CN116562659A - Steel surface slag inclusion defect cause analysis system and method - Google Patents
Steel surface slag inclusion defect cause analysis system and method Download PDFInfo
- Publication number
- CN116562659A CN116562659A CN202210093397.8A CN202210093397A CN116562659A CN 116562659 A CN116562659 A CN 116562659A CN 202210093397 A CN202210093397 A CN 202210093397A CN 116562659 A CN116562659 A CN 116562659A
- Authority
- CN
- China
- Prior art keywords
- slag inclusion
- steel
- data
- inclusion defect
- production
- 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
- 239000002893 slag Substances 0.000 title claims abstract description 246
- 230000007547 defect Effects 0.000 title claims abstract description 214
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 173
- 239000010959 steel Substances 0.000 title claims abstract description 173
- 238000000034 method Methods 0.000 title claims abstract description 117
- 238000004458 analytical method Methods 0.000 title claims abstract description 88
- 238000012544 monitoring process Methods 0.000 claims abstract description 43
- 238000009628 steelmaking Methods 0.000 claims abstract description 26
- 238000005098 hot rolling Methods 0.000 claims abstract description 22
- 238000005516 engineering process Methods 0.000 claims abstract description 12
- 238000004519 manufacturing process Methods 0.000 claims description 90
- 230000008569 process Effects 0.000 claims description 48
- 239000002994 raw material Substances 0.000 claims description 34
- 238000012545 processing Methods 0.000 claims description 30
- 238000005266 casting Methods 0.000 claims description 27
- 238000009749 continuous casting Methods 0.000 claims description 13
- 238000003066 decision tree Methods 0.000 claims description 12
- 238000010438 heat treatment Methods 0.000 claims description 12
- 238000007689 inspection Methods 0.000 claims description 12
- 239000000843 powder Substances 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 11
- 230000004907 flux Effects 0.000 claims description 11
- 238000010801 machine learning Methods 0.000 claims description 10
- XKRFYHLGVUSROY-UHFFFAOYSA-N Argon Chemical compound [Ar] XKRFYHLGVUSROY-UHFFFAOYSA-N 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 8
- 239000007788 liquid Substances 0.000 claims description 8
- 230000000903 blocking effect Effects 0.000 claims description 6
- 238000005097 cold rolling Methods 0.000 claims description 6
- 239000002436 steel type Substances 0.000 claims description 5
- 229910052786 argon Inorganic materials 0.000 claims description 4
- 238000007664 blowing Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000010924 continuous production Methods 0.000 claims description 3
- 238000010219 correlation analysis Methods 0.000 claims description 3
- 238000000556 factor analysis Methods 0.000 claims description 3
- 238000012880 independent component analysis Methods 0.000 claims description 3
- 238000000513 principal component analysis Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000005096 rolling process Methods 0.000 description 5
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 4
- 238000010835 comparative analysis Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 229910052742 iron Inorganic materials 0.000 description 2
- 230000003647 oxidation Effects 0.000 description 2
- 238000007254 oxidation reaction Methods 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- 238000010079 rubber tapping Methods 0.000 description 2
- 238000003723 Smelting Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 230000001012 protector Effects 0.000 description 1
- 238000000275 quality assurance Methods 0.000 description 1
- 239000013049 sediment Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- 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
- G01N33/204—Structure thereof, e.g. crystal structure
- G01N33/2045—Defects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- 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
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Chemical & Material Sciences (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- General Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Manufacturing & Machinery (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Primary Health Care (AREA)
- Evolutionary Computation (AREA)
- Crystallography & Structural Chemistry (AREA)
- Game Theory and Decision Science (AREA)
- Food Science & Technology (AREA)
- Medicinal Chemistry (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating And Analyzing Materials By Characteristic Methods (AREA)
Abstract
The invention discloses a steel surface slag inclusion defect cause analysis system which comprises a data platform module, an online monitoring module and an intelligent analysis module. The invention also discloses a steel surface slag inclusion defect cause analysis method, which respectively utilizes a data platform module, an online monitoring module and an intelligent analysis module to analyze the causes of the surface slag inclusion defects found in the steelmaking process and the hot rolling stage, integrates all data related to the surface slag inclusion defects and analysis technology by the steel surface slag inclusion defect cause analysis system and method, establishes a unified surface slag inclusion defect cause analysis platform, improves the analysis efficiency and accuracy of the steel surface slag inclusion defect causes, integrates a plurality of methods, analyzes the surface slag inclusion defect causes from a plurality of dimensions, reduces the occurrence of the surface slag inclusion defects from various angles, and continuously improves the product quality.
Description
Technical Field
The invention relates to the technical field of steel surface defects, in particular to a system and a method for analyzing the cause of slag inclusion defects on the steel surface.
Background
The surface slag inclusion defect is one of the surface defects frequently occurring in the steel smelting process, and is mainly formed by slag inclusion after casting powder, nozzle sediments and other non-secondary oxidation products enter molten steel and solidified blank shells in the continuous casting process due to abnormal casting powder types, abnormal liquid level fluctuation of a crystallizer, unstable casting pulling speed and the like. The covering slag can cause poor heat conductivity of the solidified blank shell, the solidified shell becomes thin, and steel leakage accidents are easy to occur; the surface slag inclusion which is not stripped in links such as a heating furnace and the like can cause surface defects of subsequent hot rolling and cold rolling procedures, metal oxidation at the position where the slag inclusion defect occurs is serious, and when a finished product is subjected to deep processing operation of stamping, slag inclusion in a thin plate is easy to fall off or crack, so that a great hidden danger can be brought to the safety performance of a component, and serious quality accidents are caused.
