CN113835411A - Comprehensive diagnosis method for abnormal quality of steel rolling process flow - Google Patents
Comprehensive diagnosis method for abnormal quality of steel rolling process flow Download PDFInfo
- Publication number
- CN113835411A CN113835411A CN202111045843.XA CN202111045843A CN113835411A CN 113835411 A CN113835411 A CN 113835411A CN 202111045843 A CN202111045843 A CN 202111045843A CN 113835411 A CN113835411 A CN 113835411A
- Authority
- CN
- China
- Prior art keywords
- quality
- fault
- hierarchical
- fault propagation
- variables
- 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
- 238000000034 method Methods 0.000 title claims abstract description 102
- 238000005096 rolling process Methods 0.000 title claims abstract description 65
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 60
- 239000010959 steel Substances 0.000 title claims abstract description 60
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 28
- 238000003745 diagnosis Methods 0.000 title claims abstract description 27
- 230000005856 abnormality Effects 0.000 claims abstract description 59
- 238000004519 manufacturing process Methods 0.000 claims abstract description 35
- 238000012544 monitoring process Methods 0.000 claims abstract description 13
- 230000008569 process Effects 0.000 claims description 44
- 239000013598 vector Substances 0.000 claims description 19
- 230000001364 causal effect Effects 0.000 claims description 16
- 238000004458 analytical method Methods 0.000 claims description 15
- 230000004927 fusion Effects 0.000 claims description 13
- 230000008859 change Effects 0.000 claims description 12
- 238000009826 distribution Methods 0.000 claims description 12
- 238000010586 diagram Methods 0.000 claims description 9
- 238000007726 management method Methods 0.000 claims description 9
- 238000004886 process control Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 7
- 230000008878 coupling Effects 0.000 claims description 6
- 238000010168 coupling process Methods 0.000 claims description 6
- 238000005859 coupling reaction Methods 0.000 claims description 6
- 239000002994 raw material Substances 0.000 claims description 6
- 230000007704 transition Effects 0.000 claims description 5
- 238000012546 transfer Methods 0.000 claims description 4
- 238000013439 planning Methods 0.000 description 4
- 239000000047 product Substances 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000001816 cooling Methods 0.000 description 2
- 238000010219 correlation analysis Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000010438 heat treatment Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000005452 bending Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000012824 chemical production Methods 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 238000005098 hot rolling Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000010206 sensitivity analysis Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000012384 transportation and delivery Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0259—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the response to fault detection
- G05B23/0262—Confirmation of fault detection, e.g. extra checks to confirm that a failure has indeed occurred
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24065—Real time diagnostics
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- General Factory Administration (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention provides a comprehensive diagnosis method for abnormal quality of a steel rolling process flow, and belongs to the technical field of control and monitoring of a production process. The method comprises the following steps: constructing a hierarchical fault propagation network fusing hierarchical quality information; identifying a fault propagation path under a variable working condition according to the constructed hierarchical fault propagation network; and performing online diagnosis on the quality abnormality related fault according to the fault propagation path identification result under the variable working condition. By adopting the method and the device, the quality abnormity of the steel rolling process flow can be comprehensively diagnosed on line.
Description
Technical Field
The invention relates to the technical field of control and monitoring of a production process, in particular to a comprehensive diagnosis method for quality abnormity of a steel rolling process flow.
Background
The modern industrial process has the characteristics of large production scale, high automation degree, obvious system integration and the like. The steel rolling process flow is often composed of a plurality of production processes/links/loops (also called as subsystems), and the subsystems are mutually influenced under the constraints of material flow, energy flow and information flow; meanwhile, the corresponding integrated automation system levels (such as a real-time control layer, a process control layer, a manufacturing execution layer and the like) are clearly divided, the work of each level is clear and the levels are in cooperative association, and the abnormal change of each level has the possibility of triggering quality abnormity; in addition, the variety and specification of the customized production are diversified, and the production working conditions are complex and changeable and the switching is frequent due to the uncertainty of raw materials, the difference of equipment states, external environments, process routes and the like; the combined action of the three-dimensional manufacturing modes of multi-level, full flow and variable working conditions brings challenges to the comprehensive diagnosis of abnormal quality of the steel rolling process flow.
Disclosure of Invention
The embodiment of the invention provides a comprehensive diagnosis method for quality abnormity of a steel rolling process flow, which can perform online diagnosis on the quality abnormity of the steel rolling process flow. The technical scheme is as follows:
the embodiment of the invention provides a comprehensive diagnosis method for abnormal quality of a steel rolling process flow, which comprises the following steps:
constructing a hierarchical fault propagation network fusing hierarchical quality information;
identifying a fault propagation path under a variable working condition according to the constructed hierarchical fault propagation network;
and performing online diagnosis on the quality abnormality related fault according to the fault propagation path identification result under the variable working condition.
Further, the constructing a hierarchical fault propagation network fusing the hierarchical quality information comprises:
on the basis of the quality abnormal working condition monitoring result, the hierarchical fault propagation network fusing the hierarchical quality information is constructed by taking the knowledge of the steel rolling process flow as the basis and combining different levels of information and the analysis results of the correlation and the causal relationship of the process variables and the quality abnormal related fault variables of different levels.
Further, on the basis of the quality abnormal working condition monitoring result, the step of constructing a hierarchical fault propagation network fusing the hierarchical quality information by taking the knowledge of the steel rolling process flow as the basis and combining different levels of information and the analysis results of the correlation and the causal relationship of the process variables and the quality abnormal related fault variables of different levels comprises the following steps:
on the basis of a quality abnormality working condition monitoring result, on the basis of steel rolling process flow knowledge and on the basis of different levels of information, the identification of quality abnormality related faults is realized by using methods such as a generalized reconstruction contribution diagram and a relative reconstruction contribution diagram, and a screening criterion of a quality abnormality related fault source target candidate set is constructed;
when the statistical correlation between the process variables and the quality abnormality related fault variables of different levels is easy to determine, knowledge and data are combined, direct and indirect causal relationships between the process variables and the quality abnormality related fault variables of different levels are obtained, a topological network is constructed, the constructed topological network is described in the form of a hierarchical symbol directed graph, collected data samples of different levels are used for assigning values to the hierarchical symbol directed graph, a quality abnormality related fault root source target candidate set is screened by reverse search according to constructed screening criteria, a quality abnormality related fault root source is determined by forward search, and the construction of a hierarchical fault propagation network fusing level quality information is realized;
wherein, combining knowledge and data means: and integrating the information of different levels of the whole process of the steel rolling process flow, the quality process model information and the result obtained by carrying out causal relationship analysis on the collected data samples of different levels by using a time sequence causal relationship analysis method.
Further, the hierarchy includes: the system comprises an equipment layer, a real-time control layer, a process control layer, a manufacturing execution layer, an enterprise management layer and an enterprise strategic layer.
Further, on the basis of the quality abnormal working condition monitoring result, based on the knowledge of the steel rolling process flow, and in combination with different levels of information and the analysis results of the correlation and the causal relationship of the process variables and the quality abnormal related fault variables of different levels, the construction of the hierarchical fault propagation network fusing the level quality information further comprises:
when the statistical correlation between process variables and quality anomaly related fault variables at different levels is difficult to determine, combines knowledge and data, establishes an evidence graph model taking quality abnormality related fault variables as nodes based on an evidence theory, the node state is the reliability distribution of the values of the fault variables related to the quality abnormity in the evidence theory, by analyzing process connectivity knowledge, related information of different levels and historical data of quality abnormality related fault variables, condition reliability distribution of adjacent quality abnormality related fault variables of different levels is obtained, the propagation relation of the quality abnormality related faults in an evidence graph is described by using the condition reliability distribution, and by using a conditional reliability distribution linear updating method, the node with the highest confidence in the updating result is used as a quality abnormality related fault source, and the construction of a hierarchical fault propagation network fusing hierarchical quality information is realized.
Further, the identifying a fault propagation path under a variable working condition according to the constructed hierarchical fault propagation network includes:
on the basis of the constructed hierarchical fault propagation network fusing the hierarchical quality information, analyzing the coupling relation of a multi-process quality index set aiming at the subsystem fault propagation network to realize the coarse identification of a fault propagation path;
and combining the working condition change information and the working condition transition probability model between the working procedures to realize a refined identification method of the full-flow fault propagation path under the variable working conditions.
Further, on the basis of the constructed hierarchical fault propagation network fusing the hierarchical quality information, analyzing the coupling relationship of the multi-process quality index set aiming at the subsystem fault propagation network, and realizing the coarse identification of the fault propagation path comprises the following steps:
on the basis of the constructed hierarchical fault propagation network fusing hierarchical quality information, aiming at the subsystem fault propagation network, a fault propagation network based on the dynamic Bayesian network is preliminarily constructed by using the quality process model of each subsystem of the steel rolling process flow and the dominant variable relations among process variables and between the process variables and quality indexes under different working condition modes as the basis and learning the probability dependence relation among the variables and the time variation rule thereof through the dynamic Bayesian network, so as to realize the rough identification of the fault propagation path.
Further, the method for realizing fine identification of the full-flow fault propagation path under the variable working conditions by combining the working condition change information and the working procedure variable working condition transition probability model comprises the following steps:
working condition change information is fused, a working condition probability transfer model among working procedures is constructed on the basis of a fault propagation network based on a dynamic Bayesian network which is preliminarily constructed, and an accurate dynamic Bayesian network is constructed by combining an identification result of a new working condition and an incidence relation among subsystems, so that the fine identification of a fault propagation path related to the quality abnormity of the whole flow under the variable working conditions is realized;
wherein the operating condition change information includes: process operating point changes, operating state changes, raw material changes, and production plan changes.
Further, the online diagnosis of the quality abnormality related fault according to the fault propagation path identification result under the variable working condition includes:
based on the established common and individual subspace models, respectively extracting characteristic vectors in the normal historical data of the steel rolling process flow under the common and individual subspace models by utilizing a subspace identification method to obtain reference vectors;
determining a characteristic vector of fault data identified by a fault propagation path under a variable working condition, and respectively obtaining the deviation degrees between quality abnormality related fault variables in the common subspace and the reference vector on the basis of the determined characteristic vector of the fault data;
fusing the obtained deviation degrees between the common and quality abnormality related fault variables in the individual subspace and the reference vector to obtain a fusion deviation degree of the common and individual subspaces;
determining a fusion deviation threshold, and classifying the fault grades according to the determined fusion deviation threshold and the obtained fusion deviation to realize the online diagnosis of the quality abnormality related fault.
The technical scheme provided by the embodiment of the invention has the beneficial effects that at least:
in the embodiment of the invention, a hierarchical fault propagation network fusing hierarchical quality information is constructed; identifying a fault propagation path under a variable working condition according to the constructed hierarchical fault propagation network; and performing online diagnosis on the quality abnormality related fault according to the fault propagation path identification result under the variable working condition. Therefore, the quality abnormity of the steel rolling process flow can be comprehensively diagnosed on line, and the problem of comprehensive diagnosis of the quality abnormity of the steel rolling process flow is solved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic diagram of a steel rolling process flow arrangement provided by an embodiment of the invention;
FIG. 2 is a schematic view of a steel rolling process flow of multiple levels, a whole flow, variable working conditions and three dimensions provided by an embodiment of the invention;
FIG. 3 is a schematic flow chart of a comprehensive diagnosis method for abnormal quality of a steel rolling process flow according to an embodiment of the present invention;
fig. 4 is a detailed flow diagram of a comprehensive diagnosis method for quality abnormality of a steel rolling process flow provided by an embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
The present embodiment takes a steel rolling process flow as an example. It should be noted that the comprehensive diagnosis method for quality abnormality of the present invention is not limited to the steel rolling process flow, and is also applicable to other production processes, such as chemical production processes.
FIG. 1 is a schematic diagram of a steel rolling process flow according to example 1 of the present invention. As can be seen from fig. 1, the steel rolling process flow described in this embodiment mainly includes a plurality of production processes such as heating, rough rolling, flying shears, finish rolling, laminar cooling, and coiling, and a long product processing flow mainly including a series structure is formed from raw materials to final products.
FIG. 2 is a schematic view of a steel rolling process flow of the present invention, which is "multi-level, full flow, variable working condition" and "three-dimensional". It can be seen that the corresponding integrated automation system of the steel rolling process flow has obvious levels, mainly comprises an equipment layer, a real-time control layer, a process control layer, a manufacturing execution layer, an enterprise management layer, an enterprise strategic layer and the like, and all levels have clear and mutually cooperative association; wherein the content of the first and second substances,
the device layer includes: the system comprises a heating furnace, a rough rolling unit, a flying shear, a finishing rolling unit, a laminar cooling and coiling machine, and mainly achieves the functions of main and auxiliary transmission, electric, hydraulic and pneumatic operation execution, instrument data acquisition and the like;
the real-time control layer mainly completes the sequence and logic control of the whole-line equipment according to an operation instruction issued by the process control layer, and undertakes the task of controlling the overall length and quality of the strip steel, thereby realizing the basic automation;
the process control layer is mainly used for tracking each production procedure of the hot continuous rolling whole line in real time, acquiring data and optimally setting process parameters according to an operation plan issued by the manufacturing execution layer, obtaining the optimally set parameters of various production equipment by calculating through a series of mathematical models according to actual working conditions at a proper moment, and determining the quality of a final product, particularly the quality of a head part, and greatly influencing the production sequence;
the manufacturing execution layer mainly completes the functions of production planning, production scheduling, quality management, inventory management, logistics tracking and the like, fully considers the production constraints of all the working procedures, gives consideration to different quality requirements and contract delivery date, and adopts an integrated scheduling strategy and scheduling strategy to realize material flow matching and energy flow matching;
the enterprise strategic layer and the enterprise management layer take decision management and production and general management as cores respectively, emphasize the planning of enterprises, and simultaneously take customer orders and market demands as planning sources to carry out macroscopic planning and grasp, so that various resources in the enterprises are fully utilized, and the enterprise benefit is improved.
In addition, the variety and specification of the steel rolling process flow are diversified through the customized production of the steel rolling process flow, and the production working conditions are complicated and changeable due to the uncertainty of raw materials, the difference of equipment states, external environments, process technologies and the like. Therefore, the hot rolling production process of the strip steel presents complexity in three dimensions (system level, full flow and multiple working conditions). The three-dimensional manufacturing process makes the safety and stability analysis of the system more complicated and changeable, and the abnormity of any dimension can cause the three-dimensional propagation of the fault to evolve, thereby bringing great challenges to the quality abnormity diagnosis of the steel rolling process flow.
As shown in fig. 3, an embodiment of the present invention provides a method for comprehensively diagnosing quality abnormality of a steel rolling process flow, including:
s101, constructing a hierarchical fault propagation network fusing hierarchical quality information, wherein the quality information comprises: longitudinal or transverse thickness difference, surface flatness and other information of the rolled steel product;
in this embodiment, on the basis of the monitoring result of the quality abnormal working condition, based on the knowledge of the steel rolling process flow, and in combination with the information of different levels and the analysis results of the correlations and causal relationships between the process variables of different levels and the quality abnormal related fault variables, a hierarchical fault propagation network fusing the quality information of the levels is constructed, as shown in fig. 4, which specifically includes the following steps:
a1, based on the monitoring result of the quality abnormal working condition, based on the knowledge of the steel rolling process flow, based on the production scheduling information in the manufacturing execution layer, the production process or the control loop in the process control layer and other different levels of information, realizing the identification of the quality abnormal related fault by using methods such as a generalized reconstruction contribution diagram and a relative reconstruction contribution diagram, and constructing the screening criterion of the quality abnormal related fault root source target candidate set to realize the screening of the target candidate set;
a2, when the statistical correlation between the process variable and the quality anomaly related fault variable of different levels is easily determined (easy determination means that the correlation between the process variable and the quality anomaly related fault variable is easily determined by a correlation analysis method), combines knowledge and data to obtain direct and indirect causal relations among process variables and quality abnormality related fault variables of different levels and construct a topological network, the constructed topological network is described in the form of a hierarchical symbolic directed graph, and the hierarchical symbolic directed graph is assigned with values by utilizing collected data samples (including historical production data, monitoring data, maintenance records and the like) of different levels, reverse searching is carried out according to the constructed screening criteria to screen a quality abnormality related fault root target candidate set, forward searching is carried out to determine the quality abnormality related fault root, and construction of a hierarchical fault propagation network fusing hierarchical quality information is realized;
wherein, combining knowledge and data means: the method is characterized by integrating the information of different levels of the whole process of the steel rolling process, the information of quality process models (such as bending force, rolling force, roll gap and the like) of a rough rolling unit, a finishing rolling unit and the like, and the result obtained by carrying out causal relationship analysis on the collected data samples of different levels by using a time sequence causal relationship analysis method such as Glan's Jack, transfer entropy, dynamic time warping and the like.
Wherein the hierarchy comprises: the system comprises an equipment layer, a real-time control layer, a process control layer, a manufacturing execution layer, an enterprise management layer and an enterprise strategic layer.
A3, when the statistical correlation between the process variable and the quality abnormal related fault variable of different levels is difficult to determine (difficult to determine means that the correlation between the process variable and the quality abnormal related fault variable is difficult to determine by correlation analysis method), combining the knowledge and the data, establishing an evidence graph model using the quality abnormal related fault variable as a node based on the evidence theory, wherein the node state is the reliability distribution of the value of the quality abnormal related fault variable in the evidence theory, obtaining the condition reliability distribution of the adjacent quality abnormal related fault variable nodes of different levels by analyzing the process connectivity knowledge (such as process pipeline and instrument flow chart, etc.), the related information of different levels and the historical data of the quality abnormal related fault variable, describing the propagation relationship of the quality abnormal related fault in the evidence graph by using the condition reliability distribution, and using the condition reliability distribution linear updating method, and taking the node with the highest confidence in the updating result as a quality abnormality related fault root, and realizing the construction of a hierarchical fault propagation network fusing hierarchical quality information.
S102, identifying a fault propagation path under a variable working condition according to the constructed hierarchical fault propagation network, as shown in fig. 4, specifically including the following steps:
b1, analyzing the coupling relation of the multi-process quality index set aiming at the subsystem fault propagation network on the basis of the constructed hierarchical fault propagation network fusing the hierarchical quality information, and realizing the coarse identification of the fault propagation path;
in the embodiment, on the basis of the constructed hierarchical fault propagation network fusing hierarchical quality information, the fault propagation network of the subsystem is preliminarily constructed on the basis of the coupling relation of a multi-process quality index set (specifically, the quality process model of each subsystem of the steel rolling process flow and the dominant variable relation among process variables and between the process variables and the quality indexes under different working condition modes) by using a dynamic Bayesian network to learn the probability dependence relation among the variables and the time-varying rule thereof, so as to realize the rough identification of the fault propagation path.
And B2, combining the working condition change information and the working condition transition probability model among the working procedures to realize a fine identification method of the full-flow fault propagation path under the variable working conditions.
In the embodiment, working condition change information is fused, a working procedure variable working condition probability transfer model is constructed on the basis of a preliminarily constructed fault propagation network based on the dynamic Bayesian network, and an accurate dynamic Bayesian network is constructed by combining the identification result of a new working condition and the incidence relation between subsystems (specifically, the fault propagation network under the variable working condition based on the accurate dynamic Bayesian network is constructed), so that the refined identification of the fault propagation path related to the full-flow quality abnormity under the variable working condition is realized;
wherein the operating condition change information includes: process operating point changes, operating state changes, raw material changes, and production plan changes.
S103, performing online diagnosis on the quality abnormality related fault according to the fault propagation path identification result under the variable working condition, as shown in fig. 4, specifically including the following steps:
c1, based on the established common and individual subspace models, respectively extracting the characteristic vectors in the normal historical data (the historical data obtained under the normal operation of the system) of the steel rolling process flow under the common and individual subspace models by utilizing subspace identification methods such as typical variable analysis and the like to obtain reference vectors;
c2, determining the characteristic vector of the fault data identified by the fault propagation path under the variable working condition by using methods of support vector data description, sensitivity analysis and the like, and further respectively obtaining the deviation degrees between the quality abnormality related fault variables in the common and individual subspaces and the reference vector based on the determined characteristic vector of the fault data;
c3, fusing the obtained similarity and the deviation between the quality abnormity related fault variables in the individual subspace and the reference vector by means of Bayes fusion, evidence theory, ensemble learning and the like to obtain the fusion deviation of the similarity and the individual subspace;
and C4, determining a fusion deviation threshold value by adopting a method of combining artificial intelligence methods such as a neural network and the like with a fuzzy technology, and dividing fault grades (including fault symptoms, slight faults, common faults and serious faults) according to the determined fusion deviation threshold value and the obtained fusion deviation, thereby realizing the online diagnosis of the quality abnormality related faults.
The comprehensive diagnosis method for the quality abnormity of the steel rolling process flow in the embodiment of the invention constructs a hierarchical fault propagation network fusing hierarchical quality information; identifying a fault propagation path under a variable working condition according to the constructed hierarchical fault propagation network; and carrying out online diagnosis on the quality abnormality related fault according to the identified fault propagation path under the variable working condition. Therefore, the quality abnormity of the steel rolling process flow can be comprehensively diagnosed on line, and the problem of comprehensive diagnosis of the quality abnormity of the steel rolling process flow is solved.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (9)
1. A comprehensive diagnosis method for abnormal quality of a steel rolling process flow is characterized by comprising the following steps:
constructing a hierarchical fault propagation network fusing hierarchical quality information;
identifying a fault propagation path under a variable working condition according to the constructed hierarchical fault propagation network;
and performing online diagnosis on the quality abnormality related fault according to the fault propagation path identification result under the variable working condition.
2. The method for comprehensively diagnosing quality abnormality of a steel rolling process flow according to claim 1, wherein the constructing of the hierarchical fault propagation network fusing hierarchical quality information comprises:
on the basis of the quality abnormal working condition monitoring result, the hierarchical fault propagation network fusing the hierarchical quality information is constructed by taking the knowledge of the steel rolling process flow as the basis and combining different levels of information and the analysis results of the correlation and the causal relationship of the process variables and the quality abnormal related fault variables of different levels.
3. The method for comprehensively diagnosing the quality abnormality of the steel rolling process flow according to claim 2, wherein the step of constructing the hierarchical fault propagation network fusing the hierarchical quality information based on the monitoring result of the quality abnormality working condition and the steel rolling process flow knowledge and the analysis results of the correlation and the causal relationship of the process variables and the quality abnormality related fault variables of different levels by combining the information of different levels comprises the following steps:
on the basis of a quality abnormality working condition monitoring result, on the basis of steel rolling process flow knowledge and on the basis of different levels of information, the identification of quality abnormality related faults is realized by using methods such as a generalized reconstruction contribution diagram and a relative reconstruction contribution diagram, and a screening criterion of a quality abnormality related fault source target candidate set is constructed;
when the statistical correlation between the process variables and the quality abnormality related fault variables of different levels is easy to determine, knowledge and data are combined, direct and indirect causal relationships between the process variables and the quality abnormality related fault variables of different levels are obtained, a topological network is constructed, the constructed topological network is described in the form of a hierarchical symbol directed graph, collected data samples of different levels are used for assigning values to the hierarchical symbol directed graph, a quality abnormality related fault root source target candidate set is screened by reverse search according to constructed screening criteria, a quality abnormality related fault root source is determined by forward search, and the construction of a hierarchical fault propagation network fusing level quality information is realized;
wherein, combining knowledge and data means: and integrating the information of different levels of the whole process of the steel rolling process flow, the quality process model information and the result obtained by carrying out causal relationship analysis on the collected data samples of different levels by using a time sequence causal relationship analysis method.
4. The method for comprehensively diagnosing quality abnormality of a steel rolling process flow according to claim 3, wherein the hierarchy includes: the system comprises an equipment layer, a real-time control layer, a process control layer, a manufacturing execution layer, an enterprise management layer and an enterprise strategic layer.
5. The method for comprehensively diagnosing the quality abnormality of the steel rolling process flow according to claim 3, wherein the step of constructing the hierarchical fault propagation network fusing the hierarchical quality information based on the monitoring result of the quality abnormality working condition and the steel rolling process flow knowledge and the analysis results of the correlation and the causal relationship between the process variables of different levels and the quality abnormality related fault variables further comprises the following steps:
when the statistical correlation between process variables and quality anomaly related fault variables at different levels is difficult to determine, combines knowledge and data, establishes an evidence graph model taking quality abnormality related fault variables as nodes based on an evidence theory, the node state is the reliability distribution of the values of the fault variables related to the quality abnormity in the evidence theory, by analyzing process connectivity knowledge, related information of different levels and historical data of quality abnormality related fault variables, condition reliability distribution of adjacent quality abnormality related fault variables of different levels is obtained, the propagation relation of the quality abnormality related faults in an evidence graph is described by using the condition reliability distribution, and by using a conditional reliability distribution linear updating method, the node with the highest confidence in the updating result is used as a quality abnormality related fault source, and the construction of a hierarchical fault propagation network fusing hierarchical quality information is realized.
6. The method for comprehensively diagnosing the quality abnormality of the steel rolling process flow according to claim 1, wherein the step of identifying the fault propagation path under the variable working condition according to the constructed hierarchical fault propagation network comprises the following steps:
on the basis of the constructed hierarchical fault propagation network fusing the hierarchical quality information, analyzing the coupling relation of a multi-process quality index set aiming at the subsystem fault propagation network to realize the coarse identification of a fault propagation path;
and combining the working condition change information and the working condition transition probability model between the working procedures to realize a refined identification method of the full-flow fault propagation path under the variable working conditions.
7. The method for comprehensively diagnosing quality abnormality of a steel rolling process flow according to claim 6, wherein the step of analyzing the coupling relationship of the multi-process quality index set for the subsystem fault propagation network on the basis of the constructed hierarchical fault propagation network fusing the hierarchical quality information to realize the coarse identification of the fault propagation path comprises the following steps:
on the basis of the constructed hierarchical fault propagation network fusing hierarchical quality information, aiming at the subsystem fault propagation network, a fault propagation network based on the dynamic Bayesian network is preliminarily constructed by using the quality process model of each subsystem of the steel rolling process flow and the dominant variable relations among process variables and between the process variables and quality indexes under different working condition modes as the basis and learning the probability dependence relation among the variables and the time variation rule thereof through the dynamic Bayesian network, so as to realize the rough identification of the fault propagation path.
8. The method for comprehensively diagnosing the quality abnormality of the steel rolling process flow according to claim 6, wherein the method for realizing the fine identification of the full-flow fault propagation path under the variable working condition by combining the working condition change information and the working condition transition probability model between the working procedures comprises the following steps:
working condition change information is fused, a working condition probability transfer model among working procedures is constructed on the basis of a fault propagation network based on a dynamic Bayesian network which is preliminarily constructed, and an accurate dynamic Bayesian network is constructed by combining an identification result of a new working condition and an incidence relation among subsystems, so that the fine identification of a fault propagation path related to the quality abnormity of the whole flow under the variable working conditions is realized;
wherein the operating condition change information includes: process operating point changes, operating state changes, raw material changes, and production plan changes.
9. The method for comprehensively diagnosing the quality abnormality of the steel rolling process flow according to claim 1, wherein the online diagnosis of the quality abnormality related fault according to the fault propagation path recognition result under the variable working conditions comprises:
based on the established common and individual subspace models, respectively extracting characteristic vectors in the normal historical data of the steel rolling process flow under the common and individual subspace models by utilizing a subspace identification method to obtain reference vectors;
determining a characteristic vector of fault data identified by a fault propagation path under a variable working condition, and respectively obtaining the deviation degrees between quality abnormality related fault variables in the common subspace and the reference vector on the basis of the determined characteristic vector of the fault data;
fusing the obtained deviation degrees between the common and quality abnormality related fault variables in the individual subspace and the reference vector to obtain a fusion deviation degree of the common and individual subspaces;
determining a fusion deviation threshold, and classifying the fault grades according to the determined fusion deviation threshold and the obtained fusion deviation to realize the online diagnosis of the quality abnormality related fault.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111045843.XA CN113835411B (en) | 2021-09-07 | 2021-09-07 | Comprehensive diagnosis method for quality abnormality of steel rolling process flow |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111045843.XA CN113835411B (en) | 2021-09-07 | 2021-09-07 | Comprehensive diagnosis method for quality abnormality of steel rolling process flow |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113835411A true CN113835411A (en) | 2021-12-24 |
| CN113835411B CN113835411B (en) | 2023-10-27 |
Family
ID=78958614
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111045843.XA Active CN113835411B (en) | 2021-09-07 | 2021-09-07 | Comprehensive diagnosis method for quality abnormality of steel rolling process flow |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113835411B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115780530A (en) * | 2023-01-19 | 2023-03-14 | 北京科技大学 | Steel rolling fault tracing method and device based on expert knowledge and data combined drive |
| CN116020879A (en) * | 2023-02-15 | 2023-04-28 | 北京科技大学 | Technological parameter-oriented strip steel hot continuous rolling space-time multi-scale process monitoring method and device |
| CN116719293A (en) * | 2023-08-04 | 2023-09-08 | 江苏甬金金属科技有限公司 | Production monitoring method and system for cold-rolled steel strip |
| CN116809652A (en) * | 2023-03-28 | 2023-09-29 | 材谷金带(佛山)金属复合材料有限公司 | Abnormality analysis method and system for hot rolling mill control system |
| CN117688496A (en) * | 2024-01-25 | 2024-03-12 | 南京航空航天大学 | Abnormality diagnosis method, system and equipment for satellite telemetry multidimensional time sequence data |
| CN118180161A (en) * | 2024-05-17 | 2024-06-14 | 太原理工大学 | Method and device for constructing multidimensional coupling vibration identification model of strip mill roller system |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5914875A (en) * | 1996-01-11 | 1999-06-22 | Kabushiki Kaisha Toshiba | Method and apparatus for diagnosing plant anomaly |
| CN104597892A (en) * | 2014-12-16 | 2015-05-06 | 北京航天测控技术有限公司 | Hierarchical fault diagnosis method for electronic information equipment |
| CN105955241A (en) * | 2016-06-03 | 2016-09-21 | 北京科技大学 | Quality fault locating method based on federated data driven production process |
| CN106406229A (en) * | 2016-12-20 | 2017-02-15 | 吉林大学 | Numerical control machine tool fault diagnosis method |
| CN110333709A (en) * | 2019-06-19 | 2019-10-15 | 北京科技大学 | A kind of rolling process fault diagnosis method and system |
| CN110989561A (en) * | 2019-12-26 | 2020-04-10 | 中国航空工业集团公司沈阳飞机设计研究所 | Method for constructing fault propagation model and method for determining fault propagation path |
| CN111695780A (en) * | 2020-05-18 | 2020-09-22 | 北京科技大学 | Process flow quality multi-fault autonomous detection method and system |
| CN111724126A (en) * | 2020-06-12 | 2020-09-29 | 北京科技大学顺德研究生院 | Accurate tracing method and system for process flow quality abnormity |
| CN112306036A (en) * | 2019-08-02 | 2021-02-02 | 中国石油化工股份有限公司 | Method for diagnosing operation fault of chemical process |
| US20210190882A1 (en) * | 2019-12-10 | 2021-06-24 | Wuhan University | Transformer failure identification and location diagnosis method based on multi-stage transfer learning |
| CN113311715A (en) * | 2021-07-30 | 2021-08-27 | 中国人民解放军海军工程大学 | Fault identification and tracing method for complex electromechanical equipment |
-
2021
- 2021-09-07 CN CN202111045843.XA patent/CN113835411B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5914875A (en) * | 1996-01-11 | 1999-06-22 | Kabushiki Kaisha Toshiba | Method and apparatus for diagnosing plant anomaly |
| CN104597892A (en) * | 2014-12-16 | 2015-05-06 | 北京航天测控技术有限公司 | Hierarchical fault diagnosis method for electronic information equipment |
| CN105955241A (en) * | 2016-06-03 | 2016-09-21 | 北京科技大学 | Quality fault locating method based on federated data driven production process |
| CN106406229A (en) * | 2016-12-20 | 2017-02-15 | 吉林大学 | Numerical control machine tool fault diagnosis method |
| CN110333709A (en) * | 2019-06-19 | 2019-10-15 | 北京科技大学 | A kind of rolling process fault diagnosis method and system |
| CN112306036A (en) * | 2019-08-02 | 2021-02-02 | 中国石油化工股份有限公司 | Method for diagnosing operation fault of chemical process |
| US20210190882A1 (en) * | 2019-12-10 | 2021-06-24 | Wuhan University | Transformer failure identification and location diagnosis method based on multi-stage transfer learning |
| CN110989561A (en) * | 2019-12-26 | 2020-04-10 | 中国航空工业集团公司沈阳飞机设计研究所 | Method for constructing fault propagation model and method for determining fault propagation path |
| CN111695780A (en) * | 2020-05-18 | 2020-09-22 | 北京科技大学 | Process flow quality multi-fault autonomous detection method and system |
| CN111724126A (en) * | 2020-06-12 | 2020-09-29 | 北京科技大学顺德研究生院 | Accurate tracing method and system for process flow quality abnormity |
| CN113311715A (en) * | 2021-07-30 | 2021-08-27 | 中国人民解放军海军工程大学 | Fault identification and tracing method for complex electromechanical equipment |
Non-Patent Citations (3)
| Title |
|---|
| 彭开香等: "面向系统层级的复杂工业过程全息故障诊断", 化工学报, vol. 70, no. 2, pages 590 - 598 * |
| 王梦园;张雄;马亮;彭开香;: "基于因果拓扑图的工业过程故障诊断", 山东大学学报(工学版), vol. 47, no. 05 * |
| 袁海斌;安茂春;: "分布式层次化故障诊断模型和诊断策略的研究", 微计算机信息, no. 19 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115780530A (en) * | 2023-01-19 | 2023-03-14 | 北京科技大学 | Steel rolling fault tracing method and device based on expert knowledge and data combined drive |
| CN116020879A (en) * | 2023-02-15 | 2023-04-28 | 北京科技大学 | Technological parameter-oriented strip steel hot continuous rolling space-time multi-scale process monitoring method and device |
| CN116809652A (en) * | 2023-03-28 | 2023-09-29 | 材谷金带(佛山)金属复合材料有限公司 | Abnormality analysis method and system for hot rolling mill control system |
| CN116809652B (en) * | 2023-03-28 | 2024-04-26 | 材谷金带(佛山)金属复合材料有限公司 | Abnormality analysis method and system for hot rolling mill control system |
| CN116719293A (en) * | 2023-08-04 | 2023-09-08 | 江苏甬金金属科技有限公司 | Production monitoring method and system for cold-rolled steel strip |
| CN116719293B (en) * | 2023-08-04 | 2023-11-17 | 江苏甬金金属科技有限公司 | Production monitoring method and system for cold-rolled steel strip |
| CN117688496A (en) * | 2024-01-25 | 2024-03-12 | 南京航空航天大学 | Abnormality diagnosis method, system and equipment for satellite telemetry multidimensional time sequence data |
| CN117688496B (en) * | 2024-01-25 | 2024-04-16 | 南京航空航天大学 | Abnormality diagnosis method, system and equipment for satellite telemetry multidimensional time sequence data |
| CN118180161A (en) * | 2024-05-17 | 2024-06-14 | 太原理工大学 | Method and device for constructing multidimensional coupling vibration identification model of strip mill roller system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113835411B (en) | 2023-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113835411A (en) | Comprehensive diagnosis method for abnormal quality of steel rolling process flow | |
| Diez-Olivan et al. | Data fusion and machine learning for industrial prognosis: Trends and perspectives towards Industry 4.0 | |
| Saldivar et al. | Self-organizing tool for smart design with predictive customer needs and wants to realize Industry 4.0 | |
| García | Multi-agent system for anomaly detection in Industry 4.0 using Machine Learning techniques | |
| Goyal et al. | Optimization of condition-based maintenance using soft computing | |
| Bastos et al. | A maintenance prediction system using data mining techniques | |
| CN109902931A (en) | A kind of polymorphic manufacture system operation risk modeling method based on operation data fusion | |
| Chen et al. | Application of data-driven models to predictive maintenance: Bearing wear prediction at TATA steel | |
| CN116522088B (en) | Nuclear power plant operation data analysis method and system based on machine learning | |
| Navinchandran et al. | Studies to predict maintenance time duration and important factors from maintenance workorder data | |
| Becherer et al. | Intelligent choice of machine learning methods for predictive maintenance of intelligent machines | |
| Tripathy et al. | Explaining Anomalies in Industrial Multivariate Time-series Data with the help of eXplainable AI | |
| Yıldız et al. | Integrating preventive and predictive maintenance policies with system dynamics: A decision table approach | |
| Lieber et al. | Sustainable interlinked manufacturing processes through real-time quality prediction | |
| Varaprasad et al. | Stock Price Prediction using Machine Learning | |
| Merkt | Predictive models for maintenance optimization: an analytical literature survey of industrial maintenance strategies | |
| Ejlali et al. | Developing hybrid machine learning models to assign health score to railcar fleets for optimal decision making | |
| Sultana et al. | Economic design of exponentially weighted moving average chart with variable sampling interval at fixed times scheme incorporating Taguchi loss function | |
| Salehi | Bayesian-based predictive analytics for manufacturing performance metrics in the era of industry 4.0 | |
| Bahar et al. | Predictive Maintenance System Integrated with Periodic Maintenance: Machine Learning and Classical Approaches | |
| CN113828638B (en) | Steel rolling process flow composite fault tracing method | |
| Dundulienė et al. | Trade risk identification and prediction using experts’ knowledge and machine learning | |
| Lee et al. | Design of decision support system for yield management in semiconductor industry: application to artificial intelligence | |
| Turkoglu et al. | Application of data mining in failure estimation of cold forging machines: An industrial research | |
| Sharma et al. | Industry 4.0 Technologies for Smart Manufacturing: A Systematic Review of Machine Learning Methods for Predictive Maintenance |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |