the cluster number.
8. The blast furnace burden level clustering method of claim 1, wherein said outputting clustering parameters according to said cluster analysis strategy comprises calculating time, iteration number and/or sum of squares of error.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010898856.0A CN112063782A (en) | 2020-08-31 | 2020-08-31 | Blast furnace burden surface clustering method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010898856.0A CN112063782A (en) | 2020-08-31 | 2020-08-31 | Blast furnace burden surface clustering method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112063782A true CN112063782A (en) | 2020-12-11 |
Family
ID=73666136
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010898856.0A Withdrawn CN112063782A (en) | 2020-08-31 | 2020-08-31 | Blast furnace burden surface clustering method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112063782A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101492750A (en) * | 2008-12-30 | 2009-07-29 | 北京科技大学 | High furnace burden face measurement and control system based on industrial phased array radar |
CN102732660A (en) * | 2012-06-27 | 2012-10-17 | 浙江大学 | Burden surface temperature field detection method based on multi-source information fusion |
CN102816883A (en) * | 2012-06-18 | 2012-12-12 | 北京科技大学 | Radar, video and laser system combined device for measuring blast furnace burden surface |
CN105002321A (en) * | 2015-06-16 | 2015-10-28 | 内蒙古科技大学 | Method for processing coal gas flow center dynamic tracking and monitoring coal gas utilization rate |
CN105483305A (en) * | 2016-01-12 | 2016-04-13 | 北京科技大学 | Material bed distribution visualization method based on blast furnace radar data |
CN109852748A (en) * | 2019-02-27 | 2019-06-07 | 内蒙古科技大学 | Monitor the Gas Flow development process of blast furnace material distribution period and prediction gas utilization rate method |
-
2020
- 2020-08-31 CN CN202010898856.0A patent/CN112063782A/en not_active Withdrawn
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101492750A (en) * | 2008-12-30 | 2009-07-29 | 北京科技大学 | High furnace burden face measurement and control system based on industrial phased array radar |
CN102816883A (en) * | 2012-06-18 | 2012-12-12 | 北京科技大学 | Radar, video and laser system combined device for measuring blast furnace burden surface |
CN102732660A (en) * | 2012-06-27 | 2012-10-17 | 浙江大学 | Burden surface temperature field detection method based on multi-source information fusion |
CN105002321A (en) * | 2015-06-16 | 2015-10-28 | 内蒙古科技大学 | Method for processing coal gas flow center dynamic tracking and monitoring coal gas utilization rate |
CN105483305A (en) * | 2016-01-12 | 2016-04-13 | 北京科技大学 | Material bed distribution visualization method based on blast furnace radar data |
CN109852748A (en) * | 2019-02-27 | 2019-06-07 | 内蒙古科技大学 | Monitor the Gas Flow development process of blast furnace material distribution period and prediction gas utilization rate method |
Non-Patent Citations (2)
Title |
---|
张海刚 等: "高炉料面特征提取与聚类分析", 《控制理论与应用》 * |
曹铭 等: "高炉料面的分类与案例匹配算法", 《控制理论与应用》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113689428B (en) | Mechanical part stress corrosion detection method and system based on image processing | |
Gola et al. | Objective microstructure classification by support vector machine (SVM) using a combination of morphological parameters and textural features for low carbon steels | |
CN115294120B (en) | Valve surface quality detection method based on image recognition | |
CN111915596A (en) | Method and device for predicting benign and malignant pulmonary nodules | |
CN110906875B (en) | Visual processing method for aperture measurement | |
CN112633337A (en) | Unbalanced data processing method based on clustering and boundary points | |
CN115661147A (en) | Metering detection data identification method based on machine vision | |
CN115294159B (en) | Method for dividing corroded area of metal fastener | |
CN114359288B (en) | Medical image cerebral aneurysm detection and positioning method based on artificial intelligence | |
CN116137036B (en) | Gene detection data intelligent processing system based on machine learning | |
CN116703251B (en) | Rubber ring production quality detection method based on artificial intelligence | |
CN115345894A (en) | Welding seam ray detection image segmentation method | |
CN110457550B (en) | Method for correcting abnormal operation data in sintering process | |
CN117390471A (en) | Hydrogeological data management method and system | |
CN112063782A (en) | Blast furnace burden surface clustering method | |
CN115294527A (en) | Subway tunnel damage detection method based on computer vision | |
CN102346817A (en) | Prediction method for establishing allergen of allergen-family featured peptides by means of SVM (Support Vector Machine) | |
CN116228776B (en) | Electromechanical equipment welding defect identification method and system | |
CN111647698A (en) | Active state detection method and device for blast furnace hearth | |
Xie et al. | A majority affiliation based under-sampling method for class imbalance problem | |
CN111860863A (en) | Active learning method based on covariance matrix | |
CN108229541B (en) | Shore bridge middle pull rod stress data classification method based on K nearest neighbor algorithm | |
CN114861782A (en) | Electronic component near-field scanning electromagnetic pattern clustering analysis method and system | |
CN114972353A (en) | Quality evaluation method and system for metal high-pressure special pipe fitting | |
CN113129314B (en) | Intelligent image processing method for lung shadows |
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 | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20201211 |