With the improvement of safety coefficient and quality requirements of users on steel products, surface slag inclusion of steel has gradually become a key technology for producing high-quality and high-added-value high-grade sheet products, but the defect cause of the surface slag inclusion is influenced by various factors such as steel types, process paths, mold flux types, production stability and the like; besides the surface slag inclusion detected by the continuous casting billet, the surface slag inclusion of the subsurface layer can be found only when the subsequent hot rolling, cold rolling and other procedures are carried out in deep processing, even part of the surface slag inclusion can be found only when a user is required to process, and the density and the serious grade tolerance of the surface slag inclusion defect of different users are different. In addition, the analysis of the cause of the surface slag inclusion defect has the characteristics of complex data sources, various analysis angles and dimensions, and a unified surface slag inclusion defect cause analysis platform needs to be established by integrating all relevant data of the surface slag inclusion defect and common analysis tools so as to improve the analysis efficiency and accuracy of the surface slag inclusion defect cause analysis.
The publication No. CN 102207503A discloses a prediction method before slag inclusion rolling of the surface of a continuous casting billet, an evaluation system is constructed through a crystallizer liquid level value and casting length, and the prediction method before slag inclusion rolling of the surface of the continuous casting billet can be used for predicting the surface quality of the continuous casting billet and guiding the treatment before rolling, but because the prediction method before slag inclusion rolling of the surface of the continuous casting billet lacks analysis based on big data, the threshold value and the weight of the evaluation are relatively solidified, which is unfavorable for continuous optimization.
Disclosure of Invention
The invention aims to provide a system and a method for analyzing the causes of slag inclusion defects on the surface of steel, which are used for gathering and integrating related information of slag inclusion scattered on the surface in a steel enterprise, constructing a one-stop analysis system and a one-stop analysis method for the causes of slag inclusion defects on the surface, which are used for all quality personnel such as quality assurance and quality inspection, by utilizing an industrial data technology, analyzing the causes of the slag inclusion defects on the surface from multiple dimensions by utilizing technologies such as a machine learning algorithm, reducing the occurrence of the slag inclusion defects on the surface from various angles, improving the accuracy and efficiency of analysis, and further continuously improving the quality and qualification rate of products.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
as an aspect of the present invention, there is provided a steel surface slag inclusion defect cause analysis system including:
the data platform module is used for collecting and processing the whole flow data of steel production related to the surface slag inclusion defect; the online monitoring module is connected with the data platform module in a matched manner and is used for carrying out online monitoring on the conditions of finding out the surface slag inclusion defects in online production in the dimensions of working procedures, process paths, steel grades and the like; and the intelligent analysis module is connected with the online monitoring module in a matched manner and is used for analyzing the causes of the steel covering slag from the covering slag raw materials and the production actual score dimension by using a chart and a machine learning algorithm.
As a steel surface slag inclusion defect cause analysis system according to the above aspect of the present invention, wherein the data platform module comprises: the data acquisition part is used for acquiring typical data in the whole flow data of the steel production related to the slag inclusion defect on the surface of the steel; and/or a data processing part connected with the output end of the data acquisition part and used for processing the whole flow data content of steel production related to the surface slag inclusion defect.
The system for analyzing the cause of the slag inclusion defect on the steel surface comprises an online monitoring module, a control module and a control module, wherein the online monitoring module comprises a rule editing part and is used for setting a steel grade, a working procedure or a machine set to be monitored, a production state threshold value and a threshold value of the occurrence ratio of the slag inclusion defect on the steel surface; and/or a monitoring alarm part connected with the output end of the rule editing part, monitoring the online production according to the rule set by the rule editing part, and carrying out online reminding when the threshold value is reached.
As a steel surface slag inclusion defect cause analysis system according to the above aspect of the present invention, wherein the intelligent analysis module includes a casting powder raw material dimension analysis section for analyzing the influence of selection of different steel casting powder raw materials on the surface slag inclusion defect; and/or a production actual performance dimension intelligent analysis part, analyzing key influencing factors of surface slag inclusion defects from data such as technological parameter actual performance values, set values, characteristic values and the like of procedures such as steelmaking continuous casting, converter, hot rolling heating furnace and the like.
As another aspect of the present invention, there is provided a steel surface slag inclusion defect cause analysis method for the steel surface slag inclusion defect cause analysis system, comprising the steps of:
s1, acquiring and processing full-flow data of steel production related to surface slag inclusion defects by utilizing a data platform module;
s2, carrying out multi-dimensional online monitoring on the condition of finding out the surface slag inclusion defect in online production by utilizing an online analysis module;
s3, intelligent analysis is carried out on the covering slag cause by using an intelligent analysis module from the covering slag raw material and the production performance dimension through a chart and a machine learning algorithm.
As a method for analyzing the cause of slag inclusion defect on the surface of steel according to the above aspect of the present invention, S1 comprises the steps of:
s11, collecting typical data in the whole process data of the steel production related to slag inclusion defects on the surface of the steel through a data collecting part;
s12, processing full-flow data of steel production related to slag inclusion defects on the surface of steel through a data processing part.
As a steel surface slag inclusion defect cause analysis method of the above aspect of the present invention, wherein typical data in the steel production full-flow data in S11 includes steel coil full-flow production basic information, namely steel coil full-flow corresponding production units from steelmaking to finished product output, production status, steel coil in/out coil number, steel coil in/out weight, width and thickness specification information, production start/end time; and/or
Technological parameter data, namely set value data of crystallizer liquid level actual performance values, tundish molten steel temperature, casting speed, argon blowing flow production actual performance data and related data in steelmaking, heating furnaces and the like are collected; and/or mold flux raw material information, namely, supplier, model and batch information of mold flux used by each heat is collected; and/or
And collecting surface slag inclusion defect information, namely collecting surface slag inclusion defect data found by each process, wherein the surface slag inclusion defect information comprises steel making, hot rolling and cold rolling process meter inspection data, manual inspection data, process blocking data and user feedback quality objection data.
As a method for analyzing the cause of slag inclusion defect on the surface of steel according to the above aspect of the present invention, S12 comprises the steps of:
s121, preprocessing data, namely processing repeated data and missing values in the steel production process;
s122, calculating a characteristic value, namely extracting key technological parameters related to the surface slag inclusion defect;
s123, the whole process data are connected in series, namely typical data in the whole process data of steel production are associated with coil numbers and unit numbers according to the production sequence of steel coils, and the whole process data of steel production are connected in series according to the logic that the outlet coil number of a former unit is the inlet coil number of a latter unit.
