CN114565314A - Hot rolled steel coil end face quality control system and method based on digital twinning - Google Patents
Hot rolled steel coil end face quality control system and method based on digital twinning Download PDFInfo
- Publication number
- CN114565314A CN114565314A CN202210233179.XA CN202210233179A CN114565314A CN 114565314 A CN114565314 A CN 114565314A CN 202210233179 A CN202210233179 A CN 202210233179A CN 114565314 A CN114565314 A CN 114565314A
- Authority
- CN
- China
- Prior art keywords
- steel coil
- quality
- data
- production
- defect
- 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.)
- Granted
Links
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 224
- 239000010959 steel Substances 0.000 title claims abstract description 224
- 238000000034 method Methods 0.000 title claims abstract description 78
- 238000003908 quality control method Methods 0.000 title claims abstract description 12
- 230000007547 defect Effects 0.000 claims abstract description 126
- 238000004519 manufacturing process Methods 0.000 claims abstract description 71
- 238000003745 diagnosis Methods 0.000 claims abstract description 22
- 238000013135 deep learning Methods 0.000 claims abstract description 11
- 238000004088 simulation Methods 0.000 claims description 30
- 230000008569 process Effects 0.000 claims description 27
- 230000008859 change Effects 0.000 claims description 25
- 238000005096 rolling process Methods 0.000 claims description 25
- 238000005098 hot rolling Methods 0.000 claims description 21
- 238000007781 pre-processing Methods 0.000 claims description 16
- 238000011156 evaluation Methods 0.000 claims description 14
- 238000007726 management method Methods 0.000 claims description 12
- 230000006399 behavior Effects 0.000 claims description 10
- 238000010438 heat treatment Methods 0.000 claims description 10
- 238000012423 maintenance Methods 0.000 claims description 6
- 230000002708 enhancing effect Effects 0.000 claims description 5
- 239000000203 mixture Substances 0.000 claims description 5
- 230000000704 physical effect Effects 0.000 claims description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 5
- 238000010923 batch production Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 4
- 239000000463 material Substances 0.000 claims description 4
- 230000006872 improvement Effects 0.000 claims description 3
- 238000002788 crimping Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000005034 decoration Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 239000002436 steel type Substances 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 229910000677 High-carbon steel Inorganic materials 0.000 description 1
- 229910001209 Low-carbon steel Inorganic materials 0.000 description 1
- 229910000742 Microalloyed steel Inorganic materials 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000003889 chemical engineering Methods 0.000 description 1
- 238000005336 cracking Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 230000008929 regeneration Effects 0.000 description 1
- 238000011069 regeneration method Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
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
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- 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
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
-
- 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)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Manufacturing & Machinery (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a digital twin-based hot rolled steel coil end face quality control system and method, which comprises the following steps: the twin model building module is used for building a virtual model aiming at the production of the hot rolled steel coil; the real-time data acquisition and image acquisition module acquires relevant parameters in production in real time through a configured data interface, acquires a picture of the end face of the steel coil through an industrial camera, and preprocesses the picture; the quality judging module classifies and identifies defects through a deep learning method, judges the steel coil in a grading way based on a steel coil end face quality grading judging method, and sorts out the steel coil with unqualified quality; and the quality diagnosis module is used for analyzing reasons of the unqualified steel coil to determine an improved scheme, applying the improved scheme to production in a production workshop, and performing recheck after acquiring data to realize closed-loop control on the quality of the steel coil. The invention provides a new way for managing and controlling the end surface quality of a hot-rolled steel coil so as to improve the efficiency.
Description
Technical Field
The invention relates to the field of digital twins and computer science, in particular to a system and a method for managing and controlling the end face quality of a hot rolled steel coil based on the digital twins.
Background
The hot rolled steel coil is one of main products in the steel industry, and becomes an indispensable raw material for industries such as mechanical manufacturing, chemical engineering, aerospace, shipbuilding and the like, and the quality of the hot rolled steel coil directly influences the performance and the quality of a final product. The terminal surface quality is one of the most important index of hot rolling coil of strip, receives influence such as raw and other materials, processing equipment and processing technology in batching process, and defects such as decorative pattern limit, limit loss, burr, folding, limit crack of different degrees can appear, consequently how to find the defect fast accurate and carry out comprehensive evaluation to it, optimize production parameter, prevent the regeneration of defect, this is the coil of strip quality management and control an important problem that needs to solve urgently. The traditional steel coil quality control is only limited to the detection of the steel coil defects, and how to improve the steel coil quality control and the steel coil quality control after the defects is not solved in time. Therefore, the invention provides a steel coil end face quality control method based on a digital twin model, so as to realize online management, diagnosis and analysis of hot rolled steel coil quality data.
Disclosure of Invention
The technical problem of the invention is solved: the defects of the prior art are overcome, the digital twinning-based hot rolled steel coil end face quality control system and method are provided, effective control over the quality of a hot rolled steel coil is achieved based on a digital twinning model, and therefore efficiency is greatly improved.
