CN111724126B - Accurate tracing method and system for quality abnormality of process flow - Google Patents
Accurate tracing method and system for quality abnormality of process flow Download PDFInfo
- Publication number
- CN111724126B CN111724126B CN202010535022.3A CN202010535022A CN111724126B CN 111724126 B CN111724126 B CN 111724126B CN 202010535022 A CN202010535022 A CN 202010535022A CN 111724126 B CN111724126 B CN 111724126B
- Authority
- CN
- China
- Prior art keywords
- quality
- flow
- information
- subsystem
- process flow
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 351
- 230000008569 process Effects 0.000 title claims abstract description 269
- 230000005856 abnormality Effects 0.000 title claims abstract description 27
- 238000012544 monitoring process Methods 0.000 claims abstract description 109
- 238000013507 mapping Methods 0.000 claims abstract description 33
- 238000000354 decomposition reaction Methods 0.000 claims description 28
- 230000004927 fusion Effects 0.000 claims description 18
- 238000004458 analytical method Methods 0.000 claims description 16
- 238000010276 construction Methods 0.000 claims description 14
- 238000010586 diagram Methods 0.000 claims description 13
- 230000007246 mechanism Effects 0.000 claims description 12
- 238000010206 sensitivity analysis Methods 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 10
- 230000035945 sensitivity Effects 0.000 claims description 10
- 238000007500 overflow downdraw method Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 238000010219 correlation analysis Methods 0.000 claims description 5
- 230000008447 perception Effects 0.000 claims description 5
- 230000003068 static effect Effects 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000010801 machine learning Methods 0.000 claims description 4
- 238000005065 mining Methods 0.000 claims description 4
- 230000009467 reduction Effects 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 238000004519 manufacturing process Methods 0.000 abstract description 33
- 239000010410 layer Substances 0.000 description 21
- 238000005096 rolling process Methods 0.000 description 15
- 229910000831 Steel Inorganic materials 0.000 description 12
- 239000010959 steel Substances 0.000 description 12
- 230000015654 memory Effects 0.000 description 8
- 239000000047 product Substances 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 6
- 238000003860 storage Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 230000001364 causal effect Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 239000002994 raw material Substances 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 2
- 239000012467 final product Substances 0.000 description 2
- 239000007789 gas Substances 0.000 description 2
- 238000010438 heat treatment Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 239000007791 liquid phase Substances 0.000 description 2
- 239000007790 solid phase Substances 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000004566 building material Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 239000013256 coordination polymer Substances 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000011229 interlayer Substances 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000004886 process control Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
-
- 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
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- General Factory Administration (AREA)
Abstract
The invention discloses a process flow quality abnormity accurate tracing method and a system, wherein the method comprises the following steps: decomposing the process flow under the constraint of mass flow, and dividing the whole flow into a plurality of subsystems; constructing a quality index set and a quality index inverse mapping model which are sensitive to the quality fluctuation of the whole process in each subsystem; based on the quality index inverse mapping model, combining system level information to construct a subsystem quality anomaly monitoring model integrating the level information, so as to realize quality anomaly monitoring of each subsystem; fusing quality anomaly monitoring information of different subsystems to realize full process quality anomaly monitoring of the process flow fusing local associated information; and constructing a quality-related fault propagation network, and based on the quality-related fault propagation network, realizing quality-related fault tracing according to the whole process quality anomaly monitoring information of the process flow. The invention can effectively monitor and judge the quality abnormality of the production process in time and accurately, and has important practical application and popularization values.
Description
Technical Field
The invention relates to the technical field of control and monitoring of a process production process, in particular to a process flow quality abnormality accurate tracing method and a process flow quality abnormality accurate tracing system.
Background
The industrial processes of steel, nonferrous, petrochemical industry, building materials and the like are important components of manufacturing industry, are important prop industries for national economy and social development, and are important supporting forces for the continuous growth of the national economy. The method has the advantages of long flow, multi-variable coupling in the working procedure, quality inheritance among the working procedures, multiple system levels and quality indexes, and the production process relates to the coexistence of gas, liquid and solid phases of complex physicochemical reactions.
The strip steel hot continuous rolling production process mainly comprises a plurality of production processes of heating, rough rolling, flying shears, finish rolling, laminar flow, coiling and the like, and a product processing long process taking a serial structure as a main body is formed from raw materials to a final product; meanwhile, the corresponding comprehensive automation system has obvious level and mainly comprises an equipment layer, a real-time control layer, a process control layer, a manufacturing execution layer and the like. The steps are separated and associated in a coordinated manner, and raw material components, equipment states, process parameters, product quality and the like of the steps cannot be sensed in real time or comprehensively, so that the safety and stability analysis of the steps are complex and changeable, any one or more links are abnormal, fault propagation and even evolution can be caused, and the enterprise is stopped and maintained due to the return of a quality dissatisfied user, so that the economic benefit of the enterprise is influenced.
The continuous uninterrupted operation of the production process leads any unit or subsystem to fail, so that the quality of the whole production line and the final product can be influenced, the raw material components, the equipment state, the process parameters, the product quality and the like of the production process cannot be sensed in real time or comprehensively, and the long-term stable, high-quality and high-efficiency operation of the production process is difficult to ensure. Therefore, in the face of the increasingly strong international market competition, ensuring the high-quality and high-efficiency operation of the industrial process through reasonable quality anomaly monitoring and tracing technology becomes an important content in the sustainable development of the national manufacturing industry.
Disclosure of Invention
The invention provides a process flow quality abnormality accurate tracing method and a process flow quality abnormality accurate tracing system, which aim to solve the problem of quality abnormality accurate tracing of flow industry and realize the technical effect of effectively monitoring and judging the quality abnormality in the process of flow industry in time and accurately.
In order to solve the technical problems, the invention provides the following technical scheme:
in one aspect, the invention provides a process flow quality anomaly accurate tracing method, which comprises the following steps:
based on process knowledge and historical data of the process flow to be detected, decomposing the process flow to be detected under the constraint of mass flow, and dividing the whole flow into a plurality of subsystems;
Constructing a quality index set sensitive to the full-flow quality fluctuation in each subsystem, and analyzing the full-flow process and quality data of the process flow to be detected to construct a quality index inverse mapping model;
based on the quality index inverse mapping model, combining system level information of the process flow to be detected, constructing a subsystem quality anomaly monitoring model integrating the level information, and realizing quality anomaly monitoring of each subsystem;
fusing quality anomaly monitoring information of different subsystems of the process flow to be detected to realize overall process quality anomaly monitoring of the process flow fused with local associated information;
and constructing a quality-related fault propagation network of the process flow to be detected, and based on the quality-related fault propagation network, realizing quality-related fault tracing according to the quality anomaly monitoring information of the whole process flow.
Further, the process flow to be detected is decomposed under the mass flow constraint based on the process knowledge and the history data of the process flow to be detected, and the whole flow is divided into a plurality of subsystems, including:
using the process knowledge and history data of the process flow to be detected, taking control association, reaction association, position association, type association, structure association and function association as reference sequences, analyzing association and propagation relations of quality information among process flow variables, and determining variable candidate sets of the whole-flow quality flow information;
Determining subsystem dominant variables based on the variable candidate set;
measuring the correlation between the residual variable and the dominant variable in the variable candidate set by using a statistical correlation analysis method, and determining an auxiliary variable; the auxiliary variables are gathered around the dominant variables with the correlation strength larger than a preset threshold value by taking the dominant variables as the center, so that the whole-flow preliminary decomposition is realized;
based on the hierarchical information of the process flow, the rationality of the preliminary decomposition result is evaluated, the preset threshold is adjusted, the full-flow fine decomposition under the constraint of the mass flow is realized, and the full flow is divided into a plurality of subsystems.
Further, the process knowledge includes hierarchical information of the process flow to be detected, process mechanisms, and expert knowledge.
Further, the determining a subsystem dominant variable based on the variable candidate set includes:
determining dominant variables of a subsystem with definite process mechanism by using priori knowledge; for a subsystem with an ambiguous process mechanism, a partial least squares or causality analysis method is utilized to determine the dominant variable.
Further, the constructing a quality index set sensitive to the quality fluctuation of the whole process in each subsystem, analyzing the whole process and quality data of the process to be detected to construct a quality index inverse mapping model, including:
Analyzing the sensitivity of each subsystem process variable to the typical quality fluctuation of the whole process;
based on the sensitivity analysis result, a quality index set sensitive to the quality fluctuation of the whole process in each subsystem is constructed, and the optimization of the quality index set is realized by combining with a process model of the process to be detected;
based on the optimized quality index set, determining the qualitative relation between the hierarchical characteristics of the quality information and the quality information among the hierarchies by utilizing process knowledge, and mining the quantitative quality information shared among all hierarchies and specific in the hierarchies by adopting a dynamic time window combined multivariate data modeling method to realize the analysis of the whole flow process and the quality data;
and constructing a high-dimensional multi-quality index inverse mapping model characterized by information depth perception by using a method combining semi-supervised learning and multi-variable regression.
Further, analyzing the sensitivity of each subsystem process variable to typical quality fluctuations of the overall process, including:
the sensitivity of each subsystem process variable to the typical quality fluctuation of the whole flow is analyzed by adopting variance decomposition, sensitivity analysis based on partial derivatives and sensitivity analysis method based on connection weight.
Further, the establishing a subsystem quality anomaly monitoring model based on the quality index inverse mapping model and combining system level information of the process flow to be detected includes:
Based on the quality index inverse mapping model, synchronizing the multi-level information scale of the process flow to be detected by using a scale synchronization method, and estimating the quality index of a preset type by using a soft sensing model;
for multi-level quality information, constructing a high-dimensional data tensor by the quality information in a system hierarchical structure, and extracting inter-level common characteristics and intra-level individual characteristic information of a process flow by adopting a preset tensor decomposition technology;
aiming at the common characteristics among layers and the personalized characteristic information in the layers, a variable-fraction reasoning combined deep neural network method is adopted to construct a probability monitoring model related to subsystem quality;
and carrying out decision fusion on probability monitoring models of different levels of the subsystem by a preset decision fusion method, and constructing a subsystem quality anomaly monitoring model fused with level information.
Further, the fusing of the quality anomaly monitoring information of different subsystems of the process flow to be detected to realize the process flow full-flow quality anomaly monitoring of fusing the local association information includes:
acquiring a static association and a dynamic cooperative relationship among all subsystems;
compressing related information of the associated subsystem by using a distributed computing framework and adopting a multivariable dimension reduction method, and transmitting the compressed related information to the target subsystem;
Fusing the quality anomaly monitoring information of the associated subsystem and the quality anomaly monitoring information of the target subsystem to realize the quality anomaly collaborative monitoring among the subsystems;
and fusing quality anomaly monitoring information of different subsystems through a preset information fusion and machine learning method, so as to realize the whole process quality anomaly monitoring of the process flow fusing the local associated information.
Further, a quality-related fault propagation network of the process flow to be detected is constructed, quality-related fault tracing is realized based on the quality-related fault propagation network and based on the monitoring information of the quality abnormality of the whole process flow, and the method comprises the following steps:
the state association influence relation among all subsystems is constructed, so that the directed graph construction among all-flow-layer-oriented subsystems is realized, and the propagation and influence process of quality related faults among all the subsystems is described;
building a structural element and fault mode association model in each subsystem, and realizing quality related fault and logic network diagram building of a subsystem layer;
the constructed directed graph facing the full flow layer and the quality related fault of the subsystem layer are associated and integrated with the logic network graph, so that the quality related fault propagation network construction integrating the multi-source data and knowledge is realized;
Determining a subsystem in which quality related faults occur according to the quality anomaly monitoring information of the whole process flow;
identifying quality related faults and constructing a screening criterion of an effective target candidate set;
and in the determined subsystem, based on the quality-related fault propagation network, the reason of the quality-related fault is found out by utilizing element reasoning in the effective target candidate set, so that hierarchical quality-related fault tracing is realized.
On the other hand, the invention also provides a process flow quality abnormity accurate tracing system, which comprises:
the depth extraction and modeling module of the quality information is used for decomposing the technological process to be detected under the constraint of the quality flow based on the technological knowledge and the historical data of the technological process to be detected, and dividing the whole process into a plurality of subsystems; constructing a quality index set sensitive to the full-flow quality fluctuation in each subsystem, and analyzing the full-flow process and quality data of the process flow to be detected to construct a quality index inverse mapping model;
the quality anomaly monitoring module is used for constructing a subsystem quality anomaly monitoring model integrating the level information based on the quality index inverse mapping model and combining the system level information of the process flow to be detected, so as to realize quality anomaly monitoring of each subsystem; fusing quality anomaly monitoring information of different subsystems to realize full process quality anomaly monitoring of the process flow fusing local associated information;
The hierarchical quality abnormality accurate tracing module is used for constructing a quality related fault propagation network of a process flow to be detected, and realizing quality related fault tracing according to the whole process flow quality abnormality monitoring information based on the quality related fault propagation network.
In yet another aspect, the present invention also provides an electronic device including a processor and a memory; wherein the memory stores at least one instruction that is loaded and executed by the processor to implement the above-described method.
In yet another aspect, the present invention also provides a computer readable storage medium having at least one instruction stored therein, the instruction being loaded and executed by a processor to implement the above method.
The technical scheme provided by the invention has the beneficial effects that at least:
the process flow quality anomaly accurate tracing method comprehensively considers the level information of the process flow industrial automation system, and realizes the deep extraction and modeling of the quality information in the process flow, the quality anomaly monitoring of the fusion level information and the hierarchical quality anomaly accurate tracing. On the basis of methods such as multi-quality index inverse mapping construction, layering modeling, multi-source data and knowledge fusion and the like, a quality anomaly accurate tracing method suitable for flow industry is provided, the key challenge problem in modern flow industry quality anomaly monitoring and tracing is aimed at, practical and effective solution strategies are explored, and important practical application and popularization values are achieved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a process flow quality anomaly accurate traceability method provided by a first embodiment of the present invention;
FIG. 2 is a schematic diagram of the process arrangement in the hot continuous rolling production process of strip steel;
fig. 3 is a schematic diagram of an implementation route of a process quality anomaly accurate tracing method according to a second embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the embodiments of the present invention will be described in further detail with reference to the accompanying drawings.
First embodiment
The embodiment provides a process flow quality abnormity accurate tracing method, the execution flow of which is shown in fig. 1, comprising the following steps:
s101, decomposing the technological process to be detected under the constraint of mass flow based on the technological knowledge and the historical data of the technological process to be detected, and dividing the whole process into a plurality of subsystems;
In this embodiment, the step S101 specifically includes the following steps:
using the process knowledge and history data of the process flow to be detected, taking control association, reaction association, position association, type association, structure association and function association as reference sequences, analyzing association and propagation relations of quality information among process flow variables, and determining variable candidate sets of the whole-flow quality flow information; determining subsystem dominant variables based on the variable candidate set; measuring the correlation between the residual variable and the dominant variable in the variable candidate set by using a statistical correlation analysis method, and determining an auxiliary variable; the auxiliary variables are gathered around the dominant variables with the correlation strength larger than a preset threshold value by taking the dominant variables as the center, so that the whole-flow preliminary decomposition is realized; based on the hierarchical information of the process flow, the rationality of the preliminary decomposition result is evaluated, a preset threshold is adjusted, the complete flow fine decomposition under the constraint of the mass flow is realized, and the complete flow is divided into a plurality of subsystems.
The process knowledge used in this embodiment includes hierarchical information, process mechanism and expert knowledge of the process flow to be detected. Determining subsystem dominant variables based on the variable candidate set, comprising: determining dominant variables of a subsystem with definite process mechanism by using priori knowledge; for a subsystem with an ambiguous process mechanism, a partial least squares or causality analysis method is utilized to determine the dominant variable.
S102, a quality index set sensitive to the quality fluctuation of the whole process in each subsystem is constructed, and the whole process and quality data of the process flow to be detected are analyzed to construct a quality index inverse mapping model;
in this embodiment, the step S102 specifically includes the following steps:
analyzing the sensitivity of each subsystem process variable to the typical quality fluctuation of the whole process; based on the sensitivity analysis result, a quality index set sensitive to the quality fluctuation of the whole process in each subsystem is constructed, and the optimization of the quality index set is realized by combining with a process model of the process to be detected; based on the optimized quality index set, determining the qualitative relation between the hierarchical characteristics of the quality information and the quality information among the hierarchies by utilizing process knowledge, and mining the quantitative quality information shared among all hierarchies and specific in the hierarchies by adopting a dynamic time window combined multivariate data modeling method to realize the analysis of the whole flow process and the quality data; and constructing a high-dimensional multi-quality index inverse mapping model characterized by information depth perception by using a method combining semi-supervised learning and multi-variable regression.
Wherein analyzing the sensitivity of each subsystem process variable to typical quality fluctuations of the overall process comprises: and analyzing the sensitivity of the subsystem process variables to the typical quality fluctuation of the whole flow by adopting sensitivity analysis methods such as variance decomposition, sensitivity analysis based on partial derivatives, sensitivity analysis based on connection weights and the like.
S103, based on the quality index inverse mapping model, combining system level information of the process flow to be detected, constructing a subsystem quality anomaly monitoring model integrating the level information, and realizing quality anomaly monitoring of each subsystem;
in this embodiment, the step S103 specifically includes the following steps:
based on the quality index inverse mapping model, synchronizing the multi-level information scale of the process flow to be detected by using a scale synchronization method, and estimating the quality index of a preset type by using a soft sensing model; for multi-level quality information, constructing a high-dimensional data tensor by the quality information in a system hierarchical structure, and extracting inter-level common characteristics and intra-level individual characteristic information of a process flow by adopting a preset tensor decomposition technology; aiming at the common characteristics among layers and the personalized characteristic information in the layers, a variable-fraction reasoning combined deep neural network method is adopted to construct a probability monitoring model related to subsystem quality; and carrying out decision fusion on probability monitoring models of different levels of the subsystem by a preset decision fusion method, and constructing a subsystem quality anomaly monitoring model fused with level information.
S104, fusing quality anomaly monitoring information of different subsystems of the process flow to be detected to realize overall process quality anomaly monitoring of the process flow fused with local associated information;
In this embodiment, the step S104 specifically includes the following steps:
acquiring a static association and a dynamic cooperative relationship among all subsystems; compressing related information of the associated subsystem by using a distributed computing framework and adopting a multivariable dimension reduction method, and transmitting the compressed related information to the target subsystem; fusing the quality anomaly monitoring information of the associated subsystem and the quality anomaly monitoring information of the target subsystem to realize the quality anomaly collaborative monitoring among the subsystems; and fusing quality anomaly monitoring information of different subsystems through a preset information fusion and machine learning method, so as to realize the whole process quality anomaly monitoring of the process flow fusing the local associated information.
S105, constructing a quality-related fault propagation network of the process flow to be detected, and realizing quality-related fault tracing according to the process flow total-flow quality anomaly monitoring information based on the quality-related fault propagation network.
In this embodiment, the step S105 specifically includes the following steps:
the state association influence relation among all subsystems is constructed, so that the directed graph construction among all-flow-layer-oriented subsystems is realized, and the propagation and influence process of quality related faults among all the subsystems is described; building a structural element and fault mode association model in each subsystem, and realizing quality related fault and logic network diagram building of a subsystem layer; the constructed directed graph facing the full flow layer and the quality related fault of the subsystem layer are associated and integrated with the logic network graph, so that the quality related fault propagation network construction integrating the multi-source data and knowledge is realized; determining a subsystem in which quality related faults occur according to the quality anomaly monitoring information of the whole process flow; identifying quality related faults and constructing a screening criterion of an effective target candidate set; and in the determined subsystem, based on the quality-related fault propagation network, the reason of the quality-related fault is found out by utilizing element reasoning in the effective target candidate set, so that hierarchical quality-related fault tracing is realized.
The process flow quality anomaly accurate tracing method comprehensively considers the level information of the process flow industrial automation system, and realizes the deep extraction and modeling of the quality information in the process flow, the quality anomaly monitoring of the fusion level information and the hierarchical quality anomaly accurate tracing. On the basis of methods such as multi-quality index inverse mapping construction, layering modeling, multi-source data and knowledge fusion and the like, a quality anomaly accurate tracing method suitable for flow industry is provided, the key challenge problem in modern flow industry quality anomaly monitoring and tracing is aimed at, practical and effective solution strategies are explored, and important practical application and popularization values are achieved.
Second embodiment
Referring to fig. 2 and fig. 3, this example further illustrates an implementation of the process flow quality anomaly accurate tracing method of the present invention, taking a hot continuous rolling production process of a strip steel driven by data combination as an example; of course, it can be understood that the process flow quality anomaly accurate tracing method is not limited to the hot continuous rolling process of the strip steel, and is also suitable for other data combined driving production processes, such as an automobile part production process.
As shown in fig. 2, in the hot continuous rolling production process of the strip steel, the flow is long, the multi-variable coupling is carried out in the working procedure, the quality inheritance is carried out among the working procedures, the system level and the quality index are more, and the production process involves the coexistence of gas, liquid and solid phases of complex physical and chemical reactions. The production line comprises a plurality of production procedures such as heating, rough rolling, flying shears, finish rolling, laminar flow, coiling and the like. The continuous uninterrupted operation of the complex production process can cause that any unit or subsystem is possibly spread and evolved on different levels through mass flow, energy flow and information flow, two or more faults are generated simultaneously or sequentially, the stable operation of the production process and the quality of products are seriously influenced, and the early monitoring and judgment of the quality abnormality can be ensured by the process flow quality abnormality accurate traceability method, so that the smooth progress of the production process and the quality of the products can be ensured.
As shown in fig. 3, on the basis of the actual engineering application driving of hot continuous rolling of the strip steel shown in fig. 2, the process flow quality abnormality accurate tracing method of the embodiment comprises the following steps:
s1, deep extraction and modeling of quality information;
specifically, in this embodiment, the implementation procedure of the above steps is specifically as follows:
S101, analyzing the structural characteristics of the whole process of the hot continuous rolling process of the strip steel through knowledge and data fusion technology on the basis of fully knowing the technological knowledge of the whole process of the hot continuous rolling process of the strip steel, researching a dominant-auxiliary variable selection method of each process under the constraint of mass flow, dividing the whole process into a plurality of subsystems, and realizing the preliminary decomposition of the whole process.
Specifically, the implementation process of S101 is as follows:
firstly, using process knowledge (hierarchical information, process mechanism, expert knowledge and the like) and taking control association, reaction association, position association, type association, structure association and function association as reference sequences, analyzing association and propagation relations of quality information among industrial process variables, and determining a whole-flow quality flow information variable candidate set;
then, determining dominant variables by using priori knowledge aiming at the subsystem with clear mechanism; determining subsystem dominant variables by utilizing Partial Least Squares (PLS), causal relation analysis and the like aiming at a subsystem with an ambiguous mechanism;
finally, statistical correlation analysis methods such as Mutual Information (MI), maximum Correlation Coefficient (MCC), maximum Information Coefficient (MIC) and the like are adopted to construct main and auxiliary variable indexes, and the correlation between the residual variable and the main variable is measured and divided into a far domain and a near domain, wherein the correlation between the near domain variables is strong, and the correlation between the far domain variables is weak. And collecting auxiliary variables which are near-domain variables around the dominant variables with stronger correlation with the near-domain variables by taking the dominant variables as centers, so as to realize the full-flow preliminary decomposition.
S102, comprehensively considering different level information such as production scheduling, tracking areas, process model setting and the like, analyzing the influence of the information on the whole process decomposition, and taking the influence as constraint guidance to realize the whole process refined decomposition of the hot continuous rolling of the strip steel under the constraint of mass flow.
Specifically, the implementation process of S102 is as follows: and (3) analyzing the variables repeatedly covered by a plurality of local spaces, comprehensively considering different-level information such as production scheduling, tracking areas and process model setting and variable information not contained by any local space, taking the information as constraint conditions, evaluating the rationality of the whole-flow preliminary decomposition result, further adjusting the threshold value, and realizing the whole-flow refined decomposition under the mass flow constraint.
S103, on the basis of the total flow fine decomposition under the mass flow constraint, a quality index set sensitive to the typical quality fluctuation of the total flow in each subsystem is constructed, and the optimization of the quality index set is realized by combining a process model, process knowledge and a data driving method.
Specifically, the implementation process of S103 is as follows: on the basis of refined decomposition, the sensitivity of each subsystem process variable to the typical quality fluctuation of the whole flow is analyzed by adopting methods such as variance decomposition, sensitivity analysis based on partial derivatives, sensitivity analysis based on connection weights and the like, and the optimization of a quality index set is realized by combining a process model.
S104, performing full-dimensional and multi-level intelligent analysis on the whole flow process and the quality data by using the internal relation and the change rule between the technological parameters and the product quality, and constructing high-dimensional and multi-quality index inverse mapping characterized by information depth perception.
Specifically, the implementation process of S104 is as follows:
firstly, determining the hierarchical characteristics of quality information and the qualitative relation of the quality information among the hierarchies by utilizing process knowledge, and mining quantitative quality information shared among all hierarchies and peculiar in the hierarchies by adopting a dynamic time window combined multivariate data modeling method to realize the full-dimension and multi-hierarchy intelligent analysis of the full-flow process and the quality data;
and then, constructing a high-dimensional multi-quality index inverse mapping model which is characterized by information depth perception by utilizing a method combining semi-supervised learning and multi-variable regression, and laying a foundation for research on aspects such as quality anomaly monitoring of fusion level information.
S2, monitoring quality abnormality of the fusion level information;
specifically, in this embodiment, the implementation procedure of the above steps is specifically as follows:
s201, on the basis of fully considering the problems of process knowledge, scheduling information, process data characteristics, interaction relation of quality information among different levels and the like of all subsystems of different levels, a unified characterization method of the quality information of multiple levels is researched, and the inter-level commonality-individuation feature extraction of the quality features is realized.
Specifically, the implementation process of S201 is as follows:
firstly, fully considering different level information such as production scheduling information, production process or control loop and the like on the basis of full-flow fine decomposition under the constraint of mass flow, optimization of a quality index set and construction of an inverse mapping model, synchronizing multi-level information scales by using a scale synchronization method, and estimating key quality indexes which are difficult to measure or not measured in real time by using a soft sensing model;
then, for multi-level quality information, the quality information is constructed into a high-dimensional data tensor by a system hierarchical structure, and tensor decomposition technologies based on CP decomposition, tucker decomposition and the like are adopted to extract inter-level common characteristics and intra-level personalized characteristic information.
S202, constructing a subsystem quality anomaly monitoring model integrating the level information so as to mine more local process information and reduce the complexity of quality on-line monitoring.
Specifically, the implementation process of S202 is as follows:
firstly, respectively aiming at the inter-layer commonality characteristic and intra-layer individuation characteristic information, constructing a subsystem quality-related probability monitoring model by adopting a variational reasoning combined deep neural network method;
and then, carrying out decision fusion on monitoring models of different levels of the subsystem by using decision fusion methods such as statistical decision, bayesian information fusion and the like, and constructing a subsystem quality abnormity monitoring model fused with level information.
S203, based on fully considering the static association and dynamic cooperative relationship among the subsystems, the quality anomaly monitoring information of different subsystems is fused by information fusion and machine learning methods such as Bayesian fusion, ensemble learning and the like, so that the full-flow quality anomaly monitoring of fused local association information is realized.
Specifically, in the present embodiment, the implementation of S203 described above is as follows:
firstly, aiming at the mutual association and coupling characteristics among subsystems, adopting methods such as typical correlation analysis (CCA), correlation projection analysis and the like to analyze the static association relationship among the subsystems, and adopting a method of combining dynamic CCA (DCCA) with TE to analyze the dynamic cooperative relationship among the subsystems;
secondly, aiming at the problem of mutual information coordination among subsystems, a distributed computing framework is utilized, a multivariable dimension reduction method is adopted to compress related information of the related subsystems, and the compressed information is transmitted to a target subsystem;
then, on the basis of a subsystem quality anomaly monitoring model integrating hierarchical information, integrating the auxiliary information of the associated system and the monitoring information of the target subsystem to realize the quality distributed collaborative monitoring among the subsystems;
finally, because the model setting information, the scheduling information and the data characteristics of different subsystems are different, the designed monitoring model, statistics and control limits are different, and on the basis of the distributed collaborative monitoring of the quality anomalies among the subsystems, the monitoring statistics, control limits and other information of the different subsystems are fused by a Bayesian fusion method, so that the overall process quality anomaly monitoring of fused local associated information is realized.
S3, hierarchical quality abnormality accurate tracing;
specifically, in this embodiment, the implementation procedure of the above steps is specifically as follows:
s301, on the basis of comprehensive analysis of production scheduling, tracking areas, process model setting information and the like of each subsystem, a knowledge and causal relationship analysis combined method is utilized to respectively construct a quality related fault propagation network from two layers of a whole flow and a subsystem according to different characteristics of quality related fault propagation inside the subsystem and among the subsystems.
Specifically, the implementation process of S301 is as follows:
firstly, combining different-level information such as full-flow production scheduling, tracking areas and process model setting and semantic expression information of different subsystem knowledge, and constructing state association influence relations among all subsystems by combining results obtained by a Granger equivalent sequence causal relation analysis method, so as to realize directed graph construction among all-flow-layer-oriented subsystems and be used for describing propagation and influence processes of quality related faults among all subsystems;
and then, at the subsystem level, carrying out fault mode analysis on each subsystem and structural elements thereof to determine complex association relations between the structural elements and fault modes and between the fault modes, constructing a structural element-fault mode association model in each subsystem, and realizing quality-related fault-logic network diagram construction at the subsystem level.
S302, performing association integration between the two layers of the whole flow and the subsystem through the equipment function state variables to realize quality related fault propagation network construction integrating multi-source data and knowledge.
Specifically, the implementation process of S302 is as follows:
and carrying out association integration on the constructed directed graph facing the full flow layer and the quality related fault-logic network graph of the subsystem layer, so as to realize the construction of the quality related fault propagation network fusing the multi-source data and knowledge.
S303, on the basis of the total flow of the integrated level information and the subsystem quality anomaly monitoring result, the constructed total flow layer and the constructed subsystem layer quality related fault propagation network are utilized, and the effective target candidate set is constructed, so that the layering quality related fault is precisely traced, and information support is provided for quickly and accurately locking the occurrence range of quality problems, finding the occurrence reason of the quality problems and maintaining decisions.
Specifically, the implementation procedure of S303 is as follows:
firstly, positioning quality related faults in a specific subsystem on the basis of the whole flow of top-down fusion level information and subsystem quality abnormality monitoring results, and reducing the fault searching range;
Then, utilizing methods such as generalized reconstruction contribution Graph (GRBC) relative reconstruction contribution graph (rRBC) and the like to realize identification of quality related faults, constructing screening criteria of effective target candidate sets, and improving efficiency of quality related fault tracing;
finally, the reason of the quality-related fault is found out by element reasoning in the effective target candidate set through the quality-related fault logic network diagram for a specific subsystem, so that layering quality-related fault accurate tracing is realized, and information support is provided for quickly and accurately locking the occurrence range of the quality problem, finding out the reason of the quality problem and maintaining decision.
In summary, the method for precisely tracing the abnormal quality of the strip steel hot continuous rolling process focuses on the problems of quality safety and stability of products in the process of strip steel hot continuous rolling and other processes, completely accords with the multi-level and large-scale industrial characteristics and development directions of the process industry, has important scientific significance for preventing the quality of the products from being reduced and exerting the running potential of the process to the maximum extent, and has important theoretical value and wide engineering application prospect.
Third embodiment
The embodiment provides a process flow quality abnormality accurate traceability system, which comprises the following modules:
The depth extraction and modeling module of the quality information is used for decomposing the technological process to be detected under the constraint of the quality flow based on the technological knowledge and the historical data of the technological process to be detected, and dividing the whole process into a plurality of subsystems; constructing a quality index set sensitive to the full-flow quality fluctuation in each subsystem, and analyzing the full-flow process and quality data of the process flow to be detected to construct a quality index inverse mapping model;
the quality anomaly monitoring module is used for constructing a subsystem quality anomaly monitoring model integrating the level information based on the quality index inverse mapping model and combining the system level information of the process flow to be detected, so as to realize quality anomaly monitoring of each subsystem; fusing quality anomaly monitoring information of different subsystems to realize full process quality anomaly monitoring of the process flow fusing local associated information;
the hierarchical quality abnormality accurate tracing module is used for constructing a quality related fault propagation network of a process flow to be detected, and realizing quality related fault tracing according to the whole process flow quality abnormality monitoring information based on the quality related fault propagation network.
The process flow quality abnormality accurate tracing system of the embodiment corresponds to the process flow quality abnormality accurate tracing method of the embodiment; the functions realized by the functional modules in the process flow quality abnormality accurate tracing system in the embodiment are in one-to-one correspondence with the flow steps in the process flow quality abnormality accurate tracing method in the embodiment; therefore, the description is omitted here.
Fourth embodiment
The embodiment provides an electronic device, which comprises a processor and a memory; wherein the memory stores at least one instruction that is loaded and executed by the processor to implement the method of the above embodiment.
The electronic device may vary considerably in configuration or performance and may include one or more processors (central processing units, CPU) and one or more memories, wherein the memories store at least one instruction that is loaded by the processors and performs the following steps:
step one, decomposing the technological process to be detected under the constraint of mass flow based on the technological knowledge and historical data of the technological process to be detected, and dividing the whole process into a plurality of subsystems; constructing a quality index set sensitive to the full-flow quality fluctuation in each subsystem, and analyzing the full-flow process and quality data of the process flow to be detected to construct a quality index inverse mapping model;
Step two, based on the quality index inverse mapping model, combining system level information of the process flow to be detected, constructing a subsystem quality anomaly monitoring model integrating the level information, and realizing quality anomaly monitoring of each subsystem; fusing quality anomaly monitoring information of different subsystems of the process flow to be detected to realize overall process quality anomaly monitoring of the process flow fused with local associated information;
and thirdly, constructing a quality-related fault propagation network of the process flow, and based on the quality-related fault propagation network, realizing quality-related fault tracing according to the quality anomaly monitoring information of the whole process flow.
Fifth embodiment
The present embodiment provides a computer-readable storage medium having stored therein at least one instruction that is loaded and executed by a processor to implement the above-described method. The computer readable storage medium may be, among other things, ROM, random Access Memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, etc. The instructions stored therein may be loaded by a processor in the terminal and perform the steps of:
step one, decomposing the technological process to be detected under the constraint of mass flow based on the technological knowledge and historical data of the technological process to be detected, and dividing the whole process into a plurality of subsystems; constructing a quality index set sensitive to the full-flow quality fluctuation in each subsystem, and analyzing the full-flow process and quality data of the process flow to be detected to construct a quality index inverse mapping model;
Step two, based on the quality index inverse mapping model, combining system level information of the process flow to be detected, constructing a subsystem quality anomaly monitoring model integrating the level information, and realizing quality anomaly monitoring of each subsystem; fusing quality anomaly monitoring information of different subsystems of the process flow to be detected to realize overall process quality anomaly monitoring of the process flow fused with local associated information;
and thirdly, constructing a quality-related fault propagation network of the process flow, and based on the quality-related fault propagation network, realizing quality-related fault tracing according to the quality anomaly monitoring information of the whole process flow.
Furthermore, it should be noted that the present invention can be provided as a method, an apparatus, or a computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the invention may take the form of a computer program product on one or more computer-usable storage media having computer-usable program code embodied therein.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.
It is finally pointed out that the above description of the preferred embodiments of the invention, it being understood that although preferred embodiments of the invention have been described, it will be obvious to those skilled in the art that, once the basic inventive concepts of the invention are known, several modifications and adaptations can be made without departing from the principles of the invention, and these modifications and adaptations are intended to be within the scope of the invention. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the embodiments of the invention.
Claims (5)
1. A process flow quality abnormity accurate tracing method is characterized by comprising the following steps:
based on process knowledge and historical data of the process flow to be detected, decomposing the process flow to be detected under the constraint of mass flow, and dividing the whole flow into a plurality of subsystems;
constructing a quality index set sensitive to the full-flow quality fluctuation in each subsystem, and analyzing the full-flow process and quality data of the process flow to be detected to construct a quality index inverse mapping model;
based on the quality index inverse mapping model, combining system level information of the process flow to be detected, constructing a subsystem quality anomaly monitoring model integrating the level information, and realizing quality anomaly monitoring of each subsystem;
Fusing quality anomaly monitoring information of different subsystems of the process flow to be detected to realize overall process quality anomaly monitoring of the process flow fused with local associated information;
constructing a quality-related fault propagation network of a process flow to be detected, and based on the quality-related fault propagation network, realizing quality-related fault tracing according to the quality anomaly monitoring information of the whole process flow;
the process flow to be detected is decomposed under the constraint of mass flow based on the process knowledge and the historical data of the process flow to be detected, and the whole flow is divided into a plurality of subsystems, including:
using the process knowledge and history data of the process flow to be detected, taking control association, reaction association, position association, type association, structure association and function association as reference sequences, analyzing association and propagation relations of quality information among process flow variables, and determining variable candidate sets of the whole-flow quality flow information;
determining subsystem dominant variables based on the variable candidate set;
measuring the correlation between the residual variable and the dominant variable in the variable candidate set by using a statistical correlation analysis method, and determining an auxiliary variable; the auxiliary variables are gathered around the dominant variables with the correlation strength larger than a preset threshold value by taking the dominant variables as the center, so that the whole-flow preliminary decomposition is realized;
Based on the hierarchical information of the process flow, evaluating the rationality of the preliminary decomposition result, adjusting the preset threshold value, realizing the complete flow fine decomposition under the constraint of the mass flow, and dividing the complete flow into a plurality of subsystems;
the construction of the quality index set sensitive to the whole process quality fluctuation in each subsystem, and the analysis of the whole process and the quality data of the technological process to be detected to construct the quality index inverse mapping model comprises the following steps:
analyzing the sensitivity of each subsystem process variable to the typical quality fluctuation of the whole process;
based on the sensitivity analysis result, a quality index set sensitive to the quality fluctuation of the whole process in each subsystem is constructed, and the optimization of the quality index set is realized by combining with a process model of the process to be detected;
based on the optimized quality index set, determining the qualitative relation between the hierarchical characteristics of the quality information and the quality information among the hierarchies by utilizing process knowledge, and mining the quantitative quality information shared among all hierarchies and specific in the hierarchies by adopting a dynamic time window combined multivariate data modeling method to realize the analysis of the whole flow process and the quality data;
constructing a high-dimensional multi-quality index inverse mapping model characterized by information depth perception by using a method combining semi-supervised learning and multi-variable regression;
The subsystem quality anomaly monitoring model integrating the level information is constructed based on the quality index inverse mapping model and combined with the system level information of the process flow to be detected, and comprises the following steps:
based on the quality index inverse mapping model, synchronizing the multi-level information scale of the process flow to be detected by using a scale synchronization method, and estimating the quality index of a preset type by using a soft sensing model;
for multi-level quality information, constructing a high-dimensional data tensor by the quality information in a system hierarchical structure, and extracting inter-level common characteristics and intra-level individual characteristic information of a process flow by adopting a preset tensor decomposition technology;
aiming at the common characteristics among layers and the personalized characteristic information in the layers, a variable-fraction reasoning combined deep neural network method is adopted to construct a probability monitoring model related to subsystem quality;
carrying out decision fusion on probability monitoring models of different levels of the subsystem by a preset decision fusion method, and constructing a subsystem quality anomaly monitoring model fused with level information;
the process flow total process quality anomaly monitoring method for fusing the quality anomaly monitoring information of different subsystems of the process flow to be detected, which comprises the following steps:
Acquiring a static association and a dynamic cooperative relationship among all subsystems;
compressing related information of the associated subsystem by using a distributed computing framework and adopting a multivariable dimension reduction method, and transmitting the compressed related information to the target subsystem;
fusing the quality anomaly monitoring information of the associated subsystem and the quality anomaly monitoring information of the target subsystem to realize the quality anomaly collaborative monitoring among the subsystems;
fusing quality anomaly monitoring information of different subsystems through a preset information fusion and machine learning method to realize the whole process quality anomaly monitoring of the process flow fusing the local associated information;
constructing a quality-related fault propagation network of a process flow to be detected, and based on the quality-related fault propagation network and based on process flow overall flow quality anomaly monitoring information, realizing quality-related fault tracing, comprising:
the state association influence relation among all subsystems is constructed, so that the directed graph construction among all-flow-layer-oriented subsystems is realized, and the propagation and influence process of quality related faults among all the subsystems is described;
building a structural element and fault mode association model in each subsystem, and realizing quality related fault and logic network diagram building of a subsystem layer;
The constructed directed graph facing the full flow layer and the quality related fault of the subsystem layer are associated and integrated with the logic network graph, so that the quality related fault propagation network construction integrating the multi-source data and knowledge is realized;
determining a subsystem in which quality related faults occur according to the quality anomaly monitoring information of the whole process flow;
identifying quality related faults and constructing a screening criterion of an effective target candidate set;
and in the determined subsystem, based on the quality-related fault propagation network, the reason of the quality-related fault is found out by utilizing element reasoning in the effective target candidate set, so that hierarchical quality-related fault tracing is realized.
2. The process flow quality anomaly accurate traceability method of claim 1, wherein the process knowledge comprises hierarchical information, process mechanisms and expert knowledge of the process flow to be detected.
3. The process flow quality anomaly accurate traceability method of claim 2, wherein said determining subsystem dominant variables based on said variable candidate set comprises:
determining dominant variables of a subsystem with definite process mechanism by using priori knowledge; for a subsystem with an ambiguous process mechanism, a partial least squares or causality analysis method is utilized to determine the dominant variable.
4. The process flow quality anomaly accurate traceability method of claim 1, wherein said analyzing the sensitivity of each subsystem process variable to a full flow typical quality fluctuation comprises:
the sensitivity of each subsystem process variable to the typical quality fluctuation of the whole flow is analyzed by adopting variance decomposition, sensitivity analysis based on partial derivatives and sensitivity analysis method based on connection weight.
5. A process flow quality anomaly accurate tracing system for implementing the process flow quality anomaly accurate tracing method of any one of claims 1 to 4, comprising:
the depth extraction and modeling module of the quality information is used for decomposing the technological process to be detected under the constraint of the quality flow based on the technological knowledge and the historical data of the technological process to be detected, and dividing the whole process into a plurality of subsystems; constructing a quality index set sensitive to the full-flow quality fluctuation in each subsystem, and analyzing the full-flow process and quality data of the process flow to be detected to construct a quality index inverse mapping model;
the quality anomaly monitoring module is used for constructing a subsystem quality anomaly monitoring model integrating the level information based on the quality index inverse mapping model and combining the system level information of the process flow to be detected, so as to realize quality anomaly monitoring of each subsystem; fusing quality anomaly monitoring information of different subsystems to realize full process quality anomaly monitoring of the process flow fusing local associated information;
The hierarchical quality abnormality accurate tracing module is used for constructing a quality related fault propagation network of a process flow to be detected, and realizing quality related fault tracing according to the whole process flow quality abnormality monitoring information based on the quality related fault propagation network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010535022.3A CN111724126B (en) | 2020-06-12 | 2020-06-12 | Accurate tracing method and system for quality abnormality of process flow |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010535022.3A CN111724126B (en) | 2020-06-12 | 2020-06-12 | Accurate tracing method and system for quality abnormality of process flow |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111724126A CN111724126A (en) | 2020-09-29 |
| CN111724126B true CN111724126B (en) | 2024-03-12 |
Family
ID=72568099
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010535022.3A Active CN111724126B (en) | 2020-06-12 | 2020-06-12 | Accurate tracing method and system for quality abnormality of process flow |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111724126B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113835411B (en) * | 2021-09-07 | 2023-10-27 | 北京科技大学顺德研究生院 | Comprehensive diagnosis method for quality abnormality of steel rolling process flow |
| CN113828638B (en) * | 2021-09-07 | 2022-06-03 | 北京科技大学 | Steel rolling process flow composite fault tracing method |
| CN113792090B (en) * | 2021-09-17 | 2024-08-30 | 中冶赛迪信息技术(重庆)有限公司 | Steel rolling data monitoring method, system, medium and electronic terminal |
| CN114324368B (en) * | 2022-03-10 | 2022-07-29 | 广东祥利科技有限公司 | Modified plastic production detection method and system based on mechanical vision detection |
| CN114580562A (en) * | 2022-03-16 | 2022-06-03 | 北京珞安科技有限责任公司 | Abnormal data detection method and device based on process flow |
| CN115496424B (en) * | 2022-11-09 | 2023-04-28 | 广东能创科技有限公司 | Safety management method and system for methanol hydrogen production process |
| CN116020879B (en) * | 2023-02-15 | 2023-06-16 | 北京科技大学 | Technological parameter-oriented strip steel hot continuous rolling space-time multi-scale process monitoring method and device |
| CN115907569B (en) * | 2023-03-02 | 2023-05-09 | 昆山市恒达精密机械工业有限公司 | Plastic product safety monitoring method and system based on Internet of things |
| CN115953397B (en) * | 2023-03-13 | 2023-06-02 | 山东金帝精密机械科技股份有限公司 | Method and equipment for monitoring process preparation flow of conical bearing retainer |
| CN116756531B (en) * | 2023-08-23 | 2023-10-13 | 北京科技大学 | Method and device for monitoring hot continuous rolling process of strip steel for model interaction linkage |
| CN116992308B (en) * | 2023-09-25 | 2024-01-16 | 中国科学院自动化研究所 | Data and knowledge fusion process fluctuation analysis and optimization method and device |
| CN117171896A (en) * | 2023-11-03 | 2023-12-05 | 鲜一汽配（南通）有限公司 | Method and system for optimizing machining of input shaft of automobile steering gear |
| CN117689269B (en) * | 2024-01-30 | 2024-05-14 | 深圳市微克科技股份有限公司 | Fault tracing method for intelligent watch processing stage |
| CN118297332A (en) * | 2024-04-10 | 2024-07-05 | 北京科技大学 | Method and device for monitoring key performance index abnormality in float glass production process |
| CN118245782B (en) * | 2024-05-28 | 2024-08-06 | 浙江省白马湖实验室有限公司 | Double-layer-level fault diagnosis method for green methanol production process |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7480640B1 (en) * | 2003-12-16 | 2009-01-20 | Quantum Leap Research, Inc. | Automated method and system for generating models from data |
| WO2019233047A1 (en) * | 2018-06-07 | 2019-12-12 | 国电南瑞科技股份有限公司 | Power grid dispatching-based operation and maintenance method |
| WO2020046261A1 (en) * | 2018-08-27 | 2020-03-05 | Siemens Aktiengesellschaft | Systematic prognostic analysis with dynamic causal model |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6907416B2 (en) * | 2001-06-04 | 2005-06-14 | Honeywell International Inc. | Adaptive knowledge management system for vehicle trend monitoring, health management and preventive maintenance |
| US8725667B2 (en) * | 2008-03-08 | 2014-05-13 | Tokyo Electron Limited | Method and system for detection of tool performance degradation and mismatch |
| CN104850893A (en) * | 2014-12-01 | 2015-08-19 | 厦门易联创质检技术服务有限公司 | Quality perception information management method and system based on three dimensional evaluation and time domain tracing |
| EP3500897A1 (en) * | 2016-08-29 | 2019-06-26 | Siemens Aktiengesellschaft | Method and system for anomaly detection in a manufacturing system |
| US10635094B2 (en) * | 2016-09-16 | 2020-04-28 | Siemens Aktiengesellschaft | Failure models for embedded analytics and diagnostic/prognostic reasoning |
-
2020
- 2020-06-12 CN CN202010535022.3A patent/CN111724126B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7480640B1 (en) * | 2003-12-16 | 2009-01-20 | Quantum Leap Research, Inc. | Automated method and system for generating models from data |
| WO2019233047A1 (en) * | 2018-06-07 | 2019-12-12 | 国电南瑞科技股份有限公司 | Power grid dispatching-based operation and maintenance method |
| WO2020046261A1 (en) * | 2018-08-27 | 2020-03-05 | Siemens Aktiengesellschaft | Systematic prognostic analysis with dynamic causal model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111724126A (en) | 2020-09-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111724126B (en) | Accurate tracing method and system for quality abnormality of process flow | |
| CN111695780B (en) | Process flow quality multi-fault autonomous detection method and system | |
| CN109358582B (en) | Big-data-based large-mining-height fully-mechanized coal mining face multi-equipment cooperative operation control method | |
| CN113305645B (en) | Numerical control machine tool cutter residual life prediction method based on hybrid neural model | |
| CN117389236B (en) | Propylene oxide production process optimization method and system | |
| You et al. | Applications of artificial intelligence for coal mine gas risk assessment | |
| Antomarioni et al. | A predictive association rule-based maintenance policy to minimize the probability of breakages: application to an oil refinery | |
| CN113036913B (en) | Method and device for monitoring state of comprehensive energy equipment | |
| CN112101439B (en) | High-speed wire quality defect diagnosis and tracing method based on distributed Bayesian network | |
| CN117393076B (en) | Intelligent monitoring method and system for heat-resistant epoxy resin production process | |
| CN116224947B (en) | Deep learning industrial control system capable of actively integrating domain rule knowledge | |
| CN114741844A (en) | Static equipment corrosion monitoring method, device, equipment and storage medium | |
| CN105653791A (en) | Data mining based corrosion failure prediction system for on-service oil pipe column | |
| Li et al. | Integrated predictive maintenance approach for multistate manufacturing system considering geometric and non-geometric defects of products | |
| Ak et al. | Data analytics and uncertainty quantification for energy prediction in manufacturing | |
| Han et al. | An intelligent moving window sparse principal component analysis-based case based reasoning for fault diagnosis: Case of the drilling process | |
| Dong et al. | Hierarchical causal graph-based fault root cause diagnosis and propagation path identification for complex industrial process monitoring | |
| CN114548494A (en) | Visual cost data prediction intelligent analysis system | |
| CN113537706A (en) | Oil field production increasing measure optimization method based on intelligent integration | |
| CN116611004B (en) | Coal mine safety disaster prediction model training method, early warning method and early warning system | |
| CN117408519A (en) | Chemical reaction process risk early warning method and device based on deep learning algorithm | |
| CN117312972A (en) | Method for identifying health state of scraper conveyor speed reducer | |
| Wang et al. | MLA‐TCN: Multioutput Prediction of Dam Displacement Based on Temporal Convolutional Network with Attention Mechanism | |
| Zhao et al. | Intelligent anti-corrosion expert system based on big data analysis | |
| Han et al. | Fault subspace decomposition and reconstruction theory based online fault prognosis |
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 |