As a method for analyzing the cause of slag inclusion defect on the surface of steel according to the above aspect of the present invention, S121 specifically includes the steps of:
s1211, repeating data processing, namely eliminating repeated data according to volume numbers, time or sampling points;
s1212, processing the missing values, namely filling the missing data by using the interpolation data.
The method for analyzing the cause of slag inclusion defect on the surface of steel, which is the aspect of the invention, comprises the following steps of:
s21, setting appointed monitored steel types, working procedures or units and data sources through a rule editing part;
s22, monitoring the online production of steel through a monitoring alarm part, and carrying out online reminding through a plurality of monitoring alarm modes when the steel reaches a threshold value.
The method for analyzing the cause of the slag inclusion defect on the surface of the steel comprises the following steps of S21, wherein the data sources comprise meter detection data, manual detection data, process blocking data, user feedback quality objection data or a combination of the data; and/or
The various monitoring alarm module modes in S22 include, but are not limited to, page graphics, audible alarm and mail alarm.
The method for analyzing the cause of slag inclusion defect on the surface of steel, which is the aspect of the invention, comprises the following steps of:
s31, analyzing the influence of the selection of different casting powder raw materials on the surface slag inclusion defect through an analysis chart technology;
s32, analyzing key influencing factors of slag inclusion defects on the steel surface from technological parameter actual performance values, set values and characteristic value data of the steelmaking process through a machine learning algorithm technology.
As a method for analyzing the cause of slag inclusion defect on the surface of steel according to the above aspect of the present invention, S31 comprises the steps of:
s311, analyzing the occurrence condition of slag inclusion defects on the corresponding surface of a specified working procedure or unit, a specified time period and a specified steel grade through an analysis chart, and obtaining the occurrence concentration condition of slag inclusion defects on the surface of steel;
s312, comparing and analyzing the occurrence rate of the slag inclusion defect on the surface of the casting powder raw material corresponding to the concentrated working procedure or unit, time period or steel grade through an analysis chart; and/or
Wherein S32 comprises the steps of:
s321, selecting an analysis range and process parameters, namely selecting a unit to be analyzed, steel grade and batch information, acquiring a data range to be analyzed, and selecting process parameters related to surface slag inclusion defects;
s322, selecting the source of surface slag inclusion data, namely selecting the procedures and the quantity sources for discovering slag inclusion defects on the surface of steel to be analyzed;
s323, influence factor analysis, namely preliminarily determining key influence factors of the surface slag inclusion defect by utilizing principal component analysis, independent component analysis and correlation analysis;
s324, a surface slag inclusion defect prediction model is built by using a decision tree algorithm by utilizing selected historical data, namely, input and output of the selected model, and key influencing factors of slag inclusion defects estimated by the model are acquired.
As a method for analyzing the cause of slag inclusion defect on the surface of steel according to the above aspect of the present invention, S324 includes the following steps:
s3241, respectively selecting input and output of a surface slag inclusion defect model;
s3242, establishing a steel surface slag inclusion defect prediction model.
As a method for analyzing the cause of slag inclusion defect on the surface of steel according to the above aspect of the present invention, S3241 specifically comprises the steps of:
s32411, inputting continuous process parameters of the procedures of steelmaking continuous casting, converter, hot rolling heating furnace and the like corresponding to the slab number by the data platform module by the model;
s32412, outputting a model to determine whether a slab has surface slag inclusion defect, namely, detecting that the slab and any corresponding sub-roll have the surface slag inclusion defect in a specified working procedure and a detection mode, and marking the surface slag inclusion defect as 1, otherwise marking the surface slag inclusion defect as 0; and/or
Wherein S3242 specifically comprises the steps of:
s32421 evaluating the influence factors of technological parameters, namely selecting optimal dividing features based on Yu Jini index, and recording the selected data set D, K as classification of slag inclusion defect or not, P k For the probability of occurrence of class k, the coefficient of kunity is as follows:
for the process parameter A, a certain possible value a is selected, D is divided into two parts of D1 and D2, then under the conditions of process parameter A, the coefficient of the Kernel of sample D is defined as
S3242, establishing a steel surface slag inclusion defect prediction model, namely selecting historical data, and generating the surface slag inclusion defect prediction model by using a decision tree algorithm;
s32423 rule extraction, namely obtaining optimal division of each node in the model layer by layer to form an explicit rule.
As a method for analyzing the cause of slag inclusion defect on the surface of steel according to the above aspect of the present invention, S3242 specifically comprises the steps of:
s32421, setting a threshold, namely, when the minimum value of the corresponding sample of the node and the minimum value of the coefficient of the foundation are lower than the threshold, the model is not divided any more;
s32422, if the number of the current node samples is smaller than a threshold value or no characteristic exists, returning to a decision sub-tree, and stopping recursion by the node;
s32423, calculating a base coefficient of the sample set, and if the base coefficient is smaller than a threshold value, returning to a decision tree subtree, wherein the node stops recursion;
s32424 calculating the base index of each dividing value of each technological parameter of the current node;
s32425, selecting a technological parameter with the minimum coefficient of the radix and a dividing point thereof as the optimal characteristic and the optimal dividing point of the current node, and generating a left sub node and a right sub node of the current node;
s32426, the steps are respectively recursively carried out on the left child node and the right child node, and finally a surface slag inclusion defect prediction model is generated.
By adopting the technical scheme, the invention has the following advantages:
the invention provides a steel surface slag inclusion defect cause analysis system and a method, wherein the surface slag inclusion defect cause analysis system comprises a data platform module, an online monitoring module and an intelligent analysis module, and a unified surface slag inclusion defect cause analysis platform is built by integrating all relevant data of the surface slag inclusion defect and a common analysis tool so as to improve the analysis efficiency and accuracy of the surface slag inclusion defect cause analysis system; the related data of the surface slag inclusion defects are subjected to series connection, pretreatment and the like, so that the quality of the data is improved, the requirements of analysis of the surface slag inclusion defects of all quality business personnel can be supported, and the accuracy of the analysis is improved; the online production condition of personalized multidimensional monitoring is supported, the surface slag inclusion defect can be found in time conveniently, further, the treatment is carried out in time, batch defects are reduced, and the product qualification rate is improved.