The technical scheme of the invention is as follows: a digital twinning-based hot rolled steel coil end face quality control system comprises: the system comprises a twin model building module, a real-time data acquisition and image acquisition module, a quality judgment module and a quality diagnosis module;
the twin model building module is used for building a virtual model aiming at the production of the hot rolled steel coil;
the real-time data acquisition and image acquisition module is used for acquiring relevant parameters in real time during production through a configured data interface, loading the acquired data into the twin model, assisting in building a virtual model, acquiring a picture of the end face of the steel coil through an industrial camera, and preprocessing the picture to be used as a distinguished data set;
the quality judging module finishes the classification and identification of defects through a deep learning method according to the data set obtained after the pretreatment, performs classification and judgment on the steel coil based on a steel coil end surface quality classification judging method, finishes the quality classification of the steel coil, and then sorts out the steel coil with unqualified quality;
and the quality diagnosis module is used for finishing maintenance diagnosis of the quality of the steel coil according to the sorted out steel coil with unqualified quality. And (4) analyzing reasons of the steel coil with unqualified quality, determining an improved scheme, applying the improved scheme to production in a production workshop, and rechecking after data is acquired to realize closed-loop control on the steel coil quality.
The twin model construction is realized as follows:
drawing a geometric model, namely establishing the geometric model of the hot-rolled steel coil by acquiring four parts of information of the geometric size, the defect characteristics, the motion characteristics and the structural characteristics of the steel coil, and embodying the shape, the size, the structural composition and the assembly relation of the steel coil;
simulating physical properties, namely performing grid division on the steel coil, simulating and depicting surface pressure value change, strength change, heating conditions of different positions and pressure change in the steel coil production process, and showing specific physical quantity change through a chart and a numerical value;
establishing a behavior model, selecting a certain rolling process, obtaining a process curve, obtaining the conditions of temperature change, rolling speed, rolling width and thickness and the like of the process according to the process curve, and importing data into the model for simulation;
and adding a rule model, and adding related rules for managing and controlling the quality of the hot rolled steel coil, wherein the related rules comprise rule rules based on historical associated data, operation rules of a hot rolling system, quality standards of the end surface of the steel coil and a steel coil knowledge base.
The real-time data acquisition and image acquisition module is realized as follows:
(1) the data acquisition interface is used for communicating with a hot rolling system, an enterprise resource planning system and the like to acquire real-time data of the steel coil hot rolling production process;
(2) the image acquisition is to set a shooting point after a coiling procedure in the hot rolling process through an industrial camera to acquire image data of the end surface of the steel coil, and each steel coil acquires 16 pictures at different positions and angles;
(3) and carrying out preprocessing such as scanning, fitting, filling with overflowing water and the like on the acquired image, and removing the background and enhancing the defect characteristics.
The quality discrimination module is implemented as follows:
(1) after preprocessing operation is carried out on a batch of collected steel coil end face pictures, identifying the end face defects of the pictures by adopting a defect identification model constructed based on a deep learning method, and acquiring the end face defects of the pictures;
(2) determining the grade of the steel coil to be detected based on a steel coil end face quality grading judgment method; determining the standard quality grade of the steel coil based on the corresponding relation between the steel coil grade and the standard quality grade, and sorting out the steel coil with unqualified quality to carry out the next quality diagnosis;
the method for judging the quality grading of the end face of the steel coil comprises the following steps:
acquiring 16 pictures from different position angles of each steel coil, identifying end surface defects through a constructed defect identification model, acquiring the end surface defects of each picture, and grading each defect according to the severity of the defect, such as level 1 edge loss, level 2 edge crack and the like to acquire the grade defect of each picture;
determining the evaluation grade of each picture for the grade defects and the weight corresponding to the evaluation grade according to the number of the grade defects in each picture, and further determining the steel coil grade corresponding to each grade defect;
and thirdly, determining the weight corresponding to each steel coil grade based on the steel coil grade corresponding to each grade defect, and determining the steel coil grade of the detected steel coil.
The quality diagnosis module is implemented as follows:
(1) counting defect information of the steel coil with unqualified quality, wherein the defect information comprises a steel coil number, a defect type, a defect picture and defect position information;
(2) by comparing actual data acquired during process production with simulation data in virtual space simulation production, the abnormity among the data is discovered, accurate positioning is realized, and the reason is found out;
(3) based on the determined reasons, parameters in the process production are properly adjusted, the production is simulated in a virtual space, the steel coil data obtained after the production of the virtual space is obtained, the steel coil quality is judged from three aspects of mechanical performance, surface appearance and size precision, if the steel coil quality meets the requirements, an improved scheme is determined and applied to a physical production workshop for production, meanwhile, the production process is synchronously monitored, the obtained data is rechecked, if the quality is qualified, the batch production is carried out according to the improved scheme, and if the quality is unqualified, the management and control are continued.
