CN102870057A - Plant diagnosis device, diagnosis method, and diagnosis program - Google Patents

Plant diagnosis device, diagnosis method, and diagnosis program Download PDF

Info

Publication number
CN102870057A
CN102870057A CN2010800659112A CN201080065911A CN102870057A CN 102870057 A CN102870057 A CN 102870057A CN 2010800659112 A CN2010800659112 A CN 2010800659112A CN 201080065911 A CN201080065911 A CN 201080065911A CN 102870057 A CN102870057 A CN 102870057A
Authority
CN
China
Prior art keywords
plant equipment
operating condition
classification
judged
diagnosis object
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
CN2010800659112A
Other languages
Chinese (zh)
Other versions
CN102870057B (en
Inventor
关合孝朗
江口彻
楠见尚弘
深井雅之
清水悟
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.)
Hitachi Ltd
Original Assignee
Hitachi Ltd
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 Hitachi Ltd filed Critical Hitachi Ltd
Publication of CN102870057A publication Critical patent/CN102870057A/en
Application granted granted Critical
Publication of CN102870057B publication Critical patent/CN102870057B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B23/00—Testing or monitoring of control systems or parts thereof
    • G05B23/02—Electric testing or monitoring
    • G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
    • G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
    • G05B23/0224—Process history based detection method, e.g. whereby history implies the availability of large amounts of data
    • G05B23/0227—Qualitative history assessment, whereby the type of data acted upon, e.g. waveforms, images or patterns, is not relevant, e.g. rule based assessment; if-then decisions
    • G05B23/0229—Qualitative history assessment, whereby the type of data acted upon, e.g. waveforms, images or patterns, is not relevant, e.g. rule based assessment; if-then decisions knowledge based, e.g. expert systems; genetic algorithms
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B23/00—Testing or monitoring of control systems or parts thereof
    • G05B23/02—Electric testing or monitoring
    • G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
    • G05B23/0259—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the response to fault detection
    • G05B23/0283—Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL]

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Testing And Monitoring For Control Systems (AREA)

Abstract

Disclosed is a plant diagnosis device which detects abnormality in the subject being diagnosed from the measurement signal of the plant being diagnosed and reports the abnormality. Specifically disclosed is a plant diagnosis device which is configured of: a state variation detection unit which categorizes and stores the measurement signals of the plant being diagnosed and determines that the state has varied, when the measurement signal does not belong to the categories obtained by the categorization; an abnormality determination unit which determines that abnormality has occurred in the subject being diagnosed, using a signal from the state variation detection unit as a part of inputted signals; an alarm generation means which reports the output from the abnormality determination unit to the outside; and an erroneous report inhibition unit which detects events other than the abnormality in the plant being diagnosed and inhibits the output from the abnormality determination unit.