Drawings
Other features, objects and advantages of the present invention will become more apparent upon reading of the detailed description of non-limiting embodiments, made with reference to the following drawings.
FIG. 1 is a block diagram of a system for analyzing the cause of slag inclusion defects on the surface of steel in accordance with the present invention;
FIG. 2 is a flow chart of the analysis method of the cause of slag inclusion defect on the surface of steel according to the invention;
FIG. 3 is a graph showing comparative analysis of raw materials of a protective slag brand in an embodiment of the present invention;
FIG. 4 is a diagram showing comparative analysis of the corresponding batch of the raw material of the slag protector in the embodiment of the present invention;
FIG. 5 is an analysis result of a slag inclusion defect prediction model established based on a decision tree in an embodiment of the present invention.
Detailed Description
The detailed features and advantages of the present invention will be readily apparent to those skilled in the art from the following detailed description, claims, and drawings that follow, taken in conjunction with the accompanying drawings.
FIG. 1 shows a frame diagram of a system for analyzing the cause of slag inclusion defects on the surface of steel according to the present invention. The steel surface slag inclusion defect cause analysis system is shown in fig. 1, and specifically comprises:
the data platform module is used for collecting and processing full-flow data of steel production related to slag inclusion defects on the surface of steel; and/or an online monitoring module which is connected with the data platform module in a matching way and is used for carrying out online monitoring on the conditions of finding out the surface slag inclusion defects in online production in the dimensions of working procedures, process paths, steel grades and the like; and/or the intelligent analysis module is connected with the online monitoring module in a matched manner and is used for analyzing the steel mold flux cause from the mold flux raw material and the production actual performance dimension by using a chart and a machine learning algorithm.
Wherein: the data platform module comprises a data acquisition part for acquiring typical data in the whole flow data of the steel production related to slag inclusion defects on the surface of the steel; and/or the data processing part is connected with the output end of the data acquisition part and is used for processing the whole flow data content of steel production related to the surface slag inclusion defect.
The online monitoring module comprises a rule editing part and is used for setting a steel grade, a working procedure or a unit to be monitored, a production state threshold value and a threshold value of the occurrence ratio of slag inclusion defects on the surface of steel; and/or a monitoring alarm part connected with the output end of the rule editing part, monitoring the online production according to the rule set by the rule editing part, and carrying out online reminding when the threshold value is reached.
The intelligent analysis module comprises a casting powder raw material dimension analysis part for analyzing the influence of selection of different steel casting powder raw materials on the surface slag inclusion defect; and/or a production actual performance dimension intelligent analysis part, analyzing key influencing factors of surface slag inclusion defects from data such as technological parameter actual performance values, set values, characteristic values and the like of procedures such as steelmaking continuous casting, converter, hot rolling heating furnace and the like.
FIG. 2 shows a flow chart of the analysis method of the cause of slag inclusion defect on the steel surface of the invention; the analysis method of the cause of slag inclusion defect on the surface of steel is shown in fig. 2, and specifically comprises the following steps:
s1, acquiring and processing full-flow data of steel production related to surface slag inclusion defects by utilizing a data platform module;
wherein, S1 comprises the following steps:
s11, collecting typical data in the whole process data of the steel production related to slag inclusion defects on the surface of the steel through a data collecting part;
typical data in the whole flow data of steel production related to the surface slag inclusion defect are collected as follows:
basic information of the whole flow production of steel coil: the whole flow of the steel coil from steelmaking to finished product output corresponds to a production unit, the production state, the coil number of the steel coil inlet/outlet, the weight of the steel coil inlet/outlet, the width and thickness specification information and the production start/end time;
process parameter data: mainly collecting production performance data such as crystallizer liquid surface performance values, tundish molten steel temperature, casting speed, argon blowing flow and the like in steelmaking, heating furnaces and the like and set value data of related data;
mold flux raw material information: collecting information such as suppliers, models, batches and the like of the mold flux used by each heat;
surface slag inclusion defect information: and collecting surface slag inclusion defect data found in each process, including steel making, hot rolling and cold rolling process meter inspection data, manual inspection data, process blockage data and user feedback quality objection data.
S12, processing full-flow data of steel production related to slag inclusion defects on the surface of steel through a data processing part.
Wherein: s12 comprises the following steps:
s121, preprocessing data, namely processing repeated data and missing values in the steel production process;
wherein, S121 specifically includes the following steps:
s1211, repeating data processing, namely eliminating repeated data according to volume numbers, time or sampling points;
s1212, processing the missing values, namely filling the missing data by using the interpolation data.
S122, calculating characteristic values, namely extracting key technological parameters related to surface slag inclusion defects, such as maximum values, average values, median values, peak values and the like of time sequence data of a crystallizer liquid level actual result value, a tundish molten steel temperature and the like;
s123, the whole process data are connected in series, namely, according to the production sequence of the steel coil, the basic information, the technological parameter data, the covering slag raw material information and the surface slag inclusion defect information of the whole process of the steel coil are associated with the coil number and the unit number, and according to the logic that the outlet coil number of the former unit is the inlet coil number of the latter unit, the whole process data are connected in series.
S2, carrying out multi-dimensional online monitoring on the condition of finding out the surface slag inclusion defect in online production by utilizing an online analysis module;
wherein, S2 specifically includes the following steps:
s21, setting appointed monitored steel types, working procedures or units and data sources through a rule editing part;
in addition, the data source refers to meter detection data, manual inspection data, process blockage data, user feedback quality objection data or a combination of the data;
s22, monitoring the online production of steel through a monitoring alarm part, and carrying out online reminding through a plurality of monitoring alarm modes when the steel reaches a threshold value.
In addition, the monitoring alarm module mode includes but is not limited to page graphics, sound alarm, mail alarm and the like.
S3, intelligent analysis is carried out on the covering slag cause by using an intelligent analysis module from the covering slag raw material and the production performance dimension through a chart and a machine learning algorithm.