The invention relates to a method for measuring and controlling the quality of an end face of a hot rolled steel coil based on digital twinning, which comprises the following steps:
the method comprises the following steps of (1) twin model construction, namely constructing a high fidelity model according to physical space geometric parameters and material attributes of the hot rolled steel coil, constructing a super-realistic simulation environment by utilizing offline data such as environmental parameters and position parameters, and completing high-fidelity behavior simulation according to online data such as operating data and dynamic parameters, wherein the twin model construction is specifically realized as follows:
drawing a geometric model, namely establishing the geometric model of the hot-rolled steel coil by acquiring the information of four parts, namely the geometric size, the defect characteristics, the motion characteristics and the structure characteristics of the steel coil, and embodying the shape, the size, the structure composition and the assembly relation of the steel coil;
simulating physical properties, namely performing grid division on the steel coil, simulating and depicting surface pressure value change, strength change, heating conditions of different positions and pressure change in the steel coil production process, and showing specific physical quantity change through a chart and a numerical value;
establishing a behavior model, selecting a certain rolling process, obtaining a process curve, obtaining the conditions of temperature change, rolling speed, rolling width and thickness and the like of the process according to the process curve, and importing data into the model for simulation;
adding a rule model, and adding related rules for managing and controlling the quality of the hot rolled steel coil, wherein the related rules comprise rule rules based on historical associated data, operation rules of a hot rolling system, quality standards of the end surface of the steel coil and a steel coil knowledge base;
step (2), real-time data acquisition and image acquisition, wherein production real-time data are acquired based on an acquisition device, a twin model is loaded to assist in constructing a virtual model, and acquired pictures are preprocessed to serve as a distinguished data set, and the method is specifically realized as follows:
firstly, a data acquisition interface is used for communicating with a hot rolling system, an enterprise resource planning system and the like to acquire real-time data of a steel coil hot rolling production process;
secondly, image acquisition is carried out by setting a shooting point after a coiling procedure in the hot rolling process through an industrial camera to acquire image data of the end surface of the steel coil, and each steel coil acquires 16 pictures at different positions and angles;
preprocessing the collected image by scanning, fitting, filling with overflowing water and the like, and removing a background and enhancing defect characteristics;
and (3) quality judgment, namely completing defect classification and identification through a deep learning method according to the data set obtained after preprocessing, and completing steel coil quality classification through a quality grading judgment method, wherein the method is specifically realized as follows:
firstly, preprocessing a batch of collected steel coil end face pictures, and then identifying the end face defects of the pictures by adopting a defect identification model constructed based on a deep learning method to obtain the end face defects of the pictures;
determining the grade of the steel coil to be detected based on a steel coil end face quality grading judgment method, determining the standard quality grade of the steel coil based on the corresponding relation between the grade of the steel coil and the standard quality grade, and sorting out the steel coil with unqualified quality to carry out the next quality diagnosis;
and (4) quality diagnosis, namely finishing maintenance diagnosis of the quality of the steel coil according to the sorted out steel coils with unqualified quality, and specifically realizing the following steps:
counting defect information of a steel coil with unqualified quality, wherein the defect information comprises information such as a steel coil number, a defect type, a defect picture, a defect position and the like;
secondly, by comparing actual data acquired during process production with simulation data in virtual space simulation production, the abnormity among the data is discovered, the accurate positioning is realized, and the reason is found out;
based on the determined reasons, parameters in the process production are properly adjusted, production is simulated in a virtual space, steel coil data obtained after the virtual space is produced are obtained, the steel coil quality is judged from three aspects of mechanical performance, surface appearance and size precision, if the steel coil quality meets requirements, an improved scheme is determined, the improved scheme is applied to a physical production workshop for production, synchronous monitoring is carried out on the production process, the data is obtained for rechecking, if the quality is qualified, batch production can be carried out according to the improved scheme, and if the quality is unqualified, management and control are continued.
Compared with the prior art, the invention has the advantages that:
(1) the omnibearing functional service is realized. At present, the quality control of the steel coil only aims at the detection of the defects of the steel coil, and the comprehensive management of the quality of the steel coil is lacked, the quality control of the end surface of the steel coil is combined with digital twinning, the production behavior of the hot-rolled steel coil is simulated by using a high-fidelity digital twinning simulation model, and the whole process of gathering data in real time, intelligently finding problems and accurately finding countermeasures is realized;
(2) the method has the advantages that the digital twin model of the end surface defect of the hot-rolled steel coil is constructed, the production behavior of the steel coil is simulated, the tracing of the reason and the determination of the improvement scheme can be completed according to simulation data, the problems can be found in time, the problems are solved, and the quality maintenance management efficiency of the steel coil is improved.
Drawings
FIG. 1 is a block diagram of an embodiment of the method of the present invention.
Detailed Description
The present invention is described in further detail below with reference to the attached drawings.
The method comprises the following steps of (1) constructing a twin model, constructing a high-fidelity model according to physical space geometric parameters and material properties of the hot-rolled steel coil, constructing a super-realistic simulation environment by utilizing offline data such as environmental parameters and position parameters, and completing high-fidelity behavior simulation according to online data such as running data and dynamic parameters, wherein the twin model is specifically realized as follows:
drawing a geometric model, namely establishing the geometric model of the hot-rolled steel coil by acquiring four parts of information of the geometric size, the defect characteristics, the motion characteristics and the structural characteristics of the steel coil, and embodying the shape, the size, the structural composition and the assembly relation of the steel coil;
simulating physical properties, namely performing grid division on the steel coil, simulating and depicting surface pressure value change, strength change, heating conditions of different positions and pressure change in the steel coil production process, and showing specific physical quantity change through a chart and a numerical value;
establishing a behavior model, selecting a certain rolling process, obtaining a process curve, obtaining the conditions of temperature change, rolling speed, rolling width and thickness and the like of the process according to the process curve, and importing data into the model for simulation;
adding a rule model, and adding related rules for managing and controlling the quality of the hot rolled steel coil, wherein the related rules comprise rule rules based on historical associated data, operation rules of a hot rolling system, quality standards of the end surface of the steel coil and a steel coil knowledge base;
step (2), acquiring real-time data and images, acquiring production real-time data based on an acquisition device, loading the production real-time data into a twin model to assist in building a virtual model, preprocessing the acquired pictures to serve as a distinguished data set, and specifically realizing the following steps:
firstly, a data acquisition interface is used for communicating with a hot rolling system, an enterprise resource planning system and the like to obtain real-time data of a steel coil hot rolling production process, including finish rolling inlet temperature TSFinish rolling temperature TECrimping temperature TJSlab heating temperature TBHeating time t, rolling width L, rolling thickness B, and rolling speed VZCrimping speed VJ；
Secondly, image acquisition is carried out by setting a shooting point after a coiling procedure in the hot rolling process through an industrial camera to acquire image data of the end surface of the steel coil, and each steel coil acquires 16 pictures at different positions and angles;
preprocessing collected images, such as scanning, fitting, filling with overflowing water and the like, eliminating background and enhancing defect characteristics, setting segmentation parameters of the images, marking areas corresponding to the matrix as clear, transition and fuzzy areas, establishing a region selection template segmentation template based on definition, wherein the clear, transition and fuzzy areas respectively adopt an original image and a filling background color, further calculating multi-level definition, and reducing the region selection range to 16x16 pixels to form a region selection template data source;
and (3) quality judgment, namely completing defect classification and identification through a deep learning method according to the data set obtained after preprocessing, and completing steel coil quality classification through a quality grading judgment method, wherein the method is specifically realized as follows:
firstly, slicing a batch of collected steel coil end face pictures layer by layer through a deep neural network, amplifying local information and defect characteristics of the pictures, extracting the defect characteristics of each layer, designing and developing a defect identification system based on a deep learning method, carrying out defect detection on the batch of steel coil end face pictures by using a constructed training model, designing a corresponding identification algorithm by analyzing texture characteristics and spectral information of typical defects such as tower shapes, edge losses, folds, burrs and the like, completing accurate classification identification and quick marking of the defects, marking the defect information of each position on the pictures, indicating the defect position and showing the defect in a visual form;
determining the grade of the steel coil to be detected based on a steel coil end face quality grading judgment method, determining the standard quality grade of the steel coil based on the corresponding relation between the grade of the steel coil and the standard quality grade, and sorting out the steel coil with unqualified quality to carry out the next quality diagnosis;
and (4) quality diagnosis, namely finishing maintenance diagnosis of the quality of the steel coil according to the sorted out steel coils with unqualified quality, and specifically realizing the following steps:
recording the defect identification result of a steel coil with unqualified quality, wherein the defect identification result comprises information such as a steel coil number, a defect type, a defect picture, a defect position, defect characteristics, a defect grade and the like, establishing a five-defect database of pattern edges, edge loss, burrs, folding and edge cracking, storing the pattern into a corresponding database according to the defect type, and being beneficial to later-stage defect reason search, wherein the table 1 shows the defect identification result of the unqualified steel coil;
TABLE 1
| Time | Steel coil number | Defective picture | Location of defect | Type of defect | Defect characteristics | Grade of defect |
| xx.xx.xx | 001-1 | |||||
| xx.xx.xx | 001-2 | |||||
| xx.xx.xx | 001-3 | |||||
| xx.xx.xx | 001-4 | |||||
| xx.xx.xx | 001-5 |
Secondly, on the basis of a digital twin intelligent workshop, the production process of a simulation physical workshop is utilized, each step of procedure is truly restored, simulation data obtained by steel coil simulation production in a virtual workshop is obtained, the simulation data is compared with actual data obtained in the process production of the physical workshop, flow difference feedback and historical link tracing are realized by means of virtual-real comparison, the stress condition and the heating condition of each part of a steel coil in each procedure are analyzed, the abnormity among the data can be rapidly found, so that the precise positioning of reasons is realized, and the comparison between the actual data and the simulation data is displayed in a table 2;