Description

The diagnostic device of plant equipment, diagnostic method and diagnostic routine
Technical field
The present invention relates to diagnostic device, diagnostic method and the diagnostic routine of plant equipment.
Background technology
The diagnostic device of plant equipment is when plant equipment has produced unusual transition events, accident etc., and this is unusual based on surveying from the measurement data of plant equipment, the generation of accident.
The diagnostic device of employing adaptive resonance theory (Adaptive Resonance Theory:ART) is disclosed in patent documentation 1.Adopt the diagnostic device of ART to have the function that multidimensional data is categorized into classification according to its similar degree.
In the technology of patent documentation 1, at first use the measurement data of ART when normal to be categorized into a plurality of classifications (normal category).Secondly, use ART that current measurement data is categorized into classification.Can't be categorized in the normal category in this measurement data, generate new classification (new classification).The generation of new classification means that variation has occured the state of plant equipment.Therefore, utilize the generation of new classification to judge unusual generation, surpassed in the generation rate of new classification be diagnosed as in the situation of threshold value unusual.
Technical literature formerly
Patent documentation
Patent documentation 1: TOHKEMY 2005-165375 communique
Summary of the invention
(inventing problem to be solved)
In patent documentation 1, be based on " current measurement data can't be categorized in the normal category, be to generate new classification in the state of plant equipment has occured to change " this prerequisite, and be judged as unusual.
Yet the state of plant equipment not only just changes in having produced unusually, also can change in timeliness deteriorated (aging degradation), operating condition variation.And also can produce new classification this moment.Thereby, can not think this moment to have produced new classification, namely produced the unusual of plant equipment.
Here, operating condition refers to that amount (generated energy that plant equipment generates) that the environmental baseline (atmospheric temperature, humidity etc.) in the place that plant equipment is set, the operation by operating personnel determine etc. and instrument characteristics do not have the external environment factor of direct relation.And irrelevant with having or not of abnormal, the value of various measurement data also changes if operating condition changes then.
Although timeliness is deteriorated, operating condition changes is not unusual, it can be diagnosed as unusually in the method for patent documentation 1.That is to say, normal condition can be diagnosed as unusually that the result becomes the reason that produces wrong report.
The object of the invention is to deteriorated as the timeliness of the essential factor of state variation by difference, operating condition changes, unusually reduce the generation rate of wrong report.
(being used for solving the means of problem)
The diagnostic device of a kind of plant equipment of the present invention, unusual and the report of surveying the diagnosis object plant equipment according to the measuring-signal of diagnosis object plant equipment is unusual, the diagnostic device of described plant equipment comprises: the state variation test section, its measuring-signal with the diagnosis object plant equipment is categorized into classification and stores, and is judged to be state in the situation that does not belong to the classification of classifying variation has occured; Abnormality juding section, it will be judged to be as a part of inputting from the signal of state variation test section and produce in the diagnosis object plant equipment unusually; The alarm generation unit, its output report with this abnormality juding section arrives outside; With the wrong report blocking portion, the unusual event in addition that it surveys the diagnosis object plant equipment stops the output of abnormality juding section.
In addition, preferred wrong report blocking portion comprises timeliness degradation section, this timeliness degradation section obtains weight coefficient with the measuring-signal of diagnosis object plant equipment, surpassed to be judged to be in the peaked situation at the variable quantity of weight coefficient and broken away from the deteriorated scope of timeliness, timeliness degradation section stops the output of abnormality juding section when not breaking away from the deteriorated scope of timeliness.
In addition, preferred wrong report blocking portion comprises that operating condition changes test section, this operating condition variation test section becomes classification to store the operating condition Data classification of diagnosis object plant equipment, be judged to be operating condition and do not change in the situation that does not produce new classification, operating condition changes the output of test section prevention abnormality juding section when operating condition has occured to change.
In addition, the operating condition data that preferably change the input of test section as operating condition do not have the external environment factor of direct relation to consist of by the characteristic with the apparatus that consists of the diagnosis object plant equipment.
The diagnostic device of a kind of plant equipment of the present invention, unusual and the report of surveying the diagnosis object plant equipment according to the measuring-signal of diagnosis object plant equipment is unusual, the diagnostic device of described plant equipment comprises: the state variation test section, its measuring-signal with the diagnosis object plant equipment is categorized into classification and stores, and is judged to be state in the situation that does not belong to the classification of classifying variation has occured; Timeliness degradation section, its measuring-signal with the diagnosis object plant equipment is obtained weight coefficient, has surpassed to be judged to be in the peaked situation at the variable quantity of weight coefficient to have broken away from the deteriorated scope of timeliness; Operating condition changes test section, and its operating condition Data classification with the diagnosis object plant equipment becomes classification to store, and is judged to be operating condition and does not change in the situation that does not produce new classification; Abnormality juding section, variation has occured and has been judged to be to have broken away from the deteriorated scope of timeliness and changed test section by operating condition by timeliness degradation section to be judged to be in operating condition do not change at the state that is judged to be by the state variation test section in it, is judged to be diagnosis object and has produced unusually; With the alarm generation unit, its output report with abnormality juding section arrives outside.
The diagnostic device of a kind of plant equipment of the present invention possesses: the measuring-signal database, and it preserves the measuring-signal of diagnosis object plant equipment; The deal with data extraction unit, it extracts the diagnostic signal that uses for the state of diagnosing the diagnosis object plant equipment from the measuring-signal database; The reference signal database, it preserves diagnostic signal; Taxon, it becomes classification with the Data classification of preserving in the reference signal database; The classification results database, it is preserved classification as normal category; Diagnosis unit, it uses the information of the up-to-date diagnostic signal that is extracted by the deal with data extraction unit and the normal category of preserving in the classification results database, the state of diagnosis diagnosis object plant equipment belongs to normally, unusual, operating condition variation, which kind of deteriorated situation of timeliness; The diagnostic result database, it preserves the diagnostic result of diagnosis unit; Unit with the information of in the diagnostic result database, preserving to image display device output, wherein, diagnosis unit possesses abnormality juding section, the state variation test section, timeliness degradation section and operating condition change test section, the state variation test section possesses following function: be judged to be state under the up-to-date diagnostic signal that is extracted by the deal with data extraction unit does not belong to the situation of the normal category of preserving in the classification results database variation has occured, timeliness degradation section possesses following function: the time amplitude of variation when deriving normal condition and the relation between the weight coefficient variable quantity, be judged to be in the peaked situation of the variable quantity of the weight coefficient when in diagnostic procedure, having surpassed normal condition and broken away from the deteriorated scope of timeliness, operating condition changes test section and possesses following function: will be by becoming with operating condition Data classification that instrument characteristics does not have the external environment factor of direct relation to consist of to be judged to be operating condition in the situation that does not produce new classification in the classification and not change, in abnormality juding section, variation has occured being judged to be state by the state variation test section, and be judged to be by timeliness degradation section and broken away from the deteriorated scope of timeliness, and change test section by operating condition and be not judged to be in operating condition do not change, be judged to be unusual.
The diagnostic method of a kind of plant equipment of the present invention, unusual and the report of surveying the diagnosis object plant equipment according to the measuring-signal of diagnosis object plant equipment is unusual, wherein the measuring-signal of diagnosis object plant equipment being categorized into classification stores, in the situation that does not belong to the classification of classifying, be judged to be state variation has occured, measuring-signal with diagnosis object is obtained weight coefficient, surpassed to be judged to be in the peaked situation at the variable quantity of weight coefficient and broken away from the deteriorated scope of timeliness, become classification to store at the operating condition Data classification with diagnosis object, being judged to be operating condition in the situation that does not produce new classification does not change, at this moment, be judged to be and produced unusually in the diagnosis object plant equipment, and report is to outside.
The diagnostic routine of a kind of plant equipment of the present invention, unusual and the report of surveying the diagnosis object plant equipment according to the measuring-signal of diagnosis object plant equipment is unusual, the diagnostic routine of described plant equipment comprises: the state variation detecting step, the measuring-signal of diagnosis object is categorized into classification stores, in the situation that does not belong to the classification of classifying, be judged to be state variation has occured; The abnormality juding step, will from the signal of state variation detecting step be judged to be as the part of input produced in the diagnosis object unusual; Alarm produces step, with the output report of abnormality juding step to outside; Stop step with wrong report, survey the unusual event in addition of diagnosis object, stop the output of abnormality juding section.
In addition, preferred wrong report stops step to comprise timeliness degradation step, in this timeliness degradation step, use the measuring-signal of diagnosis object plant equipment to obtain weight coefficient, surpassed to be judged to be in the peaked situation at the variable quantity of weight coefficient and broken away from the deteriorated scope of timeliness, in this timeliness degradation step, when not breaking away from the deteriorated scope of timeliness, stop the output of abnormality juding step.
In addition, preferred wrong report stops and comprises in the step that operating condition changes detecting step, operating condition Data classification with the diagnosis object plant equipment in this operating condition variation detecting step becomes classification to store, being judged to be operating condition in the situation that does not produce new classification does not change, change in the detecting step output of prevention abnormality juding step when operating condition has occured to change at operating condition.
In addition, the operating condition data that preferably change the input of detecting step as operating condition do not have the external environment factor of direct relation to consist of by the characteristic with the apparatus that consists of diagnosis object.
The diagnostic routine of a kind of plant equipment of the present invention, unusual and the report of surveying the diagnosis object plant equipment according to the measuring-signal of diagnosis object plant equipment is unusual, the diagnostic routine of described plant equipment comprises: the state variation detecting step, the measuring-signal of diagnosis object plant equipment is categorized into classification stores, in the situation that does not belong to the classification of classifying, be judged to be state variation has occured; Timeliness degradation step is obtained weight coefficient with the measuring-signal of diagnosis object plant equipment, has surpassed to be judged to be in the peaked situation at the variable quantity of weight coefficient to have broken away from the deteriorated scope of timeliness; Operating condition changes detecting step, becomes classification to store the operating condition Data classification of diagnosis object plant equipment, is judged to be operating condition and does not change in the situation that does not produce new classification; Variation has occured and has been judged to be to have broken away from the deteriorated scope of timeliness and changed detecting step by operating condition by timeliness degradation step to be judged to be operating condition and to be judged to be diagnosis object in not changing and to have produced unusually at the state that is judged to be by the state variation detecting step in the abnormality juding step; Produce step with alarm, with the output report of abnormality juding step to outside.
(invention effect)
Change, unusually distinguish by, operating condition deteriorated to timeliness, thus enough generation rates that reduces wrong report that subtracts.
In addition, the data trend causes of change belongs to that timeliness is deteriorated by providing to operating personnel, operating condition changes, unusual which kind of situation, thereby helps the maintenance plan of plant equipment to draw up a plan for.
Description of drawings
Fig. 1 is the block diagram of expression diagnostic device of the present invention.
Fig. 2 (a) is the figure of the process flow diagram under the expression normal condition mode of learning.
Fig. 2 (b) is the figure of the process flow diagram under the expression diagnostic mode.
Fig. 3 (a) is the figure that is illustrated in the diagnosis that makes the action of 2 patterns in each sampling period.
Fig. 3 (b) is illustrated in the figure that carries out the normal condition mode of learning in specified time limit, carries out the diagnosis of diagnostic mode within each cycle.
Fig. 3 (c) is the figure that is illustrated in the example that moment that operating personnel set carries out.
Fig. 4 (a) is the block diagram of expression data pre-processing device and ART module.
Fig. 4 (b) is the block diagram of the F0 layer of expression ART module.
Fig. 4 (c) is the block diagram of the F1 layer of expression ART module.
Fig. 5 (a) be presentation graphs 4 embodiment carrying out result's the curve map of category classification.
Fig. 5 (b) is the curve map of the time dependent example of measuring-signal when abnormal is shown.
Fig. 6 (a) is the picture of setting diagnostic data items.
Fig. 6 (b) is the block diagram of the taxon 400 of key diagram 1.
Fig. 7 (a) is the figure of picture that is illustrated in the form of the data of preserving in the measuring-signal database.
Fig. 7 (b) is the figure of picture that is illustrated in the form of the data of preserving in the reference signal database.
Fig. 7 (c) is the figure of picture that is illustrated in the form of the data of preserving in the classification results database.
Fig. 8 (a) is the figure of the formation of expression abnormality juding section 510.
Fig. 8 (b) is the figure of form that demonstrates the data of diagnostic result database at picture.
Fig. 8 (c) is the picture disply example that demonstrates the diagnostic result of diagnostic result database in time series.
Fig. 9 (a) is the figure of the action of explanation timeliness degradation section.
Fig. 9 (b) is the figure of the determinating reference of the deteriorated scope affirmation of explanation timeliness.
Figure 10 (a) is the figure of the action of explanation operating condition the 1st embodiment that changes test section.
Figure 10 (b) is the figure of the action of explanation operating condition the 2nd embodiment that changes test section.
Figure 10 (c) is the figure of the action of explanation operating condition the 3rd embodiment that changes test section.
Figure 11 is the block diagram of expression gas turbo-generator tool equipment.
Figure 12 is the figure of the result of the action example of expression when using diagnostic device of the present invention to gas turbo-generator tool equipment.
Figure 13 is the figure that has summarized the operation condition of each one of diagnosis unit.
Embodiment
Below, with accompanying drawing the present invention is described.
Embodiment
Fig. 1 is the block diagram of expression diagnostic device of the present invention.In this figure, diagnosed the state of plant equipment 100 by diagnostic device 200.
In diagnostic device 200, possess as arithmetic unit: deal with data extraction unit 300, taxon 400, diagnosis unit 500, alarm generation unit 600.In addition, in diagnostic device 200, possess as database: measuring-signal database 230, reference signal database 240, classification results database 250, diagnostic result database 260.In addition, in this figure, database slightly is designated as DB.Database is the inscape of diagnostic device 200 referred in this, and the information that records in each database is commonly called e-file (electronic data) by electronization.
In addition, in diagnostic device 200, as possessing with the interface of outside: outer input interface 210 and outside output interface 220.And, via outer input interface 210, will comprise that the external input signal 2 that the operation of the measuring-signal 1 of value of the various quantity of states of having measured plant equipment 100 and the external input device 910 that comprises keyboard 920 and mouse 930 that possesses by running management chamber 900 generates inputs to diagnostic device 200.In addition, via outside output interface 220, export image display information 14 in the running management chamber 900 image display device 940 (image displaying part) from diagnostic device 200.
In addition, in the present embodiment, deal with data extraction unit 300, taxon 400, diagnosis unit 500, alarm generation unit 600, measuring-signal database 230, reference signal database 240, classification results database 250, diagnostic result database 260 all are in the inside of diagnostic device 200, but also the part of these parts can be disposed at the outside of diagnostic device 200, only data be communicated.In addition, the plant equipment of establishing in the present embodiment as diagnosis object is 1, but also can be by many plant equipment of 1 diagnostic device, 200 diagnosis.
Below, the action of diagnostic device 200 is described.At first, the measuring-signal 3 that is transfused to via outer input interface 210 is saved in the measuring-signal database 230.
In deal with data extraction unit 300, among the measuring-signal 5 of measuring-signal database 230, preserving, extract diagnostic diagnostic signal 6, it is saved in the reference signal database 240.In reference signal database 240, preserve the measuring-signal during operating personnel are judged to be normally.In addition, in deal with data extraction unit 300, extract data items based on operating personnel's setting.Use Fig. 6 to narrate in the back this detailed content.
Taxon 400 is categorized into classification with reference signal 7.Classification results 8 is kept in the classification results database 250.In addition, use Fig. 4 to narrate in the back the contents processing of taxon 400.
In diagnosis unit 500, process up-to-date diagnostic signal 6 and the classification results 9 of preservation in classification results database 250, diagnose the state of plant equipment.Diagnosis unit 500 changes test section 540 by abnormality juding section 510, state variation test section 520, timeliness degradation section 530, operating condition and consists of.
State variation test section 520 judges whether the state of plant equipment 100 variation has occured.From the state variation result of determination 15 of state variation test section 520 outputs, be in the situation that has occured to change at the state of being judged to be " 1 ", in the situation that the state that is judged to be does not change, be " 0 ".
In state variation test section 520, compare the up-to-date diagnostic signal 6 that is extracted by deal with data extraction unit 300 and the classification results 9 of in classification results database 250, preserving, in the situation that belongs to the classification that classification results 9 comprises, diagnostic signal 6 is categorized into this classification.On the other hand, the up-to-date diagnostic signal 6 and the classification results 9 that are relatively extracted by deal with data extraction unit 300, in the situation that does not belong to the classification that classification results 9 comprises, produce new classification (below be labeled as new classification.)。Use Fig. 4 to narrate in the back the detailed contents processing of state variation test section 520.
In timeliness degradation section 530, judge that the state variation of plant equipment is whether in the deteriorated scope of timeliness.From the timeliness deterioration judging result 16 of timeliness degradation section 530 output, broken away from the state variation of plant equipment and to be " 1 " in the deteriorated scope of timeliness, be " 0 " in the deteriorated scope of timeliness.Use Fig. 9 to narrate in the back the detailed contents processing of timeliness degradation section 530.
Change in the test section 540 at operating condition, judge whether the operating condition of plant equipment variation has occured.The operating condition that changes test section 540 outputs from operating condition changes result of determination 17, is " 1 " at operating condition in not changing, and is " 0 " at operating condition in having occured to change.Use Figure 10 to narrate in the back the detailed contents processing that operating condition changes test section 540.
In abnormality juding section 510, change result of determination 17 with state variation result of determination 15, timeliness deterioration judging result 16 and operating condition and judge whether plant equipment has produced unusually.Use Fig. 8 to narrate in the back the detailed contents processing of abnormality juding section 510.
Comprising from the diagnostic result 10 of diagnosis unit 500 outputs: the result of determination of abnormity diagnosis section 510, state variation result of determination 15, timeliness deterioration judging result 16, operating condition change result of determination 17.Diagnostic result 10 is kept in the diagnostic result database 260.
In alarm generation unit 600, use the measuring-signal 4 of the diagnostic result 11 of in diagnostic result database 260, preserving and the up-to-date moment of in measuring-signal database 230, preserving to determine whether the generation alarm.
Alarm generation unit 600 has following 2 kinds of determinating references (condition 1 and condition 2) that produce alarm, and these determinating reference combination in any are determined whether producing alarm.Here, the combination in any that produces of alarm refers to: produce alarm in the situation that following condition 1 and following condition 2 these two conditions are all set up, produce alarm etc. in the situation that one of them condition among following condition 1 and the following condition 2 is set up.
Condition 1: the measuring-signal 4 in the up-to-date moment has broken away from the scope (threshold value) of regulation.
Condition 2: within specified time limit, be judged to be unusual ratio by abnormality juding section 510 and surpass certain value (threshold value).
In addition, the threshold value in condition 1 and the condition 2 is the value of being set by operating personnel.
Carried out producing in alarm generation unit 600 in the situation of judgement of alarm, alarm generation unit 600 is sent to outside output interface 220 with alarm signal 13.Alarm signal 13 is transformed into image display information 14 via outside output interface 220, is shown in image display device 940.
In the present embodiment, use image display device 940 and operator to contact alarm, but be not limited thereto, also can at first produce alarm song and cause that the operator notes the generation of alarm.In addition, also can after producing alarm song or at the generation alarm song, make simultaneously image display device 940 show image display information 14.
In addition, also the diagnostic result 12 (result of determination in the abnormity diagnosis section 510, state variation result of determination 15, timeliness deterioration judging result 16, operating condition change result of determination 17) of preserving can be shown in image display device 940 via outside output interface 220 in diagnostic result database 260.
The diagnostic device information 50 of preserving in measuring-signal database 230, reference signal database 240, classification results database 250 and diagnostic result database 260 in addition, can be shown in image display device 940.In addition, also can use as required based on the external input signal 2 of the operation of external input device 910 and revise these information.
The invention is characterized in diagnosis unit 500, to possess abnormality juding section 510, timeliness degradation section 530 and operating condition and change test section 540, be judged to be unusually thereby timeliness degradation, operating condition can not changed to detect.
That is to say, although state variation test section 520 is surveyed the unusual of plant equipment energetically, but originally should not be judged to be that unusual timeliness is deteriorated, operating condition changes owing in the content of this detection, comprise, thereby by surveying timeliness is deteriorated, operating condition changes and they are carried out comprehensive judgement via abnormality juding section 510, thereby stoped wrong report.This just means that timeliness degradation section 530 and operating condition variation test section 540 can consist of the wrong report blocking portion for state variation test section 520.
In state variation test section 520, new classification be can't be categorized in the normal category in current measurement data, be to generate in the state of plant equipment has occured to change.
Yet, change in the abnormal because the state of plant equipment is not only, also can change in, operating condition deteriorated in timeliness changes, if thereby 520 actions of state variation test section produce new classification.Here, operating condition refers to the environmental baseline (atmospheric temperature, humidity etc.) in the place that plant equipment is set, from the generated energy of plant equipment output etc., irrelevant with having or not of abnormal, if operating condition changes, then the value of various measurement data also changes.
Although timeliness is deteriorated, operating condition changes is not unusual, is only judging in the unusual method and they can be diagnosed as unusually with the generation of new classification.This becomes the reason that produces wrong report.
In the present invention, by in diagnosis unit 500, possessing abnormality juding section 510, timeliness degradation section 530, operating condition variation test section 540, thereby owing to deteriorated, the operating condition variation, unusual of timeliness of having distinguished as the essential factor of state variation, so can reduce the generation rate of wrong report.
In addition, use do not possess abnormality juding of the present invention section 510, diagnostic device that timeliness degradation section 530, operating condition change test section 540, timeliness is deteriorated in order to distinguish, operating condition changes, unusually then need long-standing service data.That is, when making the action of normal condition mode of learning, need to estimate all operating conditions and the deteriorated data of timeliness.Therefore, from mechanical equipment operation begin till begin to diagnose during elongated.Yet, the diagnostic device of the application of the invention, even if can few stage of diagnostic running also can distinguish unusually, timeliness is deteriorated, operating condition changes, thereby can shorten the importing diagnostic device required during.
Fig. 2 is the process flow diagram of elemental motion of the diagnostic device 200 of presentation graphs 1.Fig. 2 (a) is the figure of expression normal condition mode of learning, and Fig. 2 (b) is the figure of expression diagnostic mode.
Below, the inscape of putting down in writing with Fig. 1 describes.
Diagnostic device 200 has these 2 elemental motions of diagnostic mode that become the state of the normal condition mode of learning of classification and diagnosis plant equipment 100 based on the Data classification of the information of preserving in the reference signal database 240 when normal.
In Fig. 2 (a), carry out the normal condition mode of learning by carrying out successively step 1000 and 1010.
At first, in step S1000, make 300 actions of deal with data extraction unit, among the measuring-signal 5 of measuring-signal database 230, extract diagnostic signal 6.Diagnostic signal 6 is kept in the reference signal database 240.To be operating personnel (operator) be judged to be data during normal with the operating condition of plant equipment 100 to the data of preserving in the reference signal database 240.In addition, use Fig. 6 to narrate in the back the data items of preserving in the reference signal database 240.
Then, in step S1010, make taxon 400 actions, the reference signal 7 that will preserve in reference signal database 240 is classified, and classification results 8 is saved in the classification results database 250.
Like this, under the normal condition mode of learning of Fig. 2 (a), carry out deal with data extraction unit 300, measuring-signal database 230, reference signal database 240, the taxon 400 of the row in the left side in the diagnostic device 200 used Fig. 1, the processing of classification results database 250.
Under the diagnostic mode shown in Fig. 2 (b), carry out by carrying out successively step S1100, S1110 and S1120.
At first, in step S1100, will be taken in the diagnostic device 200 from the measuring-signal 1 of plant equipment 100 via outer input interface 210, and measuring-signal 3 is saved in the measuring-signal database 230.
Then, make 300 actions of deal with data extraction unit, among measuring-signal database 230, extract measuring-signal 5, and up-to-date diagnostic signal 6 is sent to diagnosis unit 500 constantly.
In step S1110, make diagnosis unit 500 actions.In diagnosis unit 500, the order that changes test section 540, abnormality juding section 510 according to state variation test section 520, timeliness degradation section 530, operating condition makes the operational part action.To send and be saved in the diagnostic result database 260 from the diagnostic result 10 of diagnosis unit 500 outputs.Be transformed into image display information 14 from the diagnostic result 12 of diagnostic result database 260 outputs via outside output interface 220, and export image display device 940 to.