Wherein, S3 specifically includes the following steps:
s31, analyzing the influence of the selection of different casting powder raw materials on the surface slag inclusion defect through an analysis chart technology; the analysis chart technology comprises, but is not limited to, technologies such as trend contrast chart, histogram, box chart and pivot table.
In addition, S31 specifically includes the following steps:
s311, analyzing the occurrence condition of slag inclusion defects on the corresponding surface of a specified working procedure or unit, a specified time period and a specified steel grade through an analysis chart, and obtaining the occurrence concentration condition of slag inclusion defects on the surface of steel;
s312, comparing and analyzing the occurrence rate of the slag inclusion defect on the surface of the casting powder raw material corresponding to the concentrated working procedure or unit, time period or steel grade through the analysis chart.
S32, analyzing key influencing factors of the surface slag inclusion defects from data such as technological parameter actual values, set values, characteristic values and the like of the procedures such as steelmaking continuous casting, converter, hot rolling heating furnace and the like through a machine learning algorithm technology.
Wherein, S32 comprises the following specific steps:
s321, selecting an analysis range and process parameters, namely selecting a unit to be analyzed, steel grade and batch information, acquiring a data range to be analyzed, and selecting process parameters related to surface slag inclusion defects;
s322, selecting the source of surface slag inclusion data, namely selecting the procedures and the quantity sources for discovering slag inclusion defects on the surface of steel to be analyzed;
s323, influence factor analysis, namely preliminarily determining key influence factors of the surface slag inclusion defect by utilizing principal component analysis, independent component analysis and correlation analysis;
s324, a surface slag inclusion defect prediction model is built by using a decision tree algorithm by utilizing selected historical data, namely, input and output of the selected model, and key influencing factors of slag inclusion defects estimated by the model are acquired.
Wherein S324 comprises the following specific steps:
s3241, respectively selecting input and output of a surface slag inclusion defect model;
s32411, inputting continuous process parameters of the procedures of steelmaking continuous casting, converter, hot rolling heating furnace and the like corresponding to the slab number by the data platform module by the model;
s32412, outputting the model to determine whether the slab has surface slag inclusion defect, namely, the slab and any sub-roll corresponding to the slab are detected to have the surface slag inclusion defect in a specified working procedure and a detection mode, and marking the surface slag inclusion defect as 1, otherwise marking the surface slag inclusion defect as 0.
S3242, establishing a steel surface slag inclusion defect prediction model.
Wherein: s3242 specifically comprises the following steps:
s32421, estimating technological parameter influence factors, namely selecting optimal dividing characteristics based on Yu Jini indexes, wherein the coefficient of the base is as follows:
wherein, the selected data set D, K is whether slag inclusion defect classification occurs or not, P k For the probability of occurrence of the kth class,
for the process parameter A, a certain possible value a is selected, D is divided into two parts of D1 and D2, then under the conditions of process parameter A, the coefficient of the Kernel of sample D is defined as
S3242, establishing a steel surface slag inclusion defect prediction model, namely selecting historical data, and generating the surface slag inclusion defect prediction model by using a decision tree algorithm;
wherein: s3242 specifically comprises the following steps:
s32421, setting a threshold, namely, when the minimum value of the corresponding sample of the node and the minimum value of the coefficient of the foundation are lower than the threshold, the model is not divided any more;
s32422, if the number of the current node samples is smaller than a threshold value or no characteristic exists, returning to a decision sub-tree, and stopping recursion by the node;
s32423, calculating a base coefficient of the sample set, and if the base coefficient is smaller than a threshold value, returning to a decision tree subtree, wherein the node stops recursion;
s32424 calculating the base index of each dividing value of each technological parameter of the current node;
s32425, selecting a technological parameter with the minimum coefficient of the radix and a dividing point thereof as the optimal characteristic and the optimal dividing point of the current node, and generating a left sub node and a right sub node of the current node;
s32426, the steps are respectively recursively carried out on the left child node and the right child node, and finally a surface slag inclusion defect prediction model is generated.
S32423 rule extraction, namely obtaining optimal division of each node in the model layer by layer to form an explicit rule.
Example 1
The embodiment provides a method and a system for analyzing the cause of the surface slag inclusion defect, which mainly aim at analyzing the cause of the surface slag inclusion defect discovered in the steelmaking process and the hot rolling stage. The system for analyzing the cause of the slag inclusion defect on the surface of the embodiment comprises a data platform module, an online monitoring module and an intelligent analysis module.
1) Data platform module
The data platform module is used for collecting and processing the whole flow data of steel production related to the surface slag inclusion defect;
wherein, typical data in the whole flow data of steel production related to the surface slag inclusion defect are collected as follows:
a) Basic information of the whole flow production of the steel coil: the whole flow of the steel coil from steelmaking, hot rolling to finished product production corresponds to a production unit, and whether the production state of hot rolling is finished, the coil number of the steel coil inlet/outlet, the weight of the steel coil inlet/outlet, the width and thickness specification information and the production start/end time are all obtained;
b) Process parameter data: collecting production performance data such as crystallizer liquid surface performance values, tundish molten steel temperature, casting speed, tundish molten steel weight average value, argon blowing flow and the like in steelmaking, heating furnaces and the like and set value data of related data;
c) Mold flux raw material information: collecting information such as suppliers, models, batches and the like of the mold flux used by each heat;
d) Surface slag inclusion defect information: collecting detection data of a steel-making and hot-rolling surface inspection instrument, manual inspection data and blocking data of hot-rolling covering slag;
wherein, the whole flow data of the steel production related to the surface slag inclusion defect is as follows:
data preprocessing: deleting the data such as null value, repeated value and the like in the data;
calculating a characteristic value: extracting characteristic values such as maximum value, average value, median, peak value and the like of time sequence data such as a crystallizer liquid level performance value, tundish molten steel temperature and the like;
serial connection of production full-flow data: and (3) associating the steel coil whole-process production basic information, the process parameter data, the covering slag raw material information and the surface slag inclusion defect information with a coil number and a unit number according to the steel coil production sequence, and connecting the steel production whole-process data in series according to the logic that the outlet coil number of the former unit is the inlet coil number of the latter unit.
2) On-line monitoring module
The module is used for carrying out online monitoring on dimensions such as a process path, steel grade and the like under the condition of finding out surface slag inclusion defects in online production, and comprises two parts of rule editing and monitoring and alarming.
The editing rule is that if the rolling proportion of the steel grade 1 in a hot rolling unit exceeds 80% and the surface slag inclusion defect finding proportion exceeds 5%, alarming is carried out, wherein the source of the data for finding the surface slag inclusion defect is designated as steelmaking, hot rolling manual detection data and blocking data;
the monitoring alarm part monitors the online production condition, and the mail pushing alarm is carried out when the rule set threshold value is reached.
3) Intelligent analysis module
In the embodiment, the intelligent analysis module is utilized to analyze the formation reason of the surface slag inclusion defect from the dimension of the casting slag raw material;
a) In the example, 283 groups of data are obtained, and the occurrence rate of surface slag inclusion defects of a plate blank with a tapping mark DV8000A1 in 2021 month-month is found to be 7.43%, the defect occurrence rate is high, and a DV8210A 1-covering slag brand raw material comparison analysis chart is shown in FIG. 3; the corresponding rolls are produced by the same casting machine, and irregular fluctuation is carried out according to the defect occurrence rate of the day statistics from 16.2% to 3.32%;
b) The raw materials of the covering slag used for analyzing the batch of slabs are marked as B-323, B-911, B-941 and B-971, the occurrence rate of the corresponding slag inclusion defect is respectively 0.00%, 7.63%, 7.57% and 5.81%, as shown in the specific figure 3, the comprehensive yield (yield = blocked slab yield + normal slab yield) and the occurrence rate of the slag inclusion defect (slag inclusion defect occurrence rate = slag inclusion defect slab yield/yield), and the raw material covering slag marked as B-941 should be paid attention to;
c) Fig. 4 shows a comparative analysis of a corresponding batch of a raw material grade B-941 of a raw material of steel mark DV8000 A1-covering slag, and further analysis of a corresponding batch information of a raw material of grade B-941, specifically, as shown in fig. 4, wherein the occurrence rate of obvious slag inclusion defects of a corresponding batch 78 and a batch 80 of the raw material grade B-941 is high, and the two batches are determined to be key influencing factors of the surface slag inclusion defects.
Example 2
The embodiment provides a method and a system for analyzing the cause of the surface slag inclusion defect, which mainly aims at analyzing the cause of the surface slag inclusion defect discovered in the steelmaking process and the hot rolling stage from the dimension of the production actual results.
1) The data platform module and the online monitoring module of this embodiment are the same as those of embodiment 1.
2) Intelligent analysis module
In the embodiment, the intelligent analysis module is utilized to analyze the formation reason of the slag inclusion defect on the surface from the dimension of the production performance
a) The occurrence rate of slab slag inclusion defects of which the tapping mark of a casting machine is DV8000B2 in 2021 is higher;
b) Selecting process parameter information which possibly affects the covering slag: drawing speed, heat extraction, casting temperature, molten iron temperature, ladle target temperature and slab surface temperature are taken as model input; for each slab, if the slab or the corresponding sub-coil is manually checked for surface slag inclusion defects in steelmaking, hot rolling or cold rolling or blocked due to the surface slag inclusion defects in the corresponding working procedure, the slab is marked as having the surface slag inclusion defects;
c) Based on a business statistics rule, selecting a sample point of 10 (namely, the number of slabs is 10), and not dividing when the coefficient of the foundation is 0.01;
d) The coefficient of the foundation of the dividing point corresponding to the drawing speed, the drawing heat, the casting temperature, the molten iron temperature, the ladle target temperature and the slab surface temperature is calculated, and the obtained data are as follows: the draw speed is the smallest when 99.1 dividing points (the coefficient of the foundation is 0.388), the heat extraction quantity is the smallest when 13.5 dividing points (the coefficient of the foundation is 0.18), the casting temperature is the smallest when 1564.22 dividing points (the coefficient of the foundation is 0.34), the ladle target temperature is the smallest when 1551.1 dividing points (the coefficient of the foundation is 0.48), and the slab surface temperature is the smallest when 1557.7 dividing points (the coefficient of the foundation is 0.52). Continuously dividing left and right child nodes of the division points which are obtained according to the minimum standard of the coefficient of the base as root nodes;
e) Based on d) finally confirming that the pulling speed, the pulling heat and the casting temperature are key influencing factors;
f) FIG. 5 shows the analysis results of building slag inclusion defect prediction models based on decision trees in an embodiment of the invention. Based on the slag inclusion defect prediction model established by the decision tree, the extraction rule is specifically shown in fig. 5, and the following conclusion can be obtained:
when 10< heat extraction quantity < = 13.5, 43.5< heat extraction quantity < = 47, and pulling speed < = 108, 62.5< heat extraction quantity and 1554< casting temperature, the slab is a normal slab, and blocking is not needed;
when 13.5< heat extraction quantity < =43.5, the pulling speed < =108, 62.5< heat extraction quantity and the casting temperature < =1554, the occurrence rate of slag inclusion defect is higher, and important attention is needed.
Finally, it is pointed out that while the invention has been described with reference to a specific embodiment thereof, it will be understood by those skilled in the art that the above embodiments are provided for illustration only and not as a definition of the limits of the invention, and various equivalent changes or substitutions may be made without departing from the spirit of the invention, therefore, all changes and modifications to the above embodiments shall fall within the scope of the appended claims.
Claims (16)
1. A steel surface slag inclusion defect cause analysis system, comprising:
the data platform module is used for collecting and processing the whole flow data of steel production related to the surface slag inclusion defect;
the online monitoring module is connected with the data platform module in a matching way and is used for carrying out online monitoring on the conditions of finding out the surface slag inclusion defect in online production in multiple dimensions of working procedures, process paths and steel types;
and the intelligent analysis module is connected with the online monitoring module in a matched manner and is used for analyzing the causes of the steel covering slag from the covering slag raw materials and the production actual score dimension by using a chart and a machine learning algorithm.