TABLE 2
| Actual data | Simulation data | |
| Inlet temperature T of finish rollingS | * | * |
| Finish rolling temperature TE | * | * |
| Crimping temperature TJ | * | * |
| Slab heating temperature TB | * | * |
| Heating time t | * | * |
| Rolling width L | * | * |
| Rolling thickness B | * | * |
| Rolling speed VZ | * | * |
| Crimping speed VJ | * | * |
Based on the reasons obtained by analysis, parameters in the process production are pertinently adjusted within a normal range, simulation test production is carried out in a virtual workshop, specific data of the steel coil after the simulation production are obtained, and whether the steel coil is within a normal range of a preset value is judged;
specifically, the judgment of the quality of the steel coil is mainly carried out from three aspects of mechanical property, surface appearance and dimensional accuracy, taking the mechanical property as an example, the mechanical property relates to strength and toughness, the Yield Strength (YS), the Tensile Strength (TS) and the Elongation (EL) are represented, several steel types with the maximum production in a steel mill are considered, including low-carbon steel, high-carbon steel and microalloyed steel, the yield strength is 700MPa, the tensile strength is 920MPa, a preset value is set for different steel types, the normal range of the yield strength and the tensile strength is +/-20 MPa, the elongation is +/-5%, after the normal range is determined, the YS, the TS and the EL of the steel coil produced in a virtual workshop simulation mode are obtained, whether the normal range of the preset value is judged, if the normal range meets the requirement, an improved scheme is determined, the improved scheme is applied to a physical production workshop for production, and the production process is synchronously monitored, obtaining data for rechecking, if the quality is qualified, carrying out batch production according to the improved scheme, recording the reason and the corresponding solution in a knowledge base, and if the quality is unqualified, continuing to control;
further, the method for judging quality grading of the end face of the steel coil in the step (3) comprises the following steps:
collecting 16 pictures from different positions of each steel coil, identifying end surface defects through a constructed defect identification model, obtaining the end surface defects of each picture, and grading each defect according to the severity of the defect, wherein 3 grades are set, namely 1 grade, 2 grade and 3 grade, and the severity is from light to heavy, namely 1 grade is light, 2 grade is medium and 3 grade is heavy. Assuming that, taking the defect of edge crack as an example, L represents the number (one) of edge cracks, H represents the depth (mm) of the crack, and level 1 is defined as L not exceeding 10 or H not exceeding 10mm, in this way, the level defect existing in each picture can be determined, such as the level 3 pattern edge existing in the 1 st picture of the steel coil;
determining the steel coil grade corresponding to each grade defect based on the grade defect of each picture, and concretely realizing the following steps: determining the evaluation grade of each picture aiming at the grade defects according to the number of the grade defects of each picture, and setting the weight corresponding to the evaluation grade according to the severity of the defects so as to determine the steel coil grade corresponding to each grade defect; specifically, the steel coil grades corresponding to the defects of different grades are calculated, the evaluation grade of each picture for the defects of the grade needs to be determined, the following table shows the number of the end surface defects of a certain grade and the evaluation grades corresponding to the different numbers, specifically, as shown in the following table 3, it needs to be noted that the numbers in the table represent the number of the defects of a certain grade in one picture, such as 3-grade pattern edges, and 1-5 defects appear in one picture, which represents that the evaluation grade of the picture for the picture is 1;
TABLE 3
Further, based on the evaluation level of each grade of defect of each picture and the set weight corresponding to each evaluation level, the steel coil grade of each grade of defect can be determined by adopting a weighted average method. Taking the defect of the pattern edge of the level 2 as AN example, the evaluation grade of the picture 1 is a1, the weight corresponding to a1 is B1, and so on, the evaluation grade of the picture N is AN, the weight corresponding to AN is BN, each steel coil has 16 pictures, the steel coil grade of the pattern edge of the level 2 can be determined according to the formula, and the steel coil grade of the pattern edge of the level 2 is (a1 × B1+ a2 × B2+ … … + AN × BN)/(B1+ B2+ … … + BN), if the calculation result contains decimal, the decimal can be rounded to AN integer;
after the steel coil grades corresponding to the defects of different grades are obtained, the whole steel coil grade is determined by continuously adopting a weighted average method, the weights corresponding to the different steel coil grades are preset, and the whole steel coil grade is determined according to which grade defects are contained in the whole steel coil, and the method specifically comprises the following steps:
the whole steel coil comprises: the defect grade D1 of C1, the coil grade of this defect is X1, the weight that X1 corresponds is Y1, so on, CN grade defect DN, the coil grade of this defect is XN, the weight that XN corresponds is YN, then the whole coil grade of coil of strip to be detected (X1 XY 1+ X2 XY 2+ … … + XN XYN)/(Y1 + Y2+ … … + YN), the decimal number appears and is directly rounded into the integer;
based on the integral steel coil grade, the quality of the steel coil to be detected can be determined according to the corresponding relation between the grade and the standard quality of the steel coil, and the corresponding relation is shown in table 4;
TABLE 4
When unqualified steel coils appear, a red alarm signal is sent out, steel coils with the grade of 1-4 are sorted out in time for diagnosis, and a green signal is sent out to indicate that the quality passes.