In step S1120, make 600 actions of alarm generation unit, judgement could produce alarm.In the situation that produces alarm, the alarm signal 13 that alarm generation unit 600 is exported is transformed into image display information 14 via outside output interface 220, and exports image display device 940 to.Thus, the operating personnel (operator) to plant equipment 100 notify alarm.
Like this, under the diagnostic mode of Fig. 2 (b), carry out diagnosis unit 500 (state variation test section 520, timeliness degradation section 530, operating condition change test section 540, abnormality juding section 510), outside output interface 220, the image display device 940 of the row on the right side in the diagnostic device 200 used Fig. 1, the processing of alarm generation unit 600.In addition, yes has used outer input interface 210, measuring-signal database 230, deal with data extraction unit 300 for its prerequisite.
Fig. 3 is the figure of sequential of the process flow diagram (Fig. 2 (b)) of the process flow diagram (Fig. 2 (a)) of the explanation normal condition mode of learning of carrying out diagnostic device 200 and diagnostic mode.
Diagnostic device 200 obtains measuring-signal 1 from plant equipment 100 in each sampling period.
In Fig. 3 (a), in each sampling period, make these two patterns of normal condition mode of learning and diagnostic mode move to diagnose.
In addition, in Fig. 3 (b), make the action of normal condition mode of learning in during the setting of each regulation, within each sampling period, only make diagnostic mode move to diagnose.
And in Fig. 3 (c), operating personnel implement to set the operation between the learning period, between diagnostic period, and constantly make normal condition mode of learning and diagnostic mode action at this.
No matter adopt any method, can both within each sampling period, carry out diagnostic mode, diagnose the state of plant equipment onlinely.
Below, particularly contents processing and the contents processing under the diagnostic mode of normal condition mode of learning being described successively by this order, its prerequisite is all to carry out the processing that the process variable that will input is categorized into classification under which pattern.Taxon 400 is equivalent to carry out the unit of this processing under the normal condition mode of learning, and state variation test section 520 is equivalent to carry out the unit of this processing under diagnostic mode.Thereby, before the action specification that enters under each pattern, the viewpoint of category classification method is described with the block diagram of Fig. 4 as public awareness.
Below, narration still also can be adopted other clustering methods such as vector quantization to the situation of taxon 400 and state variation test section 520 application self-adapting resonance theories (Adaptive Resonance Theory:ART).
Shown in Fig. 4 (a), taxon 400 and state variation test section 520 are made of data pre-processing device 710 and ART module 720.And ART module 720 possesses F0 layer 721, F1 layer 722, F2 layer 723, storer 724 and chooser system 725, and they mutually combine.In addition, F0 layer 721, F1 layer 722 for example consist of shown in Fig. 4 (b), Fig. 4 (c) like that.
In this Fig. 4 (a), at first, data pre-processing device 710 is transformed into service data the input data of ART module 720.Particularly, carry out following (1) (2) formula.Below, this program (step) is described.
At first, by each measure the item, calculate maximal value and minimum value.Come data are carried out standardization with the maximal value that calculates and minimum value.Here, take the process variable xi of plant equipment as example illustrates standardized method.
The data number of xi is N, and n measured value is made as xi (n).In addition, the maximal value in N the data and minimum value are made as respectively Max_i, Min_i, then the data Nxi after the standardization (n) represents with following (1) formula.
[mathematical expression 1]
Nxi(n)＝α+(1-α)×(xi(n)-Min_i)/(Max_i-Min_i)…(1)
At this, α (0≤α＜0.5) is constant, according to above-mentioned (1) formula and data are standardized as the scope of [α, 1-α].
Then, the complement of the data after the normalized joins in the input data.The complement CNxi (n) of standardized data Nxi (n) calculates with following (2) formula.
[mathematical expression 2]
CNxi(n)＝1-Nxi(n) …(2)
In data pre-processing device 710, a plurality of input data are carried out above-mentioned (1) (2) formula, to comprise that the data of complement CNxi (n) of the standardized data Nxi (n) that obtains as its result and standardized data as input data I i (n), input to ART module 720.Above step is included in during the service data of carrying out in the data pre-processing device 710 processes to the input data transformation of ART module 720.
In ART module 720, will input data I i (n) and be categorized into a plurality of classifications.For this reason, ART module 720 possesses F0 layer 721, F1 layer 722, F2 layer 723, storer 724 and chooser system 725, and they mutually combine.F1 layer 722 and F2 layer 723 carry out combination via weight coefficient.Weight coefficient represents to input the prototype (prototype) of the classification that data are classified.Here, other typical value of prototype representation class.
The algorithm of ART module 720 then, is described.
Shown in algorithm summary processing 1 described as follows when having inputted from data to ART module 720～processing 5.
Process 1: in Fig. 4 (b), the F0 layer 721 by the expression contents processing carries out standardization with input vector, removes noise.
Process 2: the input data by inputing to F1 layer 722 and the comparison of weight coefficient, select the candidate of suitable classification according to the contents processing of Fig. 4 (c).
Process 3: according to the appropriateness of recently estimating the classification of being selected by chooser system 725 of parameter ρ.If be judged as suitably, then input data and be classified in this classification, enter and process 4.On the other hand, if be not judged as suitably, then this classification is reset, and selects the candidate (re-treatment 2) of suitable classification among other classifications.If the value that the increases parameter ρ then classification of classification attenuates, if reduce the chap of then classifying of the value of ρ.This parameter ρ is called warning (vigilance) parameter.
Process 4: if whole known classifications is reset in processing 2, then judges the input data and belong to new classification, generate the new weight coefficient of the prototype of the new classification of expression.
Process 5: if the input data are classified into classification J, the weight coefficient WJ (new) corresponding with classification J then uses weight coefficient WJ (old) in the past and input data p (or the data that gone out by the input data fork) to be updated according to following (3) formula.
[mathematical expression 3]
WJ(new)＝Kw·p+(1-Kw)·WJ(old) …(3)
Here, Kw is Study rate parameter (0＜Kw＜1), is the value that determines input vector is reflected to the degree of new weight coefficient.
The data classification algorithm of ART module 720 is characterised in that above-mentioned processing 4.
Processing in 4, recording in the situation of the different input data of the style of (preservation) having inputted from the classification results database 250 of Fig. 1, can not change the style that records and record new style.Thereby, record new design while can record the style of learning over.
Like this, if given as the service data that gives of input data in advance, the style that gives of ART module 720 study then.Therefore, in the complete ART module 720 of study, inputted new input data, then can judge which style with the past is close according to above-mentioned algorithm.In addition, if the style for not lived through in the past then is classified into new classification.
Particularly, carry out above whole processing 1～processing 5 with following step.
At first, Fig. 4 (b) is the block diagram of the formation of expression F0 layer 721.F0 layer 721 is made of function blocks 71,72,73,74, manages throughout and carries out respectively following (4) (5) (6) (7) formula in the functional block 71,72,73,74, obtains the standardization input vector.
A succession of processing in the F0 layer 721 of Fig. 4 (b) is: will give the again standardization of input data I i of function blocks 71 constantly, and generate the standardization input vector Ui that finally exports F1 layer 721 and chooser system 725 from function blocks 74 at each.In addition, in following (4) (5) (6) (7) formula, " i " expression project data number, " 0 " expression F1 layer.
At first, according to (4) formula, i calculates Wi0 according to the input data I.Here, a is constant.
[mathematical expression 4]
Wi0＝Ii+aUi0 …(4)
Then, use Xi0 after (5) formula is calculated the Wi0 standardization that makes (4) formula.Here, W0 is with the value after the Wi0 norm.
[mathematical expression 5]
Xi0＝Wi0/W0 …(5)
Then, use (6) formula to calculate and from Xi0, removed the Vi0 behind the noise.Wherein, θ is for the constant of removing noise.According to the calculating of (6) formula, because small value becomes 0, so the noise of input data is removed.
[mathematical expression 6]
Vi0=f (Xi0)=Xi0 (when Xi0 〉=θ)
=0 (in above-mentioned situation in addition) ... (6)
At last, use (7) formula to obtain standardization input vector Ui0.Wherein, V0 is with the value after the Vi0 norm.The Ui0 that finally obtains is the input of F1 layer.
[mathematical expression 7]
Ui0＝Vi0/V0 …(7)
Fig. 4 (c) is the block diagram of the formation of expression F1 layer 722.In F1 layer 722, the Ui0 that will be obtained by (7) formula keeps in the mode of short-term storage, calculates the Pi that finally inputs to F2 layer 723.
The F2 layer is made of function blocks 75,76,77,78,79,80, manages throughout and carries out respectively following (8) (9) (10) (11) (12) (13) formula in the functional block 75,76,77,78,79,80.This calculating formula table of induction is shown (8)～(13) formula.Wherein, a, b are constant, and the f () function shown in (6) formula of serving as reasons, Tj are the grade of fit of being calculated by F2 layer 722.In addition, the denominator of (9) (11) (12) formula is the value after the norm.
[mathematical expression 8]
Wi＝Ui0+aUi …(8)
[mathematical expression 9]
Xi＝Wi/W …(9)
[mathematical expression 10]
Vi＝f(Xi)+bf(Qi)…(10)
[mathematical expression 11]
Ui＝Vi/V …(11)
[mathematical expression 12]
Qi＝Pi/P …(12)
[mathematical expression 13]
Pi＝Ui+∑g(yi)Zij
Wherein, g (yi)=d (Tj=max (Tj))
=0 (situation beyond above-mentioned) ... (13)
The process variable that utilizes the algorithm of the ART module 720 of Fig. 4 will input plant equipment carries out the results are shown among Fig. 5 of category classification.Fig. 5 (a) is the curve map of presentation class result's a example.Demonstrate as an example 2 projects in the measurement data among this figure, carry out mark with 2 dimension curve figure.The longitudinal axis and transverse axis standardize to represent to the measurement data of each project.
Measurement data is divided into a plurality of classifications 750 (circle shown in Fig. 5 (a)) by the ART module 720 of Fig. 4 (a).
In this figure, represent for the measurement data of 2 projects curve maps with 2 dimensions, but be not limited thereto, also can for 3 more than the project measurement data and carry out making of classification with the coordinate of multidimensional.
The classification results of Fig. 5 (a) is to obtain in taxon 400 under the normal condition mode of learning of the Fig. 2 (a) that illustrates before, and this result is accumulated in the classification results database 250.In addition, in state variation test section 520, obtaining under the diagnostic mode of Fig. 2 (b).
Time dependent example when Fig. 5 (b) illustrates the measuring-signal 1 that gets access to from plant equipment 100 and changes according to unusual generation.Transverse axis is got the time, and the longitudinal axis is got the generation ratio (generation frequency) of measuring-signal, classification numbering and new classification.Data D1 and D2 correspond respectively to project A and B.
In this figure, initial items A and B roughly are stabilized in certain value, reduce at the tight earlier data D1 of moment t1 (project A), then increase at the tight back of moment t1 data D2 (project B).Then, data D2 (project B) reduces, and final data D1 (project A) and data D2 (project B) increase.
In addition, before due in t1, the classification that is classified is numbered 1～4, classification, normal category when this is benchmark.With respect to this, after having passed through moment t1, the classification of project A and B is numbered 5～7, and the new classification that changes has occured the state that becomes the expression plant equipment.
Follow in this, increase through the generation ratio (generation frequency) that begins new classification after the moment t1 from firm, be diagnosed as state until surpass threshold value variation has occured.Here, the generation ratio of new classification uses the moving average of the number of the new classification that produces within specified time limit to calculate.
Such state variation is detected by state variation test section 520 under the diagnostic mode of Fig. 2 (b).In state variation test section 520, the moving average that can be chosen in the number of the new classification that produces in specified time limit is diagnosed as when having surpassed threshold value that the method that changes has occured state or according to having occured at the state that is diagnosed as in the situation of the classification that produces for new classification to change for being diagnosed as one of them method among the method that mode that state do not change diagnoses each sample in the situation of normal category.
Fig. 6 (a) is for the picture of setting the image display device 940 data items, Fig. 1 that is extracted by deal with data extraction unit 300.Purpose (want detect unusual content) by diagnosis decides diagnostic data items.
This picture 940 can roll and enclosed label (label of demonstration group 1 and group 2 in legend) in direction in length and breadth.On picture, the process variable of plant equipment (data items) A, B, C, D etc. show with its process numbering PID.In addition, the maximum, minimum value of each process variable (data items) etc. have been shown, while operating personnel judge that the purpose (want detect unusual content) of the relevant or diagnosis between each process variable etc. selects to want the combination of the process variable that divides into groups to manage.Represent that in the label of the group 1 of figure process variable (data items) A, C, D have each other relation and navigate to the group that monitor and select.
Fig. 6 (b) is the block diagram of the formation of presentation class unit 400, has disposed data pre-processing device and ART module that Fig. 4 (a) illustrated for the process variable in the group of being set by Fig. 6 (a).That is to say, in taxon 400, with the group quantity correspondingly, a succession of processing in executing data pretreating device and the ART module.
Fig. 7 (a), Fig. 7 (b), Fig. 7 (c) are illustrated in respectively the form of the data of preserving in measuring-signal database 230, reference signal database 240 and the classification results database 250 of Fig. 1.Also these figure can be thought the display frame of the image display device 940 of Fig. 1.In addition, in picture 940, the scroll bar of direction and be carried out as required the point that label shows is identical in Fig. 6 (a) or other pictures in length and breadth.
Shown in Fig. 7 (a), in measuring-signal database 230, preserve the value of a plurality of data items (project A, B, C etc.) that measured by plant equipment 100 by each sampling period (moment of the longitudinal axis).In display frame 940, on the longitudinal axis, show the value of these data items (project A, B, C etc.) by the time sequence, but thereby the data that the scroll box 56a by using cross shifting and 56b can the roll display wide regions.
In addition, in the situation of the reference signal database 240 of Fig. 7 (b), mark 57a, the 57b of the data sheet by selecting expression benchmark 1～2, thus can only gather the project that shows by the benchmark classification.
In the deal with data extraction unit 300 of Fig. 1, among measuring-signal database 230, be extracted in employed data group the diagnosis of plant equipment 100.For example, the data group that shows reference signal in Fig. 7 (b) is the state that 2 (" benchmark 1 ", " benchmark 2 ") and the data group when having selected " benchmark 1 " are made of project A, project C and project D.The data corresponding with the data items of the group 1 of being set by Fig. 6 (a) are benchmark 1, and the data corresponding with the data items of group 2 are benchmark 2.
Like this, in the measuring-signal database 230 shown in Fig. 7 (a) like that the measured value of total data project preserve in the seasonal effect in time series mode as 1 data group, and in the reference signal database 240 as Fig. 7 (b) shown in the measured value of such data items that is extracted by deal with data extraction unit 300 preserve in the seasonal effect in time series mode as a plurality of data groups.
And then Fig. 7 (c) is the display frame that is illustrated in the form of the data of preserving in the classification results database 250 of Fig. 1.Relation between the classification numbering of Fig. 7 (c) left side demonstration moment and the data of inscribing at this moment and classification, the relation between Fig. 7 (c) right side Display Category numbering and weight coefficient.Like this, in classification results database 250, preserved the classification results of each data group that reference signal database 240 preserves.
Fig. 8 (a) is the block diagram that consists of abnormality juding section 510.In abnormality juding section 510, judge with the operating condition variation result of determination 17 that the timeliness deterioration judging result 16 in the state variation result of determination 15 in the state variation test section 520, the timeliness degradation section 530, operating condition change in the test section 540 whether plant equipment has produced unusually.
As Figure 13 concludes, be in the situation that state variation result of determination 15 has occured to change at the state of being judged to be " 1 ", in the situation that the state that is judged to be does not change, be " 0 ".Timeliness deterioration judging result 16 has broken away from the state variation of plant equipment and has been " 1 " in the deteriorated scope of timeliness, is " 0 " in the deteriorated scope of timeliness.Operating condition changes result of determination 17 and is " 1 " at operating condition in not changing, and is " 0 " in then operating condition has occured to change.
The result who obtains with door 514 who these digital signals is inputed to Fig. 8 (a) becomes abnormality juding result 10.That is, change and the state variation of plant equipment has broken away from the deteriorated scope of timeliness and operating condition do not change at the state of plant equipment, abnormality juding result 10 be made as " 1 ", be judged to be produced unusual.Under condition in addition, abnormality juding result 10 becomes " 0 ".
Fig. 8 (b) is the figure that demonstrates the form of the data of preserving at picture 940 in diagnostic result database 260.State variation result of determination, operating condition variation result of determination, timeliness deterioration judging result, abnormality juding result are preserved in the seasonal effect in time series mode respectively, and are shown on the picture.In addition, in diagnostic result database 260, also preserved the moment shown in the Fig. 7 (c) that obtains as the result who makes 520 actions of state variation test section and this time data of inscribing and classification the classification numbering between relation and classification is numbered and weight coefficient between relation.
Fig. 8 (c) has shown in the seasonal effect in time series mode that by image display device 940 picture disply when the diagnostic result 12 of preserving is routine in diagnostic result database 260.Can be shown in Fig. 8 (b) like that with the result of numerical value mode displaying time sequence, and can be shown in Fig. 8 (c) like that with the result of the mode displaying time sequence of curve map.In addition, not only show " 0 ", " 1 ", also can show the moving average of certain specified time limit.In Fig. 8 (c), the moving average of curve 551 expression state variation result of determination, curve 552 expression operating conditions change the moving average of result of determination, curve 553 expression timeliness deterioration judging results' moving average, curve 554 expression abnormality juding results' moving average.
Fig. 9 (a) is the accompanying drawing of the action of explanation timeliness degradation section 530.Shown in the upper left curve map of Fig. 9 (a), measured value changes gradually in the deteriorated situation of timeliness.That is to say that if transverse axis represents the time, the longitudinal axis represents the size of process variable, then project A, B show the trend of slow variation in the situation that timeliness changes.On the other hand, shown in upper right curve map, measured value changes with breaking away from sometimes the deteriorated scope of timeliness, and be the situation that expression abnormal or operating condition change this moment mostly.In addition, about upper right curve map, also be that transverse axis represents the time, the longitudinal axis represents the size of process variable.
In addition, also can judge the identification of following the above-mentioned time to change with the variation delta z of the weight coefficient of the variable quantity of measured value, classification.Be correlated with owing between these indexs, exist, so can judge that timeliness is deteriorated by enough Δ z.The curve map of the bottom of Fig. 9 (a) all is that transverse axis is got project A, and the longitudinal axis is got project B, and has put down in writing the variation delta z of the weight coefficient of the classification between the different moment.In the timeliness in left side changed, the variation delta z of the weight coefficient of classification was showed littlely, and in the situation that measured value changes with breaking away from the deteriorated scope of timeliness, the variation delta z of the weight coefficient of classification is showed greatly.
In addition, even if when normal condition, the timeliness that is accompanied by plant equipment is deteriorated, and the trend of data also can change.If the intensity of variation of the data trend during pre-recorded normal condition is in this scope then be judged to be in the deteriorated scope of timeliness.
Fig. 9 (b) is the figure that is illustrated as the deteriorated scope of timeliness or breaks away from the determinating reference of the deteriorated scope of timeliness.In the figure, transverse axis is got the time amplitude of variation, the heavy index variation amount of longitudinal axis weighting, the maximal value that will obtain under the normal condition mode of learning is prepared as the judgement line of benchmark, according to actual measurement to time amplitude of variation and weight coefficient variable quantity Relations Among whether broken away from the border and determined whether in the scope that timeliness changes.
In the present embodiment, the time amplitude of variation when having summed up normal condition and the result of weight coefficient variable quantity Relations Among judge.Here, use following (14) formula is calculated the weight coefficient variable quantity with respect to the time amplitude of variation.
[mathematical expression 14]
Max(∑(Wi(t+Δt)-Wi(t) 2) …(14)
In (14) formula, Δ t is the time amplitude of variation, and i is for being used for the symbol of recognition data project, 1≤i≤n (n is the data items sum) wherein, and Wi (t) be the weight coefficient of the data items i under the moment t.
In addition, under diagnostic mode, in the peaked situation of the variable quantity of the weight coefficient when having surpassed normal condition, be judged to be and broken away from the deteriorated scope of timeliness.In addition, in the present embodiment, the maximal value of the variable quantity of the weight coefficient during with normal condition is judged to be threshold value, but the mean value of the variable quantity of the weight coefficient also can be with normal condition the time is set as threshold value.
Figure 10 (a) is the accompanying drawing that the explanation operating condition changes the 1st embodiment of test section 540.In the present embodiment, the ART that carries out Fig. 4 identical with the processing of implementing in taxon 400, state variation test section 520 processes (processing of data pre-processing device 710c and ART module 720d).Wherein, this moment is to the process variable of the input of ART and non-mechanical devices but the operating condition data are different in this.That is to say, utilize and the data of the group that only is made of the operating condition data to be carried out the sorted result of ART judge whether operating condition variation has occured.As the candidate of operating condition data and enumerate the external environment factor that " generation power ", " temperature " etc. and instrument characteristics do not have direct relation.
Producing in the situation of new classification in ART classification operating condition data, the possibility that operating condition changes is high.On the contrary, do not producing new classification in ART classification operating condition data, producing in the category measurement data in the situation of new classification, the reason of state variation is not the variation of operating condition.Can judge so whether operating condition variation has occured.
Figure 10 (b) is the figure that the explanation operating condition changes the 2nd embodiment of test section 540.
Also use ART here.In ART, shown in Figure 10 (b), the longitudinal axis and transverse axis represent respectively project A and project B, and the Data classification during with benchmark becomes several normal category.In this action, in the situation that has produced new classification, calculating becomes the data of object and the similar degree between the normal category, extracts the most similar (similar degree is maximum) normal category.
Similar degree S for example uses following (15) formula to calculate, and extracts (similar degree is minimum) normal category of S minimum.
[mathematical expression 15]
Sj＝∑(Di-Wij) 2 …(15)
Here, i is the symbol for the recognition data project, wherein 1≤i≤n (n is the data items sum).In addition, j is for being used for other symbol of recognition category, wherein 1≤j≤m (m is the sum of normal category).And Sj is similar degree, and Di is the value of data items i that becomes the data of object, and Wij is the weight coefficient of the data items i among the classification j.
Then, for example use following (16) formula to calculate the contribution degree Ci of each data items.
[mathematical expression 16]
Ci＝|Di-Wij| …(16)
Because the higher data items of contribution degree more separates with normal category, thereby can be described as the data items that becomes the reason that produces new classification.Be in the situation of operating condition data at this data items, it is that variation has occured operating condition that the reason that state changes is judged as.
In the example shown in Figure 10 (b), to observe for the position on the two-dimensional coordinate of normal category, the variation of what degree has occured in new classification coordinate at the longitudinal axis, transverse axis.The variation of transverse axis (contribution degree) is larger than the variation of the longitudinal axis in the figure, and the contribution degree of hence one can see that project A is large.
Figure 10 (c) is the figure that the explanation operating condition changes the 3rd embodiment of test section 540.In the present embodiment, made up the function that illustrated among Figure 10 (a), Figure 10 (b).
Contribution degree result 543 after using the result 542 after Figure 10 (a) judges and using Figure 10 (b) judgement is input to operating condition and changes detection unit 544.Change in the detection unit 544 at operating condition, implement the digital processing of AND or OR, the output operating condition changes result of determination 545.
Below, the action when diagnostic device 200 of the present invention is applied to thermal power generation machinery equipment is described.Figure 11 is the block diagram of expression thermal power generation machinery equipment.
In this figure, thermal power generation machinery equipment 100 comprises gas turbo-generator 110, control device 120 and data sending device 130.Gas turbo-generator 110 comprises generator 111, compressor 112, burner 113 and turbine 114.
In when generating, the air that is sucked into compressor 112 compressed be used as pressurized air, this pressurized air is delivered to burner 113, with fuel mixing after-combustion.Make turbine 114 rotations with the gases at high pressure that produce by burning, and generate electricity by generator 111.
In control device 120, control the output of gas turbo-generator 110 according to electricity needs.In addition, control device 120 will be arranged at service data 102 that the sensor (not shown) of gas turbo-generator 110 measures as the input data.Service data 102 is the quantity of states such as suction temperature, fuel input amount, turbine exhaust gas temperature, turbine rotation number, generator power output, turboshaft vibration, measures within each sampling period.In addition, also measure the weather informations such as atmospheric temperature.
In control device 120, calculate for the control signal 101 of controlling gas turbo-generator 110 with these service datas 102.
Signal data dispensing device 130 will comprise the service data 102 that measured by control device 120 and the measuring-signal 1 of the control signal 101 calculated by control device 120 is sent to diagnostic device 200.
Figure 12 is the result after diagnostic device 200 is diagnosed is used in explanation as object take the thermal power generation machinery equipment of narrating in Figure 11 accompanying drawing.
Figure 12 is the result when selecting generator output a, atmospheric temperature b as the data items of group 1 operating condition data and having selected generator output a, atmospheric temperature b, fuel flow rate c as the data items of group 2.
At moment T 0-T 1Between, as normal data, be used under mode of learning, learning normal condition.Even if under normal condition, Efficiency Decreasing has increased so obtain the required fuel flow rate c of identical data a because timeliness changes.
From moment T 1Work the diagnosis of having brought into use diagnostic device 200.At moment T 1-T 2Although between be normal condition, the impact that changes owing to timeliness causes fuel flow rate c to increase.Its result because the value of fuel flow rate c changes, causes producing new classification when making 520 action of state variation test section.Therefore, at process moment T 1Afterwards, state variation result of determination d moves closer to 1.In addition, in timeliness degradation section 530, because the increase ratio of fuel flow rate c is in the scope in when study, so be judged to be in the deteriorated scope of timeliness.More than, being judged to be in abnormality juding section 510 does not have abnormal.
At moment T 2-T 3Between, the temporary rising of atmospheric temperature b.If the then Efficiency Decreasing because atmospheric temperature b rises then increases in order to obtain the required fuel flow rate c of identical output a.Its result produces new classification when making 520 action of state variation test section.In addition, in timeliness degradation section 530, the scope when the increase ratio that is judged to be fuel flow rate c has broken away from study.In operating condition changes test section 540, because variation has occured atmospheric temperature b, thereby be judged to be operating condition f and change.Based on more than, being judged to be in abnormality juding section 510 does not have abnormal.
After atmospheric temperature reduces, at moment T 4Abnormal, fuel flow rate c increases together with it.Do not change because the increase ratio of fuel flow rate c breaks away from scope and the operating condition f of the deteriorated e of timeliness yet, thus in abnormality juding section 510, be judged to be produced unusual.
Implement in the result who only uses state variation test section 520 in the situation of abnormality juding, at moment T 1Can diagnose later on to produce and give birth to unusually.This is wrong report.The timeliness degradation section 530 of narrating in the application of the invention, operating condition change test section 540, thereby can be judged to be constantly T 1-T 4Scope be normal, can cut down rate of false alarm.
Like this, the diagnostic device 200 of the application of the invention has obtained to lower the effect of rate of false alarm.In addition, the data trend causes of change belongs to that timeliness is deteriorated by providing to operating personnel, operating condition changes, unusual which kind of situation, thereby helps the maintenance plan of plant equipment to draw up a plan for.
In addition, although in diagnosis unit 500 of the present invention, possessed abnormality juding section 510, state variation test section 520, timeliness degradation section 530, operating condition variation test section 540, but also can only possess the wherein side that timeliness degradation section 530 or operating condition change test section 540, wherein a kind of situation of the variation of or operating condition deteriorated for detection of timeliness.
In addition, can with more than carried out diagram and the item that illustrated consists of as diagnostic device in instructions, with computing machine it is implemented as diagnostic method, perhaps a series of program is embodied in program.
Utilizability on the industry
The present invention can be widely used in various plant equipment etc. as diagnostic device, diagnostic method, the diagnostic routine of plant equipment etc.
Symbol description
100: plant equipment
200: diagnostic device
210: outer input interface
220: outside output interface
230: the measuring-signal database
240: the reference signal database
250: the classification results database
260: the diagnostic result database
300: the deal with data extraction unit
400: taxon
500: diagnosis unit
510: abnormality juding section
520: the state variation test section
530: timeliness degradation section
540: operating condition changes test section
600: the alarm generation unit
900: the running management chamber
910: external input device
920: keyboard
930: mouse
940: image display device
Claims (according to the modification of the 19th of treaty)
1. the diagnostic device of a plant equipment, the unusual and report of surveying the diagnosis object plant equipment according to the measuring-signal of diagnosis object plant equipment is unusual, and the diagnostic device of this plant equipment is characterised in that, comprising:
The state variation test section, its measuring-signal with described diagnosis object plant equipment is categorized into classification and stores, and is judged to be state in the situation that does not belong to the classification of classifying variation has occured;
Abnormality juding section, it will be judged to be as a part of inputting from the signal of this state variation test section and produce in described diagnosis object plant equipment unusually;
The alarm generation unit, its output report with this abnormality juding section arrives outside; With
The wrong report blocking portion, the event that it is surveyed beyond described diagnosis object plant equipment unusual stops the output of described abnormality juding section.
2. the diagnostic device of (after revising) plant equipment according to claim 1 is characterized in that,
Described wrong report blocking portion comprises timeliness degradation section, this timeliness degradation section obtains the weight coefficient corresponding with classification with the measuring-signal of described diagnosis object plant equipment, surpassed to be judged to be in the peaked situation at the variable quantity of the weight coefficient corresponding with classification and broken away from the deteriorated scope of timeliness
This timeliness degradation section stops the output of described abnormality juding section when not breaking away from the deteriorated scope of timeliness.
3. the diagnostic device of plant equipment according to claim 1 is characterized in that,
Described wrong report blocking portion comprises that operating condition changes test section, this operating condition variation test section becomes classification to store the operating condition Data classification of described diagnosis object plant equipment, being judged to be operating condition in the situation that does not produce new classification does not change
This operating condition changes test section and at operating condition the output of described abnormality juding section has occured to stop when changing.
4. the diagnostic device of plant equipment according to claim 3 is characterized in that,
Change the operating condition data of the input of test section as operating condition, do not have the external environment factor of direct relation to consist of by the characteristic with the apparatus that consists of described diagnosis object plant equipment.
5. the diagnostic device of (revise after) a kind of plant equipment, the unusual and report of surveying diagnosis object according to the measuring-signal of diagnosis object plant equipment is unusual, and the diagnostic device of described plant equipment is characterised in that, comprising:
The state variation test section, its measuring-signal with described diagnosis object plant equipment is categorized into classification and stores, and is judged to be state in the situation that does not belong to the classification of classifying variation has occured;
Timeliness degradation section, its measuring-signal with the diagnosis object plant equipment is obtained the weight coefficient corresponding with classification, has surpassed to be judged to be in the peaked situation at the variable quantity of the weight coefficient corresponding with classification to have broken away from the deteriorated scope of timeliness;
Operating condition changes test section, and its operating condition Data classification with the diagnosis object plant equipment becomes classification to store, and is judged to be operating condition and does not change in the situation that does not produce new classification;
Abnormality juding section, variation has occured and has been judged to be to have broken away from the deteriorated scope of timeliness and changed test section by described operating condition by described timeliness degradation section to be judged to be in operating condition do not change at the state that is judged to be by described state variation test section in it, is judged to be described diagnosis object and has produced unusually; With
The alarm generation unit, its output report with this abnormality juding section arrives outside.
6. (revise afterwards) a kind of diagnostic device of plant equipment, possess:
The measuring-signal database, it preserves the measuring-signal of diagnosis object plant equipment;
The deal with data extraction unit, it extracts the diagnostic signal that uses for the state of diagnosing described diagnosis object plant equipment among described measuring-signal database;
The reference signal database, it preserves described diagnostic signal;
Taxon, its Data classification that will preserve in described reference signal database becomes classification;
The classification results database, it is preserved described classification as normal category;
Diagnosis unit, it uses the information of the up-to-date described diagnostic signal that is extracted by described deal with data extraction unit and the described normal category of preserving in described classification results database, diagnose the state of described diagnosis object plant equipment to belong to normally, unusual, operating condition variation, which kind of deteriorated situation of timeliness;
The diagnostic result database, it preserves the diagnostic result of described diagnosis unit; With
The unit of the information of in described diagnostic result database, preserving to image display device output,
The diagnostic device of described plant equipment is characterised in that,
Possess abnormality juding section, state variation test section, timeliness degradation section and operating condition in the described diagnosis unit and change test section,
Described state variation test section possesses following function: do not belong at the up-to-date described diagnostic signal that is extracted by described deal with data extraction unit and is judged to be state in the situation of the described normal category of in described classification results database, preserving variation has occured,
Described timeliness degradation section possesses following function: the relation between the time amplitude of variation when deriving normal condition and the weight coefficient variable quantity corresponding with classification, be judged to be in the peaked situation of the variable quantity of the weight coefficient corresponding with classification when in diagnostic procedure, having surpassed normal condition and broken away from the deteriorated scope of timeliness
Described operating condition changes test section and possesses following function: will not produce in the situation of new classification in the classification by becoming with operating condition Data classification that instrument characteristics does not have the external environment factor of direct relation to consist of, being judged to be operating condition does not change
In described abnormality juding section, variation having occured and be judged to be to have broken away from the deteriorated scope of timeliness and changed test section by described operating condition by described timeliness degradation section not to be judged to be in operating condition do not change at the state that is judged to be by described state variation test section, has been judged to be unusually.
7. the diagnostic method of (revise after) a kind of plant equipment, the unusual and report of surveying the diagnosis object plant equipment according to the measuring-signal of diagnosis object plant equipment is unusual, and the diagnostic method of described plant equipment is characterised in that,
The measuring-signal of described diagnosis object plant equipment is categorized into classification to be stored, in the situation that does not belong to the classification of classifying, be judged to be state variation has occured, obtain the weight coefficient corresponding with classification with the measuring-signal of diagnosis object plant equipment, surpassed to be judged to be in the peaked situation at the variable quantity of the weight coefficient corresponding with classification and broken away from the deteriorated scope of timeliness, become classification to store the operating condition Data classification of diagnosis object, being judged to be operating condition in the situation that does not produce new classification does not change, at this moment, be judged to be and produced unusually in the described diagnosis object plant equipment, and report is to outside.
8. the diagnostic routine of a plant equipment, the unusual and report of surveying the diagnosis object plant equipment according to the measuring-signal of diagnosis object plant equipment is unusual, and the diagnostic routine of described plant equipment is characterised in that, comprising:
The state variation detecting step is categorized into classification with the measuring-signal of described diagnosis object plant equipment and stores, and is judged to be state in the situation that does not belong to the classification of classifying variation has occured;
The abnormality juding step, will from the signal of this state variation detecting step be judged to be as the part of input in described diagnosis object, produced unusual;
Alarm produces step, with the output report of this abnormality juding step to outside; With
Wrong report stops step, surveys the unusual event in addition of described diagnosis object, stops the output of described abnormality juding step.
9. the diagnostic routine of (after revising) plant equipment according to claim 8 is characterized in that,
Described wrong report stops step to comprise timeliness degradation step, in this timeliness degradation step, obtain the weight coefficient corresponding with classification with the measuring-signal of described diagnosis object plant equipment, surpassed to be judged to be in the peaked situation at the variable quantity of the weight coefficient corresponding with classification and broken away from the deteriorated scope of timeliness
In this timeliness degradation step, when not breaking away from the deteriorated scope of timeliness, stop the output of described abnormality juding step.
10. the diagnostic routine of plant equipment according to claim 8 is characterized in that,
Described wrong report stops step to comprise that operating condition changes detecting step, operating condition Data classification with described diagnosis object plant equipment in this operating condition variation detecting step becomes classification to store, being judged to be operating condition in the situation that does not produce new classification does not change
Change in the detecting step at this operating condition, when variation has occured operating condition, stop the output of described abnormality juding step.
11. the diagnostic routine of plant equipment according to claim 10 is characterized in that,
Change the operating condition data of the input of detecting step as operating condition, do not have the external environment factor of direct relation to consist of by the characteristic with the apparatus that consists of described diagnosis object.
The diagnostic routine of (12. after revising) a kind of plant equipment, the unusual and report of surveying the diagnosis object plant equipment according to the measuring-signal of diagnosis object plant equipment is unusual, and the diagnostic routine of described plant equipment is characterised in that, comprising:
The state variation detecting step is categorized into classification with the measuring-signal of described diagnosis object plant equipment and stores, and is judged to be state in the situation that does not belong to the classification of classifying variation has occured;
Timeliness degradation step is obtained the weight coefficient corresponding with classification with the measuring-signal of diagnosis object plant equipment, has surpassed to be judged to be in the peaked situation at the variable quantity of the weight coefficient corresponding with classification to have broken away from the deteriorated scope of timeliness;
Operating condition changes detecting step, becomes classification to store the operating condition Data classification of diagnosis object plant equipment, is judged to be operating condition and does not change in the situation that does not produce new classification;
The abnormality juding step, variation having occured and be judged to be to have broken away from the deteriorated scope of timeliness and changed detecting step by described operating condition by described timeliness degradation step to be judged to be in operating condition do not change at the state that is judged to be by described state variation detecting step, is judged to be described diagnosis object and has produced unusually; With
Alarm produces step, with the output report of this abnormality juding step to outside.

Claims (12)

1. the diagnostic device of a plant equipment, the unusual and report of surveying the diagnosis object plant equipment according to the measuring-signal of diagnosis object plant equipment is unusual, and the diagnostic device of this plant equipment is characterised in that, comprising:
The state variation test section, its measuring-signal with described diagnosis object plant equipment is categorized into classification and stores, and is judged to be state in the situation that does not belong to the classification of classifying variation has occured;
Abnormality juding section, it will be judged to be as a part of inputting from the signal of this state variation test section and produce in described diagnosis object plant equipment unusually;
The alarm generation unit, its output report with this abnormality juding section arrives outside; With
The wrong report blocking portion, the event that it is surveyed beyond described diagnosis object plant equipment unusual stops the output of described abnormality juding section.
2. the diagnostic device of plant equipment according to claim 1 is characterized in that,
Described wrong report blocking portion comprises timeliness degradation section, this timeliness degradation section obtains weight coefficient with the measuring-signal of described diagnosis object plant equipment, surpassed to be judged to be in the peaked situation at the variable quantity of weight coefficient and broken away from the deteriorated scope of timeliness
This timeliness degradation section stops the output of described abnormality juding section when not breaking away from the deteriorated scope of timeliness.
3. the diagnostic device of plant equipment according to claim 1 is characterized in that,
Described wrong report blocking portion comprises that operating condition changes test section, this operating condition variation test section becomes classification to store the operating condition Data classification of described diagnosis object plant equipment, being judged to be operating condition in the situation that does not produce new classification does not change
This operating condition changes test section and at operating condition the output of described abnormality juding section has occured to stop when changing.
4. the diagnostic device of plant equipment according to claim 3 is characterized in that,
Change the operating condition data of the input of test section as operating condition, do not have the external environment factor of direct relation to consist of by the characteristic with the apparatus that consists of described diagnosis object plant equipment.
5. the diagnostic device of a plant equipment, the unusual and report of surveying diagnosis object according to the measuring-signal of diagnosis object plant equipment is unusual, and the diagnostic device of described plant equipment is characterised in that, comprising:
The state variation test section, its measuring-signal with described diagnosis object plant equipment is categorized into classification and stores, and is judged to be state in the situation that does not belong to the classification of classifying variation has occured;
Timeliness degradation section, its measuring-signal with the diagnosis object plant equipment is obtained weight coefficient, has surpassed to be judged to be in the peaked situation at the variable quantity of weight coefficient to have broken away from the deteriorated scope of timeliness;
Operating condition changes test section, and its operating condition Data classification with the diagnosis object plant equipment becomes classification to store, and is judged to be operating condition and does not change in the situation that does not produce new classification;
Abnormality juding section, variation has occured and has been judged to be to have broken away from the deteriorated scope of timeliness and changed test section by described operating condition by described timeliness degradation section to be judged to be in operating condition do not change at the state that is judged to be by described state variation test section in it, is judged to be described diagnosis object and has produced unusually; With
The alarm generation unit, its output report with this abnormality juding section arrives outside.
6. the diagnostic device of a plant equipment possesses:
The measuring-signal database, it preserves the measuring-signal of diagnosis object plant equipment;
The deal with data extraction unit, it extracts the diagnostic signal that uses for the state of diagnosing described diagnosis object plant equipment among described measuring-signal database;
The reference signal database, it preserves described diagnostic signal;
Taxon, its Data classification that will preserve in described reference signal database becomes classification;
The classification results database, it is preserved described classification as normal category;
Diagnosis unit, it uses the information of the up-to-date described diagnostic signal that is extracted by described deal with data extraction unit and the described normal category of preserving in described classification results database, diagnose the state of described diagnosis object plant equipment to belong to normally, unusual, operating condition variation, which kind of deteriorated situation of timeliness;
The diagnostic result database, it preserves the diagnostic result of described diagnosis unit; With
The unit of the information of in described diagnostic result database, preserving to image display device output,
The diagnostic device of described plant equipment is characterised in that,
Possess abnormality juding section, state variation test section, timeliness degradation section and operating condition in the described diagnosis unit and change test section,
Described state variation test section possesses following function: do not belong at the up-to-date described diagnostic signal that is extracted by described deal with data extraction unit and is judged to be state in the situation of the described normal category of in described classification results database, preserving variation has occured,
Described timeliness degradation section possesses following function: the time amplitude of variation when deriving normal condition and the relation between the weight coefficient variable quantity, be judged to be in the peaked situation of the variable quantity of the weight coefficient when in diagnostic procedure, having surpassed normal condition and broken away from the deteriorated scope of timeliness
Described operating condition changes test section and possesses following function: will not produce in the situation of new classification in the classification by becoming with operating condition Data classification that instrument characteristics does not have the external environment factor of direct relation to consist of, being judged to be operating condition does not change
In described abnormality juding section, variation having occured and be judged to be to have broken away from the deteriorated scope of timeliness and changed test section by described operating condition by described timeliness degradation section not to be judged to be in operating condition do not change at the state that is judged to be by described state variation test section, has been judged to be unusually.
7. the diagnostic method of a plant equipment, the unusual and report of surveying the diagnosis object plant equipment according to the measuring-signal of diagnosis object plant equipment is unusual, and the diagnostic method of described plant equipment is characterised in that,
The measuring-signal of described diagnosis object plant equipment is categorized into classification to be stored, in the situation that does not belong to the classification of classifying, be judged to be state variation has occured, measuring-signal with the diagnosis object plant equipment is obtained weight coefficient, surpassed to be judged to be in the peaked situation at the variable quantity of weight coefficient and broken away from the deteriorated scope of timeliness, become classification to store the operating condition Data classification of diagnosis object, being judged to be operating condition in the situation that does not produce new classification does not change, at this moment, be judged to be and produced unusually in the described diagnosis object plant equipment, and report is to outside.
8. the diagnostic routine of a plant equipment, the unusual and report of surveying the diagnosis object plant equipment according to the measuring-signal of diagnosis object plant equipment is unusual, and the diagnostic routine of described plant equipment is characterised in that, comprising:
The state variation detecting step is categorized into classification with the measuring-signal of described diagnosis object plant equipment and stores, and is judged to be state in the situation that does not belong to the classification of classifying variation has occured;
The abnormality juding step, will from the signal of this state variation detecting step be judged to be as the part of input in described diagnosis object, produced unusual;
Alarm produces step, with the output report of this abnormality juding step to outside; With
Wrong report stops step, surveys the unusual event in addition of described diagnosis object, stops the output of described abnormality juding step.
9. the diagnostic routine of plant equipment according to claim 8 is characterized in that,
Described wrong report stops step to comprise timeliness degradation step, in this timeliness degradation step, obtain weight coefficient with the measuring-signal of described diagnosis object plant equipment, surpassed to be judged to be in the peaked situation at the variable quantity of weight coefficient and broken away from the deteriorated scope of timeliness
In this timeliness degradation step, when not breaking away from the deteriorated scope of timeliness, stop the output of described abnormality juding step.
10. the diagnostic routine of plant equipment according to claim 8 is characterized in that,
Described wrong report stops step to comprise that operating condition changes detecting step, operating condition Data classification with described diagnosis object plant equipment in this operating condition variation detecting step becomes classification to store, being judged to be operating condition in the situation that does not produce new classification does not change
Change in the detecting step at this operating condition, when variation has occured operating condition, stop the output of described abnormality juding step.
11. the diagnostic routine of plant equipment according to claim 10 is characterized in that,
Change the operating condition data of the input of detecting step as operating condition, do not have the external environment factor of direct relation to consist of by the characteristic with the apparatus that consists of described diagnosis object.
12. the diagnostic routine of a plant equipment, the unusual and report of surveying the diagnosis object plant equipment according to the measuring-signal of diagnosis object plant equipment is unusual, and the diagnostic routine of described plant equipment is characterised in that, comprising:
The state variation detecting step is categorized into classification with the measuring-signal of described diagnosis object plant equipment and stores, and is judged to be state in the situation that does not belong to the classification of classifying variation has occured;
Timeliness degradation step is obtained weight coefficient with the measuring-signal of diagnosis object plant equipment, has surpassed to be judged to be in the peaked situation at the variable quantity of weight coefficient to have broken away from the deteriorated scope of timeliness;
Operating condition changes detecting step, becomes classification to store the operating condition Data classification of diagnosis object plant equipment, is judged to be operating condition and does not change in the situation that does not produce new classification;
The abnormality juding step, variation having occured and be judged to be to have broken away from the deteriorated scope of timeliness and changed detecting step by described operating condition by described timeliness degradation step to be judged to be in operating condition do not change at the state that is judged to be by described state variation detecting step, is judged to be described diagnosis object and has produced unusually; Produce step with alarm, with the output report of this abnormality juding step to outside.
CN201080065911.2A 2010-04-08 2010-04-08 Plant diagnosis device, diagnosis method, and diagnosis program Active CN102870057B (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2010/002575 WO2011125130A1 (en) 2010-04-08 2010-04-08 Plant diagnosis device, diagnosis method, and diagnosis program

Publications (2)

Publication Number Publication Date
CN102870057A true CN102870057A (en) 2013-01-09
CN102870057B CN102870057B (en) 2015-01-28

Family

ID=44762120

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201080065911.2A Active CN102870057B (en) 2010-04-08 2010-04-08 Plant diagnosis device, diagnosis method, and diagnosis program

Country Status (3)

Country Link
JP (1) JP5544418B2 (en)
CN (1) CN102870057B (en)
WO (1) WO2011125130A1 (en)

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105992977A (en) * 2013-11-28 2016-10-05 株式会社日立制作所 Diagnostic device and diagnostic method for plant
CN107024917A (en) * 2016-01-29 2017-08-08 发那科株式会社 It was found that the unit control apparatus the reason for exception of manufacture machinery
CN108369416A (en) * 2016-03-15 2018-08-03 株式会社日立制作所 Abnormity diagnostic system
CN108463736A (en) * 2016-01-20 2018-08-28 三菱电机株式会社 Abnormal detector and abnormality detection system
CN109839900A (en) * 2017-11-28 2019-06-04 发那科株式会社 Numerical control device
CN109997087A (en) * 2016-12-01 2019-07-09 住友重机械工业株式会社 Fault diagnosis system
CN110235010A (en) * 2017-02-06 2019-09-13 三菱日立电力系统株式会社 State analysis device, display methods and program
CN110543141A (en) * 2018-05-29 2019-12-06 发那科株式会社 Diagnostic device, diagnostic method, and diagnostic program
CN110823274A (en) * 2018-08-08 2020-02-21 Dmg森精机株式会社 Abnormal state determination system
CN111272454A (en) * 2018-12-04 2020-06-12 日立环球生活方案株式会社 Abnormality diagnosis device and abnormality diagnosis method
CN111542792A (en) * 2018-03-01 2020-08-14 株式会社日立制作所 Diagnostic device and diagnostic method
CN111751637A (en) * 2019-03-26 2020-10-09 株式会社日本综合研究所 Information processing apparatus and storage medium
CN112334924A (en) * 2018-07-10 2021-02-05 住友重机械工业株式会社 Display method for construction machine and support device for construction machine
CN112567306A (en) * 2018-08-31 2021-03-26 东芝三菱电机产业系统株式会社 Manufacturing process monitoring device
CN113168171A (en) * 2018-12-05 2021-07-23 三菱电机株式会社 Abnormality detection device and abnormality detection method
CN113757915A (en) * 2020-06-03 2021-12-07 日立环球生活方案株式会社 Abnormality diagnosis device and abnormality diagnosis method
CN114391093A (en) * 2019-09-17 2022-04-22 日产自动车株式会社 Abnormality determination device and abnormality determination method

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6366520B2 (en) * 2015-02-02 2018-08-01 日立Ｇｅニュークリア・エナジー株式会社 Output fluctuation monitoring apparatus and method
KR101876185B1 (en) 2016-08-29 2018-07-09 한국수력원자력 주식회사 Learning method on recent data considering external effect in early alarm system, and system using thereof
JP6810097B2 (en) 2018-05-21 2021-01-06 ファナック株式会社 Anomaly detector
WO2020144747A1 (en) * 2019-01-08 2020-07-16 眞鍋造機株式会社 Abnormality detection device, crane system, and abnormality detection method
JP7260069B1 (en) * 2022-03-29 2023-04-18 三菱電機株式会社 Anomaly detection device, mechanical system and anomaly detection method
JPWO2023199567A1 (en) * 2022-04-12 2023-10-19
JP7485473B1 (en) 2023-03-22 2024-05-16 東芝エネルギーシステムズ株式会社 Generator cooler performance monitoring device

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH04186119A (en) * 1990-11-21 1992-07-02 Hitachi Ltd Method and device for monitoring abnormality
JPH10301619A (en) * 1997-04-23 1998-11-13 Toshiba Corp Diagnostic device for fuel battery power generation plant
JP2005165375A (en) * 2003-11-28 2005-06-23 Hitachi Ltd Diagnostic device and method for facility
CN101118694A (en) * 2007-05-18 2008-02-06 李克明 Vehicle intellectualized management system

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0660280A (en) * 1992-08-10 1994-03-04 Mitsubishi Electric Corp Transformation equipment monitor
JP3745783B2 (en) * 1994-06-08 2006-02-15 株式会社アドイン研究所 Pattern classification device and state monitoring system using this device
JP4140634B2 (en) * 2006-01-26 2008-08-27 松下電工株式会社 Anomaly monitoring method and anomaly monitoring apparatus
JP4767039B2 (en) * 2006-02-28 2011-09-07 株式会社フジクラ Vibration detection sensor system and vibration detection method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH04186119A (en) * 1990-11-21 1992-07-02 Hitachi Ltd Method and device for monitoring abnormality
JPH10301619A (en) * 1997-04-23 1998-11-13 Toshiba Corp Diagnostic device for fuel battery power generation plant
JP2005165375A (en) * 2003-11-28 2005-06-23 Hitachi Ltd Diagnostic device and method for facility
CN101118694A (en) * 2007-05-18 2008-02-06 李克明 Vehicle intellectualized management system

Cited By (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105992977A (en) * 2013-11-28 2016-10-05 株式会社日立制作所 Diagnostic device and diagnostic method for plant
CN105992977B (en) * 2013-11-28 2018-12-14 株式会社日立制作所 The diagnostic device and diagnostic method of complete set of equipments
CN108463736A (en) * 2016-01-20 2018-08-28 三菱电机株式会社 Abnormal detector and abnormality detection system
CN108463736B (en) * 2016-01-20 2020-06-30 三菱电机株式会社 Abnormality detection device and abnormality detection system
US10203666B2 (en) 2016-01-29 2019-02-12 Fanuc Corporation Cell controller for finding cause of abnormality in manufacturing machine
CN107024917B (en) * 2016-01-29 2018-08-03 发那科株式会社 It was found that the unit control apparatus of the abnormal reason of manufacture machinery
CN107024917A (en) * 2016-01-29 2017-08-08 发那科株式会社 It was found that the unit control apparatus the reason for exception of manufacture machinery
US10976731B2 (en) 2016-03-15 2021-04-13 Hitachi, Ltd. Abnormality diagnostic system
CN108369416B (en) * 2016-03-15 2021-07-16 株式会社日立制作所 Abnormality diagnosis system
CN108369416A (en) * 2016-03-15 2018-08-03 株式会社日立制作所 Abnormity diagnostic system
CN109997087A (en) * 2016-12-01 2019-07-09 住友重机械工业株式会社 Fault diagnosis system
CN110235010A (en) * 2017-02-06 2019-09-13 三菱日立电力系统株式会社 State analysis device, display methods and program
CN110235010B (en) * 2017-02-06 2021-10-15 三菱动力株式会社 State analysis device, display method, and storage medium
CN109839900A (en) * 2017-11-28 2019-06-04 发那科株式会社 Numerical control device
CN109839900B (en) * 2017-11-28 2021-04-20 发那科株式会社 Numerical controller
CN111542792A (en) * 2018-03-01 2020-08-14 株式会社日立制作所 Diagnostic device and diagnostic method
CN110543141A (en) * 2018-05-29 2019-12-06 发那科株式会社 Diagnostic device, diagnostic method, and diagnostic program
CN112334924A (en) * 2018-07-10 2021-02-05 住友重机械工业株式会社 Display method for construction machine and support device for construction machine
CN110823274A (en) * 2018-08-08 2020-02-21 Dmg森精机株式会社 Abnormal state determination system
CN112567306A (en) * 2018-08-31 2021-03-26 东芝三菱电机产业系统株式会社 Manufacturing process monitoring device
CN111272454A (en) * 2018-12-04 2020-06-12 日立环球生活方案株式会社 Abnormality diagnosis device and abnormality diagnosis method
CN111272454B (en) * 2018-12-04 2022-03-29 日立环球生活方案株式会社 Abnormality diagnosis device and abnormality diagnosis method
CN113168171A (en) * 2018-12-05 2021-07-23 三菱电机株式会社 Abnormality detection device and abnormality detection method
CN113168171B (en) * 2018-12-05 2023-09-19 三菱电机株式会社 Abnormality detection device and abnormality detection method
CN111751637A (en) * 2019-03-26 2020-10-09 株式会社日本综合研究所 Information processing apparatus and storage medium
CN114391093A (en) * 2019-09-17 2022-04-22 日产自动车株式会社 Abnormality determination device and abnormality determination method
CN114391093B (en) * 2019-09-17 2023-10-24 日产自动车株式会社 Abnormality determination device and abnormality determination method
CN113757915A (en) * 2020-06-03 2021-12-07 日立环球生活方案株式会社 Abnormality diagnosis device and abnormality diagnosis method
CN113757915B (en) * 2020-06-03 2022-12-06 日立环球生活方案株式会社 Abnormality diagnosis device and abnormality diagnosis method

Also Published As

Publication number Publication date
JPWO2011125130A1 (en) 2013-07-08
WO2011125130A1 (en) 2011-10-13
CN102870057B (en) 2015-01-28
JP5544418B2 (en) 2014-07-09

Similar Documents

Publication Publication Date Title
CN102870057B (en) Plant diagnosis device, diagnosis method, and diagnosis program
CN102576227B (en) Plant diagnostic equipment
JP5530019B1 (en) Abnormal sign detection system and abnormality sign detection method
KR101995042B1 (en) Gas turbine failure prediction utilizing supervised learning methodologies
KR102618023B1 (en) Failure prediction diagnosis system and method through pattern analysis according to failure type
JP5331774B2 (en) Equipment state monitoring method and apparatus, and equipment state monitoring program
US7181365B2 (en) Diagnostic data detection and control
US7096074B2 (en) Methods and apparatus for early fault detection and alert generation in a process
JP5484591B2 (en) PLANT DIAGNOSIS DEVICE AND PLANT DIAGNOSIS METHOD
JP5292477B2 (en) Diagnostic device and diagnostic method
JP6200833B2 (en) Diagnostic equipment for plant and control equipment
CN104756029B (en) A kind of system of the parts group of monitoring device
CN104596780B (en) Diagnosis method for sensor faults of motor train unit braking system
US6636842B1 (en) System and method for controlling an industrial process utilizing process trajectories
Serdio et al. Fuzzy fault isolation using gradient information and quality criteria from system identification models
JP4430384B2 (en) Equipment diagnostic apparatus and diagnostic method
TWI780434B (en) Abnormal diagnosis device and method
JP6830414B2 (en) Diagnostic device and diagnostic method
CN116034321A (en) Interpreting machine learning output in industrial applications
JP6685124B2 (en) Diagnostic device and diagnostic method
Jang et al. Anomaly detection of 2.4 l diesel engine using one-class svm with variational autoencoder
Purarjomandlangrudi et al. Application of anomaly technique in wind turbine bearing fault detection
JP6951202B2 (en) Data classification device
Richter et al. Automatic Defect Detection by Classifying Aggregated Vehicular Behavior
AO et al. An Ensemble Learning-Based Alert Trigger System for Predictive Maintenance of Assets with In-Situ Sensors

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