2. The steel surface slag inclusion defect cause analysis system of claim 1, wherein the data platform module comprises:
the data acquisition part is used for acquiring typical data in the whole flow data of the steel production related to the slag inclusion defect on the surface of the steel; and/or
And the data processing part is connected with the output end of the data acquisition part and is used for processing the whole flow data content of steel production related to the surface slag inclusion defect.
3. The steel surface slag inclusion defect cause analysis system of claim 1, wherein the online monitoring module comprises:
the rule editing part is used for setting a steel grade, a working procedure or a machine set to be monitored, a production state threshold value and a threshold value of the occurrence ratio of slag inclusion defects on the surface of steel; and/or
And the monitoring alarm part is connected with the output end of the rule editing part, monitors on-line production according to the rule set by the rule editing part, and carries out on-line reminding when the on-line production reaches a threshold value.
4. The steel surface slag inclusion defect cause analysis system of claim 1, wherein the intelligent analysis module comprises:
a covering slag raw material dimension analysis part for analyzing the influence of the selection of different steel covering slag raw materials on the surface slag inclusion defect; and/or
And the intelligent production performance dimension analysis part is used for analyzing key influencing factors of surface slag inclusion defects from data such as technological parameter performance values, set values, characteristic values and the like of procedures such as steelmaking continuous casting, converter, hot rolling heating furnace and the like.
5. A steel surface slag inclusion defect cause analysis method for a steel surface slag inclusion defect cause analysis system according to any one of claims 1 to 5, comprising the steps of:
s1, acquiring and processing full-flow data of steel production related to surface slag inclusion defects by utilizing a data platform module;
s2, carrying out multi-dimensional online monitoring on the condition of finding out the surface slag inclusion defect in online production by utilizing an online analysis module;
s3, intelligent analysis is carried out on the covering slag cause by using an intelligent analysis module from the covering slag raw material and the production performance dimension through a chart and a machine learning algorithm.
6. The method for analyzing the cause of slag inclusion defect on the surface of steel as set forth in claim 5, wherein said step S1 comprises the steps of:
s11, collecting typical data in the whole process data of the steel production related to slag inclusion defects on the surface of the steel through a data collecting part;
s12, processing full-flow data of steel production related to slag inclusion defects on the surface of steel through a data processing part.
7. The method for analyzing the cause of slag inclusion defects on the surface of steel as claimed in claim 6, wherein typical data in the steel production full-process data in S11 includes basic information of steel coil full-process production, namely steel coil full-process corresponding production units from steelmaking to finished product output, production state, steel coil in/out coil number, steel coil in/out weight, width and thickness specification information, production start/end time; and/or
Technological parameter data, namely set value data of crystallizer liquid level actual performance values, tundish molten steel temperature, casting speed, argon blowing flow production actual performance data and related data in steelmaking, heating furnaces and the like are collected; and/or
The mold flux raw material information, namely, the supplier, model and batch information of mold flux used by each heat is collected; and/or
And collecting surface slag inclusion defect information, namely collecting surface slag inclusion defect data found by each process, wherein the surface slag inclusion defect information comprises steel making, hot rolling and cold rolling process meter inspection data, manual inspection data, process blocking data and user feedback quality objection data.
8. The method for analyzing the cause of slag inclusion defect on the surface of steel as set forth in claim 6, wherein said step S12 comprises the steps of:
s121, preprocessing data, namely processing repeated data and missing values in the steel production process;
s122, calculating a characteristic value, namely extracting key technological parameters related to the surface slag inclusion defect;
s123, the whole process data are connected in series, namely typical data in the whole process data of steel production are associated with coil numbers and unit numbers according to the production sequence of steel coils, and the whole process data of steel production are connected in series according to the logic that the outlet coil number of a former unit is the inlet coil number of a latter unit.
9. The method for analyzing the cause of slag inclusion defect on the surface of steel as set forth in claim 8, wherein said step S121 comprises the steps of:
s1211, repeating data processing, namely eliminating repeated data according to volume numbers, time or sampling points;
s1212, processing the missing values, namely filling the missing data by using the interpolation data.
10. The method for analyzing the cause of slag inclusion defect on the surface of steel as set forth in claim 5, wherein said S2 comprises the steps of:
s21, setting appointed monitored steel types, working procedures or units and data sources through a rule editing part;
s22, monitoring the online production of steel through a monitoring alarm part, and carrying out online reminding through a plurality of monitoring alarm modes when the steel reaches a threshold value.
11. The method for analyzing causes of slag inclusion defects on a steel surface according to claim 10, wherein the data sources in S21 include inspection data of a meter, manual inspection data, process blockage data, user feedback quality objection data, or a combination thereof; and/or
The various monitoring alarm module modes in S22 include, but are not limited to, page graphics, audible alarm, and mail alarm.
12. The method for analyzing the cause of slag inclusion defect on the surface of steel as set forth in claim 5, wherein said step S3 comprises the steps of:
s31, analyzing the influence of the selection of different casting powder raw materials on the surface slag inclusion defect through an analysis chart technology;
s32, analyzing key influencing factors of slag inclusion defects on the steel surface from technological parameter actual performance values, set values and characteristic value data of the steelmaking process through a machine learning algorithm technology.
13. The method for analyzing the cause of slag inclusion defect on the surface of steel as set forth in claim 12, wherein said S31 comprises the steps of:
s311, analyzing the occurrence condition of slag inclusion defects on the corresponding surface of a specified working procedure or unit, a specified time period and a specified steel grade through an analysis chart, and obtaining the occurrence concentration condition of slag inclusion defects on the surface of steel;
s312, comparing and analyzing the occurrence rate of the slag inclusion defect on the surface of the casting powder raw material corresponding to the concentrated working procedure or unit, time period or steel grade through an analysis chart; and/or
The step S32 includes the steps of:
s321, selecting an analysis range and process parameters, namely selecting a unit to be analyzed, steel grade and batch information, acquiring a data range to be analyzed, and selecting process parameters related to surface slag inclusion defects;
s322, selecting the source of surface slag inclusion data, namely selecting the procedures and the quantity sources for discovering slag inclusion defects on the surface of steel to be analyzed;
s323, influence factor analysis, namely preliminarily determining key influence factors of the surface slag inclusion defect by utilizing principal component analysis, independent component analysis and correlation analysis;
s324, a surface slag inclusion defect prediction model is built by using a decision tree algorithm by utilizing selected historical data, namely, input and output of the selected model, and key influencing factors of slag inclusion defects estimated by the model are acquired.
14. The method for analyzing the cause of slag inclusion defect on the surface of steel as set forth in claim 13, wherein said step S324 comprises the steps of:
s3241, respectively selecting input and output of a surface slag inclusion defect model;
s3242, establishing a steel surface slag inclusion defect prediction model.
15. The method for analyzing the cause of slag inclusion defect on the surface of steel as set forth in claim 14, wherein said step S3241 comprises the steps of:
s32411, inputting continuous process parameters of the procedures of steelmaking continuous casting, converter, hot rolling heating furnace and the like corresponding to the slab number by the data platform module by the model;
s32412, outputting a model to determine whether a slab has surface slag inclusion defect, namely, detecting that the slab and any corresponding sub-roll have the surface slag inclusion defect in a specified working procedure and a detection mode, and marking the surface slag inclusion defect as 1, otherwise marking the surface slag inclusion defect as 0; and/or
The step S3242 specifically includes the following steps:
s32421 evaluating the influence factors of technological parameters, namely selecting optimal dividing features based on Yu Jini index, and recording the selected data set D, K as classification of slag inclusion defect or not, P k For the probability of occurrence of class k, the coefficient of kunity is as follows:
for the process parameter A, a certain possible value a is selected, D is divided into two parts of D1 and D2, then under the conditions of process parameter A, the coefficient of the Kernel of sample D is defined as
S32422, establishing a steel surface slag inclusion defect prediction model, namely selecting historical data, and generating the surface slag inclusion defect prediction model by using a decision tree algorithm;
s32423 rule extraction, namely obtaining optimal division of each node in the model layer by layer to form an explicit rule.
16. The method for analyzing the cause of slag inclusion defect on the surface of steel as set forth in claim 15, wherein said step S3242 comprises the steps of:
s32421, setting a threshold, namely, when the minimum value of the corresponding sample of the node and the minimum value of the coefficient of the foundation are lower than the threshold, the model is not divided any more;
s32422, if the number of the current node samples is smaller than a threshold value or no characteristic exists, returning to a decision sub-tree, and stopping recursion by the node;
s32423, calculating a base coefficient of the sample set, and if the base coefficient is smaller than a threshold value, returning to a decision tree subtree, wherein the node stops recursion;
s32424 calculating the base index of each dividing value of each technological parameter of the current node;
s32425, selecting a technological parameter with the minimum coefficient of the radix and a dividing point thereof as the optimal characteristic and the optimal dividing point of the current node, and generating a left sub node and a right sub node of the current node;
s32426, the steps are respectively recursively carried out on the left child node and the right child node, and finally a surface slag inclusion defect prediction model is generated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210093397.8A CN116562659A (en) | 2022-01-26 | 2022-01-26 | Steel surface slag inclusion defect cause analysis system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210093397.8A CN116562659A (en) | 2022-01-26 | 2022-01-26 | Steel surface slag inclusion defect cause analysis system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116562659A true CN116562659A (en) | 2023-08-08 |
Family
ID=87498724
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210093397.8A Pending CN116562659A (en) | 2022-01-26 | 2022-01-26 | Steel surface slag inclusion defect cause analysis system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116562659A (en) |
-
2022
- 2022-01-26 CN CN202210093397.8A patent/CN116562659A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108038553B (en) | Rolling mill equipment state on-line monitoring and diagnosing system and monitoring and diagnosing method | |
| CN110458474B (en) | Comprehensive evaluation method for physical quality of rolled cigarettes | |
| CN108256260A (en) | A kind of continuous casting billet quality Forecasting Methodology based on extreme learning machine | |
| CN110989510A (en) | Hot galvanizing product full-process quality control and grade automatic judgment system | |
| CN116307289B (en) | Textile processing procedure parameter detection and prediction method, system and storage medium | |
| CN112418538A (en) | Continuous casting billet inclusion prediction method based on random forest classification | |
| CN105302123B (en) | The monitoring method of on-line measurement data | |
| CN114985697B (en) | Crystallizer liquid level fluctuation monitoring method and system | |
| Zhou et al. | Application of time series data anomaly detection based on deep learning in continuous casting process | |
| CN112287550A (en) | Strip steel head thickness difference process parameter optimization method based on principal component analysis controller | |
| CN106054832B (en) | Multivariable-based dynamic online monitoring method and device for intermittent chemical production process | |
| Duan et al. | Longitudinal crack detection approach based on principal component analysis and support vector machine for slab continuous casting | |
| CN115106499A (en) | Crystallizer liquid level abnormal fluctuation distinguishing method and system | |
| CN114872290A (en) | Self-adaptive production abnormity monitoring method for injection molding part | |
| CN114326593A (en) | Tool life prediction system and method | |
| CN116562659A (en) | Steel surface slag inclusion defect cause analysis system and method | |
| CN112582032B (en) | High-thermal-stability iron-based soft magnetic amorphous alloy designed based on machine learning model | |
| Zhou et al. | Online abnormal interval detection and classification of industrial time series data based on multi-scale deep learning | |
| CN108364095B (en) | Molten steel quality diagnosis method in steelmaking production process based on data mining | |
| CN116703254A (en) | Production information management system for mechanical parts of die | |
| CN117036797A (en) | Continuous casting billet longitudinal crack prediction method based on feature extraction and random forest classification | |
| CN111366702A (en) | Online intelligent accurate slitting system and online intelligent accurate slitting method for non-oriented silicon steel | |
| CN112231368A (en) | Unary linear regression analysis method based on steel production big data | |
| VALEK et al. | The Quality Results of Slabs at Segregation Area | |
| CN113325811B (en) | Online industrial process anomaly detection method based on memory and forgetting strategy |
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 |