Those skilled in the art will appreciate that the invention may be practiced without these specific details.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (6)
1. The utility model provides a hot rolled steel coil terminal surface quality management and control system based on digit twinning which characterized in that includes: the system comprises a twin model building module, a real-time data acquisition and image acquisition module, a quality judgment module and a quality diagnosis module;
the twin model building module is used for building a virtual model aiming at the production of the hot rolled steel coil;
the real-time data acquisition and image acquisition module is used for acquiring relevant parameters in production in real time through a configured data interface, loading the acquired data into a twin model, assisting in building a virtual model, acquiring a picture of the end face of the steel coil through an industrial camera, and preprocessing the picture to be used as a distinguished data set;
the quality judging module finishes the classification and identification of defects through a deep learning method according to the data set obtained after the pretreatment, performs classification and judgment on the steel coil based on a steel coil end surface quality classification judging method, finishes the quality classification of the steel coil, and then sorts out the steel coil with unqualified quality;
and the quality diagnosis module is used for finishing maintenance diagnosis of the quality of the steel coil according to the sorted out steel coil with unqualified quality, analyzing reasons of the steel coil with unqualified quality, determining an improved scheme, applying the improved scheme to production in a production workshop, and performing rechecking after data is acquired to realize closed-loop control on the quality of the steel coil.
2. The digitally twinned hot rolled steel coil end face quality management and control system as claimed in claim 1, wherein: the twin model construction is realized as follows:
drawing a geometric model, namely establishing the geometric model of the hot-rolled steel coil by acquiring four parts of information of the geometric size, the defect characteristics, the motion characteristics and the structural characteristics of the steel coil, and embodying the shape, the size, the structural composition and the assembly relation of the steel coil;
simulating physical properties, namely performing grid division on the steel coil, simulating and depicting surface pressure value change, strength change, heating conditions of different positions and pressure change in the steel coil production process, and showing specific physical quantity change through a chart and a numerical value;
establishing a behavior model, selecting a certain rolling process, obtaining a process curve, obtaining the temperature change, the rolling speed and the rolling width and thickness conditions of the process according to the process curve, and importing data into the model for simulation;
and adding a rule model, and adding related rules for quality control of the hot rolled steel coil, wherein the related rules comprise rule rules based on historical associated data, operation rules of a hot rolling system, quality standards of the end face of the steel coil and a steel coil knowledge base.
3. The digitally twinned hot rolled steel coil end face quality management and control system as claimed in claim 1, wherein: the real-time data acquisition and image acquisition module is realized as follows:
(1) the data acquisition interface is used for communicating with a hot rolling system, an enterprise resource planning system and the like to acquire real-time data of the steel coil hot rolling production process;
(2) the image acquisition is to set a shooting point after a coiling procedure in the hot rolling process through an industrial camera to acquire image data of the end surface of the steel coil, and each steel coil acquires 16 pictures at different positions and angles;
(3) and carrying out preprocessing such as scanning, fitting, filling with overflowing water and the like on the acquired image, and removing the background and enhancing the defect characteristics.
4. The digitally twinned hot rolled steel coil end face quality management and control system as claimed in claim 1, wherein: the quality discrimination module is implemented as follows:
(1) after preprocessing a batch of collected steel coil end face pictures, identifying the end face defects of the pictures by adopting a defect identification model constructed based on a deep learning method to obtain the end face defects of the pictures;
(2) determining the grade of the steel coil to be detected based on a steel coil end face quality grading judgment method; determining the standard quality grade of the steel coil based on the corresponding relation between the steel coil grade and the standard quality grade, and sorting out the steel coil with unqualified quality to carry out the next quality diagnosis;
the method for judging the quality grading of the end face of the steel coil comprises the following steps:
acquiring 16 pictures from different positions and angles of each steel coil, identifying end surface defects through a constructed defect identification model, obtaining the end surface defects of each picture, grading each defect according to the severity of the defect, and obtaining the grade defect of each picture from light to heavy;
determining the evaluation grade of each picture for the grade defects and the weight corresponding to the evaluation grade according to the number of the grade defects in each picture, and further determining the steel coil grade corresponding to each grade defect;
and thirdly, determining the weight corresponding to each steel coil grade based on the steel coil grade corresponding to each grade defect, and determining the steel coil grade of the detected steel coil.
5. The digitally twinned hot rolled steel coil end face quality management and control system as claimed in claim 1, wherein: the quality diagnosis module is implemented as follows:
(1) counting defect information of the steel coil with unqualified quality, wherein the defect information comprises a steel coil number, a defect type, a defect picture and defect position information;
(2) by comparing actual data acquired during process production with simulation data in virtual space simulation production, the abnormity among the data is discovered, the accurate positioning is realized, and the reason is found out;
(3) based on the determined reasons, parameters in the process production are properly adjusted, the production is simulated in a virtual space, the steel coil data obtained after the production of the virtual space is obtained, the steel coil quality is judged from three aspects of mechanical performance, surface appearance and size precision, if the steel coil quality meets the requirements, an improved scheme is determined and applied to a physical production workshop for production, meanwhile, the production process is synchronously monitored, the obtained data is rechecked, if the quality is qualified, the batch production is carried out according to the improved scheme, and if the quality is unqualified, the management and control are continued.
