CN103279123A - Method of monitoring faults in sections for intermittent control system - Google Patents

Method of monitoring faults in sections for intermittent control system Download PDF

Info

Publication number
CN103279123A
CN103279123A CN2013101882936A CN201310188293A CN103279123A CN 103279123 A CN103279123 A CN 103279123A CN 2013101882936 A CN2013101882936 A CN 2013101882936A CN 201310188293 A CN201310188293 A CN 201310188293A CN 103279123 A CN103279123 A CN 103279123A
Authority
CN
China
Prior art keywords
fuzzy
cpv
data
control system
carried out
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
Application number
CN2013101882936A
Other languages
Chinese (zh)
Other versions
CN103279123B (en
Inventor
张晓丹
张新民
王国柱
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenyang University of Chemical Technology
Original Assignee
Shenyang University of Chemical Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenyang University of Chemical Technology filed Critical Shenyang University of Chemical Technology
Priority to CN201310188293.6A priority Critical patent/CN103279123B/en
Publication of CN103279123A publication Critical patent/CN103279123A/en
Application granted granted Critical
Publication of CN103279123B publication Critical patent/CN103279123B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Testing And Monitoring For Control Systems (AREA)

Abstract

The invention discloses a method of monitoring faults in sections for an intermittent control system and relates to a fault monitoring method. Firstly, a plurality of batches of collected intermittent process data are standardized in a way of expanding variables, and a data matrix on each sampling time is subjected to principal component analysis; secondly, a fuzzy C-means clustering is a fuzzy clustering analysis method which is suitable for soft partition and is generated through combining a fuzzy set theory and a k-means clustering; and thirdly, after segmentation is finished, an improved MPCA (Multiway Principal Component Analysis) model with a time varying principal element covariance on the basis of expanding variables is established on each subphase, then when on-line monitoring is carried out, which phase a new batch of data belongs to is judged, whether the data exceeds the fault monitoring control limit or not is calculated and judged, if so, a fault occurs, and the fault monitoring in sections ends. According to the invention, process multi-phase partition is more accurate, misinformation and missing report rates in monitoring are reduced, and the practical application and operability are strong.

Description

Intermittent control system is carried out the method that segment failure monitors
Technical field
The present invention relates to a kind of fault monitoring method, particularly relate to a kind of method of Intermittent control system being carried out the segment failure supervision.
Background technology
Batch production process is one of production run significant during modern industry is produced.It has characteristics such as multistage property, dynamic and reaction complexity, has obtained using widely in industries such as pharmaceutical production, petrochemical complex, semiconductor machining and biological products.Because the complicacy of batch process operating conditions and reaction mechanism, be difficult to set up reliability and the high mechanism model of accuracy, therefore, setting up reasonable, an effective multivariate statistics monitoring system realizes in-service monitoring and the fault diagnosis of batch process are had important practical significance.
Multidirectional pivot analysis (MPCA) is a kind of method that batch process monitors, but often people only set up a MPCA model in whole batch process reaction time, have ignored the multistage property in the batch production, easily cause higher rate of false alarm.At the stage characteristic of batch process, traditional segmented mode all belongs to hard division, hard sorting technique, with being divided in certain class of the data strictness of each sub-period, has either-or character.And non-linear, the dynamic of batch process statistics and uncertainty are difficult to reach above requirement, and adjacent sub-period is a progressive formation, and transient data has intermediary.Therefore need to consider that each transient data belongs to the degree of membership problem of each classification, carry out soft division, thereby the stage that could realize batch process better divides.During traditional MPCA modeling simultaneously, the one, the restriction of when being subjected to needing batch track synchronization process and in-service monitoring by batch expansion following data track being estimated; The 2nd, can not remove the main non-linear restriction in batch process when launching to be subjected to standardization by variable.
Summary of the invention
The object of the present invention is to provide a kind of method of Intermittent control system being carried out the segment failure supervision, this method is used pivot analysis technology, CPV (1), fuzzy C-means clustering method and is improved the MPCA technology batch process fault is carried out the segmentation supervision, it is more accurate to make the process multistage divide, wrong report, rate of failing to report in monitoring have been reduced, practical application and workable.
The objective of the invention is to be achieved through the following technical solutions:
Intermittent control system is carried out the method that segment failure monitors, described method comprises following process:
At first, the multiple batches of batch process data that collect are launched to carry out standardization by variable, each sampling instant data matrix is carried out pivot analysis, obtain its corresponding CPV (1), in fragmentation procedure, if the dynamic relationship between data remains unchanged always, value CPV(1) will keep constant substantially; Secondly, fuzzy C-means clustering is the Fuzzy Cluster Analysis method that is fit to carry out soft division that puts forward in conjunction with fuzzy set theory and k-mean cluster; With reflecting the input of CPV (1) index of Data Dynamic characteristic transformation as fuzzy C-means clustering, realize the multistage soft division of batch process according to fuzzy maximum membership grade principle identification adjacent phases; Again, after segmentation is finished, each sub set up based on variable launch have the time become the improvement MPCA model of pivot covariance, determine under the normal condition
Figure 2013101882936100002DEST_PATH_IMAGE001
With
Figure 74297DEST_PATH_IMAGE002
The fault monitoring of statistic control limit then, judges at first during in-service monitoring that it belongs to that stage, calculate its correspondence
Figure 2013101882936100002DEST_PATH_IMAGE003
With
Figure 11160DEST_PATH_IMAGE004
Monitor statistic, judge whether it exceeds fault monitoring control limit, if exceed then show that fault takes place, institute carries out segment failure and monitors and finish.
Described method of Intermittent control system being carried out the segment failure supervision, its described CPV (1) changes the maximum information that changes in index CPV (1) acquisition procedure for corresponding dynamic perfromance, the size of its value, correlation information and dynamic relationship between the response data variable.
Described method of Intermittent control system being carried out the segment failure supervision, the index CPV (1) that described dynamic perfromance changes is, multiple batches of batch process data, data matrix by each sampling instant after the variable expansion carries out pivot analysis, obtains the index CPV (1) that the representative data dynamic perfromance changes.
Described method of Intermittent control system being carried out the segment failure supervision, the index CPV (1) that described Data Dynamic characteristic changes carries out fuzzy C-means clustering to each sampling instant, realizes soft division of multistage according to maximum fuzzy membership principle identification adjacent phases.
The described method that Intermittent control system is carried out the segment failure supervision, fuzzy clustering realizes multistage soft division to described fuzzy C-means clustering to CPV (1), each stage after the division sets up the improvement MPCA that becomes covariance when having and monitors model.
Described method of Intermittent control system being carried out the segment failure supervision, described multistage soft division realizes soft division of multistage according to maximum fuzzy membership principle identification adjacent phases, each sub after dividing is set up improved MPCA monitor model, realize that the batch process multistage monitors.
Advantage of the present invention and effect are:
1. selected cluster index CPV (1) process that more can capture changes maximum procedural information, correlativity and dynamic perfromance between simultaneously more can representative data.
2. fuzzy C-means clustering has overcome the defective of traditional hard division, has considered that the batch process data belong to the degree of membership problem of each classification, the stage is divided more meet batch process reality.
3. improve the MPCA modeling in conjunction with batch launching main non-linear and variable in batch process of eliminating advantage such as batch track synchronization when launching not need to estimate the data that batch unreacted is intact and not requiring modeling, simultaneously during in-service monitoring than fixedly covariance calculating of traditional MPCA
Figure 2013101882936100002DEST_PATH_IMAGE005
Statistic sensitivity is higher, more can embody the dynamic movement between data.
4. the segmentation of batch process segmentation modeling monitors the multistage characteristic that has taken into full account batch process, has reduced wrong report, rate of failing to report in monitoring.
5. the present invention comes from batch process reality, and deep theoretical foundation is arranged, practical application and workable.
Description of drawings
Fig. 1 is based on CPV(1) the fuzzy membership change curve;
Fig. 2 criticizes procedure processing method for improved MPCA.
Embodiment
The present invention is described in detail below in conjunction with embodiment.
The data matrix of each sampling instant after the inventive method is launched by variable multiple batches of batch process data carries out pivot analysis, calculate its corresponding dynamic perfromance and change index CPV (1), fuzzy clustering realizes multistage soft division to CPV (1) to use the fuzzy C-means clustering algorithm; Each stage after dividing is set up the improvement MPCA that becomes covariance when having monitor model.The data matrix of each sampling instant after multiple batches of batch process data are launched by variable carries out pivot analysis, obtains the index CPV (1) that the representative data dynamic perfromance changes.Data Dynamic characteristic transformation index CPV (1) to each sampling instant carries out fuzzy C-means clustering, realizes soft division of multistage according to maximum fuzzy membership principle identification adjacent phases.Each sub after dividing is set up improved MPCA monitor model, realize that the batch process multistage monitors.
The multiple batches of batch process data that collect are launched to carry out standardization by variable, each sampling instant data matrix is carried out pivot analysis obtain its corresponding CPV (1), CPV (1) has caught the maximum information that changes in the process on the one hand; On the other hand, but the size of its value also correlation information between the response data variable and dynamic relationship to a certain extent, in fragmentation procedure, if the dynamic relationship between data remains unchanged always, CPV(1 so) value will keep constant substantially.(Fuzzy C-Means Clustering FCM) is exactly the Fuzzy Cluster Analysis method that is fit to carry out soft division that puts forward in conjunction with fuzzy set theory and k-mean cluster to fuzzy C-means clustering.It has overcome hard cluster and will be divided into object strictness to be identified in certain class and either-or character, has set up the uncertainty description of sample to classification.The present invention will reflect that CPV (1) index of Data Dynamic characteristic transformation as the input of fuzzy C-means clustering, realizes the multistage soft division of batch process according to fuzzy maximum membership grade principle identification adjacent phases.After segmentation is finished, each sub set up based on variable launch have the time become the improvement MPCA model of pivot covariance, determine under the normal condition
Figure 357828DEST_PATH_IMAGE001
With
Figure 81939DEST_PATH_IMAGE002
The fault monitoring control limit of statistic.Judge at first during in-service monitoring that it belongs to that stage, calculate its correspondence
Figure 172255DEST_PATH_IMAGE003
With
Figure 166887DEST_PATH_IMAGE004
Monitor statistic, judge whether it exceeds fault monitoring control limit, if exceed then show that fault takes place.
Multistage divides: suppose three-dimensional batch process data
Figure 418877DEST_PATH_IMAGE006
, wherein,
Figure 2013101882936100002DEST_PATH_IMAGE007
For batch,
Figure 630284DEST_PATH_IMAGE008
Be variable number,
Figure 2013101882936100002DEST_PATH_IMAGE009
Be hits.Expand into by variable
Figure 71761DEST_PATH_IMAGE010
, the column criterionization of going forward side by side; Data matrix to each sampling instant carries out the pivot decomposition by formula (1) then, obtains the eigenwert of descending sort
Figure 2013101882936100002DEST_PATH_IMAGE011
, calculate CPV (1) by accumulative total variance contribution ratio formula (2) at last.Wherein m is the variable number,
Figure 747330DEST_PATH_IMAGE012
For sub matrix,
Figure 2013101882936100002DEST_PATH_IMAGE013
Be load matrix,
Figure 170221DEST_PATH_IMAGE014
Residual matrix.
Figure 2013101882936100002DEST_PATH_IMAGE015
(1)
Figure 432707DEST_PATH_IMAGE016
(2)
With the CPV (1) of each sampling instant of the obtaining input sample as fuzzy C-means clustering, according to the reaction mechanism of batch production process batch process is divided
Figure 2013101882936100002DEST_PATH_IMAGE017
Section obtains the degree of membership matrix
Figure 182269DEST_PATH_IMAGE018
As follows:
Figure 2013101882936100002DEST_PATH_IMAGE019
,
Figure 885914DEST_PATH_IMAGE020
Refer to the
Figure 2013101882936100002DEST_PATH_IMAGE021
Individual sample prescription be under the jurisdiction of with
Figure 56870DEST_PATH_IMAGE022
Centered by classification The degree of membership value, matrix
Figure 541072DEST_PATH_IMAGE018
It is one
Figure 776881DEST_PATH_IMAGE024
Matrix, wherein the element of each row shows that corresponding object is under the jurisdiction of The degree of membership of each class in the individual classification; According to the change curve (as Fig. 1 (data come from penicillin fermentation process)) of fuzzy membership, pick out according to maximum fuzzy membership principle
Figure 67103DEST_PATH_IMAGE017
The individual stage.
Multistage monitors: the stage is divided the back for the data block of each sub Set up and improve MPCA model (as Fig. 2).At first, with data by batch expanding into
Figure 38601DEST_PATH_IMAGE026
(for being less than in each batch
Figure 2013101882936100002DEST_PATH_IMAGE027
The equal zero padding of row), it is carried out being reduced to matrix again after the standardization Form, the new matrix after will reducing then
Figure 51611DEST_PATH_IMAGE025
Expand into matrix by variable
Figure 3518DEST_PATH_IMAGE028
, at last to matrix
Figure 586946DEST_PATH_IMAGE028
Carry out pivot analysis, obtain matrix of loadings
Figure DEST_PATH_IMAGE029
With sub matrix
Figure 228141DEST_PATH_IMAGE030
, will get sub matrix
Figure 275732DEST_PATH_IMAGE030
Resolve into
Figure DEST_PATH_IMAGE031
Individual
Figure 460857DEST_PATH_IMAGE032
Matrix calculates its correspondence
Figure DEST_PATH_IMAGE033
Covariance matrix constantly
In Multivariable Statistical Process Control, commonly used Statistic and
Figure 611718DEST_PATH_IMAGE002
Statistic is carried out fault monitoring to process.
Figure 154695DEST_PATH_IMAGE001
Statistic is obeyed
Figure DEST_PATH_IMAGE035
Distribute, therefore The control limit of statistic can be calculated by following formula:
Figure 491184DEST_PATH_IMAGE036
（3）
In the formula
Figure DEST_PATH_IMAGE037
Be modeling batch,
Figure 247788DEST_PATH_IMAGE038
Be that the pivot number is (when segmentation modeling
Figure 210933DEST_PATH_IMAGE038
Be the pivot number of each son section), Be to have
Figure 193933DEST_PATH_IMAGE038
With
Figure 906586DEST_PATH_IMAGE040
Individual degree of freedom, confidence level are
Figure DEST_PATH_IMAGE041
Figure 517696DEST_PATH_IMAGE035
The distribution critical value.
Figure 418787DEST_PATH_IMAGE002
Statistic is obeyed weighting
Figure 748137DEST_PATH_IMAGE042
Distribute, so the control of statistic limit can be calculated by following formula:
Figure DEST_PATH_IMAGE043
（4）
In the formula With
Figure DEST_PATH_IMAGE045
When being modeling respectively
Figure 537287DEST_PATH_IMAGE033
Variance and mean value constantly, Be that degree of freedom is
Figure DEST_PATH_IMAGE047
, degree of confidence is The distribution critical value.
During the in-service monitoring new lot, we judge that at first it belongs to that stage, with modeling hourly value and standard deviation it are carried out standardization; Calculate new lot Constantly
Figure 552386DEST_PATH_IMAGE001
Statistic:
Figure 840016DEST_PATH_IMAGE048
（5）
In the formula
Figure DEST_PATH_IMAGE049
For waiting to monitor new lot
Figure 256085DEST_PATH_IMAGE033
Score vector constantly,
Figure 961873DEST_PATH_IMAGE050
Be
Figure 297040DEST_PATH_IMAGE033
Covariance constantly (time become covariance) matrix;
New lot Constantly
Figure 472818DEST_PATH_IMAGE002
Statistic:
Figure DEST_PATH_IMAGE051
（6）
In the formula
Figure 846161DEST_PATH_IMAGE052
For waiting to monitor new lot
Figure 663813DEST_PATH_IMAGE033
Residual matrix constantly;
Check at last
Figure 880031DEST_PATH_IMAGE003
With
Figure 824853DEST_PATH_IMAGE004
Whether exceed its control limit separately.If exceed the control limit, show that this process breaks down.

Claims (6)

1. Intermittent control system is carried out the method that segment failure monitors, it is characterized in that described method comprises following process:
At first, the multiple batches of batch process data that collect are launched to carry out standardization by variable, each sampling instant data matrix is carried out pivot analysis, obtain its corresponding CPV (1), in fragmentation procedure, if the dynamic relationship between data remains unchanged always, value CPV(1) will keep constant substantially; Secondly, fuzzy C-means clustering is the Fuzzy Cluster Analysis method that is fit to carry out soft division that puts forward in conjunction with fuzzy set theory and k-mean cluster; With reflecting the input of CPV (1) index of Data Dynamic characteristic transformation as fuzzy C-means clustering, realize the multistage soft division of batch process according to fuzzy maximum membership grade principle identification adjacent phases; Again, after segmentation is finished, each sub set up based on variable launch have the time become the improvement MPCA model of pivot covariance, determine under the normal condition
Figure 2013101882936100001DEST_PATH_IMAGE001
With
Figure 210128DEST_PATH_IMAGE002
The fault monitoring of statistic control limit then, judges at first during in-service monitoring that it belongs to that stage, calculate its correspondence
Figure 2013101882936100001DEST_PATH_IMAGE003
With
Figure 526578DEST_PATH_IMAGE004
Monitor statistic, judge whether it exceeds fault monitoring control limit, if exceed then show that fault takes place, institute carries out segment failure and monitors and finish.
2. according to claim 1 Intermittent control system is carried out the method that segment failure monitors, it is characterized in that, described CPV (1) changes the maximum information that changes in index CPV (1) acquisition procedure for corresponding dynamic perfromance, the size of its value, correlation information and dynamic relationship between the response data variable.
3. according to claim 2 Intermittent control system is carried out the method that segment failure monitors, it is characterized in that, the index CPV (1) that described dynamic perfromance changes is, multiple batches of batch process data, data matrix by each sampling instant after the variable expansion carries out pivot analysis, obtains the index CPV (1) that the representative data dynamic perfromance changes.
4. according to claim 3 Intermittent control system is carried out the method that segment failure monitors, it is characterized in that, the index CPV (1) that described Data Dynamic characteristic changes carries out fuzzy C-means clustering to each sampling instant, realizes soft division of multistage according to maximum fuzzy membership principle identification adjacent phases.
5. according to claim 4 Intermittent control system is carried out the method that segment failure monitors, it is characterized in that, fuzzy clustering realizes multistage soft division to described fuzzy C-means clustering to CPV (1), and each stage after the division sets up the improvement MPCA that becomes covariance when having and monitors model.
6. according to claim 5 Intermittent control system is carried out the method that segment failure monitors, it is characterized in that, described multistage soft division realizes soft division of multistage according to maximum fuzzy membership principle identification adjacent phases, each sub after dividing is set up improved MPCA monitor model, realize that the batch process multistage monitors.
CN201310188293.6A 2013-05-21 2013-05-21 Intermittent control system is carried out to the method for segment failure supervision Expired - Fee Related CN103279123B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310188293.6A CN103279123B (en) 2013-05-21 2013-05-21 Intermittent control system is carried out to the method for segment failure supervision

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310188293.6A CN103279123B (en) 2013-05-21 2013-05-21 Intermittent control system is carried out to the method for segment failure supervision

Publications (2)

Publication Number Publication Date
CN103279123A true CN103279123A (en) 2013-09-04
CN103279123B CN103279123B (en) 2015-12-23

Family

ID=49061686

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310188293.6A Expired - Fee Related CN103279123B (en) 2013-05-21 2013-05-21 Intermittent control system is carried out to the method for segment failure supervision

Country Status (1)

Country Link
CN (1) CN103279123B (en)

Cited By (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103838217A (en) * 2014-03-10 2014-06-04 北京工业大学 Method for monitoring faults in fermentation process based on MICA-OCSVM
CN103970092A (en) * 2014-04-13 2014-08-06 北京工业大学 Multi-stage fermentation process fault monitoring method based on self-adaption FCM algorithm
CN104063542A (en) * 2014-06-20 2014-09-24 东北大学 Method of monitoring and failure recognition during continuous rolling production process of seamless steel tubes based on multi-stage MPCA (Multilinear principal component analysis) method
CN105159247A (en) * 2015-08-05 2015-12-16 北京工业大学 Information transmission based phase affiliation judgment method for real time sampling points in intermittent process
CN105259895A (en) * 2015-10-14 2016-01-20 山东科技大学 Method and monitoring system for detecting and separating micro fault in industrial process
CN105353607A (en) * 2015-11-26 2016-02-24 江南大学 Data-difference-driving batch process self-learning dynamic optimization method
CN105511445A (en) * 2015-12-01 2016-04-20 沈阳化工大学 Multi-modal process fault detection method based on local neighbor standardization matrix
CN105607631A (en) * 2016-03-24 2016-05-25 辽宁工业大学 Batch process weak fault model control limit establishment method and weak fault monitoring method
CN105629958A (en) * 2016-02-06 2016-06-01 北京工业大学 Intermittence process fault diagnosis method based on sub-period MPCA-SVM
CN105955214A (en) * 2016-05-26 2016-09-21 华东理工大学 Batch process fault detection method based on sample timing sequence and neighborhood similarity information
CN107895224A (en) * 2017-10-30 2018-04-10 北京工业大学 A kind of MKECA fermentation process fault monitoring methods based on extension nuclear entropy load matrix
CN107942669A (en) * 2017-11-30 2018-04-20 海南师范大学 The limited rolling time horizon of batch injection moulding process mixes tracking and controlling method
CN108664002A (en) * 2018-04-27 2018-10-16 中国石油大学(华东) A kind of nonlinear dynamic process monitoring method towards quality
CN109254531A (en) * 2017-11-29 2019-01-22 辽宁石油化工大学 The optimal cost control method of multistage batch process with time lag and interference
CN109754010A (en) * 2018-12-29 2019-05-14 北京化工大学 A kind of multi-modal division methods of batch process of temporal constraint fuzzy clustering
CN109932908A (en) * 2019-03-20 2019-06-25 杭州电子科技大学 A kind of multidirectional pivot analysis process monitoring method based on alarm reliability fusion
CN110288724A (en) * 2019-06-27 2019-09-27 大连海事大学 A kind of batch process monitoring method based on wavelet function pivot analysis
CN110795695A (en) * 2019-10-28 2020-02-14 华北理工大学 Multi-model dynamic PCA (principal component analysis) fault monitoring method for eliminating time sequence correlation
CN111638707A (en) * 2020-06-07 2020-09-08 南京理工大学 Intermittent process fault monitoring method based on SOM clustering and MPCA
CN113341926A (en) * 2021-06-10 2021-09-03 兰州理工大学 Multi-stage intermittent process fault detection method based on sparse weighted neighborhood preserving embedding

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
张佳等: "多段MPCA法监测间歇过程的故障", 《计算机与应用化学》 *

Cited By (31)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103838217A (en) * 2014-03-10 2014-06-04 北京工业大学 Method for monitoring faults in fermentation process based on MICA-OCSVM
CN103838217B (en) * 2014-03-10 2016-08-10 北京工业大学 A kind of sweat fault monitoring method based on MICA-OCSVM
CN103970092A (en) * 2014-04-13 2014-08-06 北京工业大学 Multi-stage fermentation process fault monitoring method based on self-adaption FCM algorithm
CN103970092B (en) * 2014-04-13 2017-02-08 北京工业大学 Multi-stage fermentation process fault monitoring method based on self-adaption FCM algorithm
CN104063542A (en) * 2014-06-20 2014-09-24 东北大学 Method of monitoring and failure recognition during continuous rolling production process of seamless steel tubes based on multi-stage MPCA (Multilinear principal component analysis) method
CN105159247A (en) * 2015-08-05 2015-12-16 北京工业大学 Information transmission based phase affiliation judgment method for real time sampling points in intermittent process
CN105259895A (en) * 2015-10-14 2016-01-20 山东科技大学 Method and monitoring system for detecting and separating micro fault in industrial process
CN105259895B (en) * 2015-10-14 2017-08-11 山东科技大学 A kind of detection of industrial process small fault and separation method and its monitoring system
CN105353607A (en) * 2015-11-26 2016-02-24 江南大学 Data-difference-driving batch process self-learning dynamic optimization method
CN105353607B (en) * 2015-11-26 2017-10-27 江南大学 A kind of batch process self study dynamic optimization method driven by data difference
CN105511445A (en) * 2015-12-01 2016-04-20 沈阳化工大学 Multi-modal process fault detection method based on local neighbor standardization matrix
CN105629958A (en) * 2016-02-06 2016-06-01 北京工业大学 Intermittence process fault diagnosis method based on sub-period MPCA-SVM
CN105629958B (en) * 2016-02-06 2018-03-16 北京工业大学 A kind of batch process method for diagnosing faults based on sub-period MPCA SVM
CN105607631A (en) * 2016-03-24 2016-05-25 辽宁工业大学 Batch process weak fault model control limit establishment method and weak fault monitoring method
CN105955214B (en) * 2016-05-26 2018-10-02 华东理工大学 Batch process fault detection method based on sample time-series and neighbour's affinity information
CN105955214A (en) * 2016-05-26 2016-09-21 华东理工大学 Batch process fault detection method based on sample timing sequence and neighborhood similarity information
CN107895224A (en) * 2017-10-30 2018-04-10 北京工业大学 A kind of MKECA fermentation process fault monitoring methods based on extension nuclear entropy load matrix
CN107895224B (en) * 2017-10-30 2022-03-15 北京工业大学 MKECA fermentation process fault monitoring method based on extended nuclear entropy load matrix
CN109254531A (en) * 2017-11-29 2019-01-22 辽宁石油化工大学 The optimal cost control method of multistage batch process with time lag and interference
CN107942669A (en) * 2017-11-30 2018-04-20 海南师范大学 The limited rolling time horizon of batch injection moulding process mixes tracking and controlling method
CN108664002A (en) * 2018-04-27 2018-10-16 中国石油大学(华东) A kind of nonlinear dynamic process monitoring method towards quality
CN109754010A (en) * 2018-12-29 2019-05-14 北京化工大学 A kind of multi-modal division methods of batch process of temporal constraint fuzzy clustering
CN109932908B (en) * 2019-03-20 2022-03-01 杭州电子科技大学 Multi-directional principal component analysis process monitoring method based on alarm reliability fusion
CN109932908A (en) * 2019-03-20 2019-06-25 杭州电子科技大学 A kind of multidirectional pivot analysis process monitoring method based on alarm reliability fusion
CN110288724A (en) * 2019-06-27 2019-09-27 大连海事大学 A kind of batch process monitoring method based on wavelet function pivot analysis
CN110288724B (en) * 2019-06-27 2021-08-03 大连海事大学 Batch process monitoring method based on wavelet function principal component analysis
CN110795695A (en) * 2019-10-28 2020-02-14 华北理工大学 Multi-model dynamic PCA (principal component analysis) fault monitoring method for eliminating time sequence correlation
CN111638707A (en) * 2020-06-07 2020-09-08 南京理工大学 Intermittent process fault monitoring method based on SOM clustering and MPCA
CN111638707B (en) * 2020-06-07 2022-05-20 南京理工大学 Intermittent process fault monitoring method based on SOM clustering and MPCA
CN113341926A (en) * 2021-06-10 2021-09-03 兰州理工大学 Multi-stage intermittent process fault detection method based on sparse weighted neighborhood preserving embedding
CN113341926B (en) * 2021-06-10 2023-11-17 兰州理工大学 Multi-stage intermittent process fault detection method based on sparse weighted neighborhood preserving embedding

Also Published As

Publication number Publication date
CN103279123B (en) 2015-12-23

Similar Documents

Publication Publication Date Title
CN103279123A (en) Method of monitoring faults in sections for intermittent control system
Chen et al. A just-in-time-learning-aided canonical correlation analysis method for multimode process monitoring and fault detection
CN110322048B (en) Fault early warning method for production logistics conveying equipment
CN104267668A (en) Bayes-method-based spaceflight valve part fault diagnosis method in machining process
CN109297689B (en) Large-scale hydraulic machinery intelligent diagnosis method introducing weight factors
CN107895224B (en) MKECA fermentation process fault monitoring method based on extended nuclear entropy load matrix
CN107146004A (en) A kind of slag milling system health status identifying system and method based on data mining
CN105607631B (en) The weak fault model control limit method for building up of batch process and weak fault monitoring method
CN110245460B (en) Intermittent process fault monitoring method based on multi-stage OICA
CN101403923A (en) Course monitoring method based on non-gauss component extraction and support vector description
CN102662321A (en) Online updating method of principal component analysis monitoring model
CN102929148A (en) Multiple product production mode statistical process control method based on T-K control chart
CN105955214B (en) Batch process fault detection method based on sample time-series and neighbour's affinity information
CN105469138A (en) Control system performer fault diagnosis method based on particle swarms and support vector machine
CN106950945B (en) A kind of fault detection method based on dimension changeable type independent component analysis model
CN105675038A (en) Device for predicting faults of instruments
CN108445867B (en) non-Gaussian process monitoring method based on distributed ICR model
CN112257767B (en) Product key part state classification method aiming at class imbalance data
CN108388234A (en) A kind of fault monitoring method dividing changeable gauge block pca model based on correlation
CN103488561A (en) kNN (k-nearest neighbor) fault detection method for online upgrading master sample model
CN111324110A (en) Fermentation process fault monitoring method based on multiple shrinkage automatic encoders
CN103926919B (en) Industrial process fault detection method based on wavelet transformation and Lasso function
CN108960309A (en) A kind of dynamic process monitoring method rejected based on RBF neural autocorrelation
CN112214006A (en) Intermittent process fault detection method and system considering two-dimensional dynamic characteristics
CN104615123B (en) K-nearest neighbor based sensor fault isolation method

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20151223

Termination date: 20170521