6. A method for controlling the quality of the end face of a hot rolled steel coil based on digital twinning is characterized by comprising the following steps:
the method comprises the following steps of (1) constructing a twin model, constructing a high fidelity model according to physical space geometric parameters and material properties of a hot rolled steel coil, constructing a super-realistic simulation environment by utilizing environment parameters and position parameter offline data, and completing high-realistic behavior simulation according to running data and dynamic parameter online data, wherein the twin model is specifically realized as follows:
drawing a geometric model, namely establishing the geometric model of the hot-rolled steel coil by acquiring the information of four parts, namely the geometric size, the defect characteristics, the motion characteristics and the structure characteristics of the steel coil, and embodying the shape, the size, the structure composition and the assembly relation of the steel coil;
simulating physical properties, namely performing grid division on the steel coil, simulating and depicting surface pressure value change, strength change, heating conditions of different positions and pressure change in the steel coil production process, and showing specific physical quantity change through a chart and a numerical value;
establishing a behavior model, selecting a certain rolling process, obtaining a process curve, obtaining the conditions of temperature change, rolling speed, rolling width and thickness and the like of the process according to the process curve, and importing data into the model for simulation;
adding a rule model, and adding related rules for managing and controlling the quality of the hot rolled steel coil, wherein the related rules comprise rule rules based on historical associated data, operation rules of a hot rolling system, quality standards of the end surface of the steel coil and a steel coil knowledge base;
step (2), acquiring real-time data and images, acquiring production real-time data based on an acquisition device, loading the production real-time data into a twin model to assist in building a virtual model, preprocessing the acquired pictures to serve as a distinguished data set, and specifically realizing the following steps:
firstly, a data acquisition interface is used for communicating with a hot rolling system, an enterprise resource planning system and the like to acquire real-time data of a steel coil hot rolling production process;
image acquisition is to set a shooting point after a coiling procedure in a hot rolling process through an industrial camera to acquire image data of the end faces of steel coils, and each steel coil acquires 16 pictures at different positions and angles;
preprocessing the collected image by scanning, fitting, filling with overflowing water and the like, and removing a background and enhancing defect characteristics;
and (3) quality judgment, namely completing defect classification and identification through a deep learning method according to the data set obtained after preprocessing, and completing steel coil quality classification through a quality grading judgment method, wherein the method is specifically realized as follows:
firstly, preprocessing a batch of collected steel coil end face pictures, and then identifying the end face defects of the pictures by adopting a defect identification model constructed based on a deep learning method to obtain the end face defects of the pictures;
determining the grade of the steel coil to be detected based on a steel coil end face quality grading judgment method, determining the standard quality grade of the steel coil based on the corresponding relation between the grade of the steel coil and the standard quality grade, and sorting out the steel coil with unqualified quality to carry out the next quality diagnosis;
and (4) quality diagnosis, namely finishing maintenance diagnosis of the quality of the steel coil according to the sorted out steel coils with unqualified quality, and specifically realizing the following steps:
counting defect information of a steel coil with unqualified quality, wherein the defect information comprises information such as a steel coil number, a defect type, a defect picture, a defect position and the like;
secondly, by comparing actual data acquired during process production with simulation data in virtual space simulation production, the abnormity among the data is discovered, the accurate positioning is realized, and the reason is found out;
and thirdly, based on the determined reasons, parameters in the process production are properly adjusted, the production is simulated in a virtual space, the steel coil data obtained after the virtual space is produced are obtained, the steel coil quality is judged from three aspects of mechanical performance, surface appearance and dimensional precision, if the steel coil quality meets the requirements, the improvement scheme is determined and applied to a physical production workshop for production, meanwhile, the production process is synchronously monitored, the data are rechecked, if the quality is qualified, the steel coil can be produced in batches according to the improvement scheme, and if the quality is unqualified, the control is continued.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210233179.XA CN114565314B (en) | 2022-03-09 | 2022-03-09 | Digital twinning-based hot rolled steel coil end face quality control system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210233179.XA CN114565314B (en) | 2022-03-09 | 2022-03-09 | Digital twinning-based hot rolled steel coil end face quality control system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114565314A true CN114565314A (en) | 2022-05-31 |
| CN114565314B CN114565314B (en) | 2024-06-18 |
Family
ID=81717915
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210233179.XA Active CN114565314B (en) | 2022-03-09 | 2022-03-09 | Digital twinning-based hot rolled steel coil end face quality control system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114565314B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115055385A (en) * | 2022-06-09 | 2022-09-16 | 黄冈师范学院 | Automatic machine vision sheet defect eliminating method |
| CN115795983A (en) * | 2023-01-29 | 2023-03-14 | 江苏沙钢集团有限公司 | Wire quality control method, device, equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160203593A1 (en) * | 2013-09-10 | 2016-07-14 | Thyssenkrupp Steel Europe Ag | Method and device for testing an inspection system for detecting surface defects |
| CN110889238A (en) * | 2019-12-06 | 2020-03-17 | 北京航空航天大学 | Autoclave management and control system driven by digital twin |
| CN111695734A (en) * | 2020-06-12 | 2020-09-22 | 中国科学院重庆绿色智能技术研究院 | Multi-process planning comprehensive evaluation system and method based on digital twin and deep learning |
| WO2021169598A1 (en) * | 2020-02-27 | 2021-09-02 | 广东工业大学 | Digital twinning-based physical simulation method for production line and system thereof |
| CN113373295A (en) * | 2021-04-22 | 2021-09-10 | 上海孟伯智能物联网科技有限公司 | Annealing furnace digital twin intelligent alarm system and method based on multilayer perceptron |
| CN113930600A (en) * | 2021-10-12 | 2022-01-14 | 大连理工大学 | Bell-type furnace annealing process monitoring and control method based on digital twin technology |
-
2022
- 2022-03-09 CN CN202210233179.XA patent/CN114565314B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160203593A1 (en) * | 2013-09-10 | 2016-07-14 | Thyssenkrupp Steel Europe Ag | Method and device for testing an inspection system for detecting surface defects |
| CN110889238A (en) * | 2019-12-06 | 2020-03-17 | 北京航空航天大学 | Autoclave management and control system driven by digital twin |
| WO2021169598A1 (en) * | 2020-02-27 | 2021-09-02 | 广东工业大学 | Digital twinning-based physical simulation method for production line and system thereof |
| CN111695734A (en) * | 2020-06-12 | 2020-09-22 | 中国科学院重庆绿色智能技术研究院 | Multi-process planning comprehensive evaluation system and method based on digital twin and deep learning |
| CN113373295A (en) * | 2021-04-22 | 2021-09-10 | 上海孟伯智能物联网科技有限公司 | Annealing furnace digital twin intelligent alarm system and method based on multilayer perceptron |
| CN113930600A (en) * | 2021-10-12 | 2022-01-14 | 大连理工大学 | Bell-type furnace annealing process monitoring and control method based on digital twin technology |
Non-Patent Citations (2)
| Title |
|---|
| 张振宇;陈晓波;: "基于深度学习模型识别热轧钢卷边部缺陷的探索", 冶金自动化, no. 03, 16 May 2019 (2019-05-16) * |
| 张雪荣: "基于深度学习的钢卷端面缺陷检测系统设计", 《计算机集成制造系统》, 8 March 2022 (2022-03-08) * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115055385A (en) * | 2022-06-09 | 2022-09-16 | 黄冈师范学院 | Automatic machine vision sheet defect eliminating method |
| CN115055385B (en) * | 2022-06-09 | 2024-03-22 | 黄冈师范学院 | Automatic defect eliminating method for machine vision sheet |
| CN115795983A (en) * | 2023-01-29 | 2023-03-14 | 江苏沙钢集团有限公司 | Wire quality control method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114565314B (en) | 2024-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115082467B (en) | Building material welding surface defect detection method based on computer vision | |
| CN114565314A (en) | Hot rolled steel coil end face quality control system and method based on digital twinning | |
| CN107607554A (en) | A kind of Defect Detection and sorting technique of the zinc-plated stamping parts based on full convolutional neural networks | |
| CN113298757A (en) | Metal surface defect detection method based on U-NET convolutional neural network | |
| CN111815555A (en) | Metal additive manufacturing image detection method and device combining anti-neural network with local binary | |
| CN114066820A (en) | Fabric defect detection method based on Swin-transducer and NAS-FPN | |
| CN111402236B (en) | Hot rolled strip steel surface defect grading method based on image gray value | |
| CN115035106A (en) | Strip steel defect intelligent detection method | |
| CN113822889A (en) | Method for detecting surface defects of hot-rolled steel plate | |
| CN115797361B (en) | Aluminum template surface defect detection method | |
| CN116519699A (en) | Structural damage detection system and structural damage detection method | |
| CN114881998A (en) | Workpiece surface defect detection method and system based on deep learning | |
| CN117252864B (en) | Steel production device smoothness detection system based on identification analysis | |
| CN114155226A (en) | Micro defect edge calculation method | |
| CN117392097A (en) | Additive manufacturing process defect detection method and system based on improved YOLOv8 algorithm | |
| CN115345846A (en) | Intelligent grading method and system for grain size of medium and low carbon steel | |
| CN110763705A (en) | Deep learning identification method and system based on X-ray image and X-ray machine | |
| CN114757916A (en) | Industrial CT image defect classification method based on feature extraction and BP network | |
| CN115457323A (en) | Visual inspection system-based method for classifying non-uniform surface defects of mobile phone glass screen | |
| CN115656182A (en) | Sheet material point cloud defect detection method based on tensor voting principal component analysis | |
| CN115546108A (en) | Intelligent detection method for appearance quality of automobile tire based on edge cloud cooperation and AR | |
| CN108133261B (en) | Automobile rubber hose mandrel quality evaluation method based on BP neural network | |
| CN114359366A (en) | Machine vision measuring method and application thereof in bearing workpiece measurement | |
| CN115830403B (en) | Automatic defect classification system and method based on deep learning | |
| CN117974665B (en) | Metal mold on-line detection method and equipment based on computer vision |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |