CN109240244A - Equipment running status health degree analysis method and system based on data-driven - Google Patents
Equipment running status health degree analysis method and system based on data-driven Download PDFInfo
- Publication number
- CN109240244A CN109240244A CN201811258951.3A CN201811258951A CN109240244A CN 109240244 A CN109240244 A CN 109240244A CN 201811258951 A CN201811258951 A CN 201811258951A CN 109240244 A CN109240244 A CN 109240244A
- Authority
- CN
- China
- Prior art keywords
- data
- equipment
- model
- measuring point
- operation frame
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM]
- G05B19/4184—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM] characterised by fault tolerance, reliability of production system
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/31—From computer integrated manufacturing till monitoring
- G05B2219/31457—Factory remote control, monitoring through internet
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/80—Management or planning
Abstract
The invention discloses the equipment running status health degree analysis method and system based on data-driven, comprising the following steps: S1: real-time watch device, according toPHMPrincipleP‑FInterval graph determines suitable device, and acquires suitable device measuring point data;S2: the measuring point data of acquisition is handled;S3: establishing device data correlation model, according to data characteristics model foundation equipment operation frame model in the device data correlation model;S4: being monitored according to equipment operation frame model, and the exception or failure for monitoring carry out early warning;S5: it is diagnosed and is handled according to early warning result.The invention has the advantages that: unit exception or the timely early warning of failure can not only be made, the time that will be broken down and position are predicted, targetedly carry out equipment maintenance and meets the rank data integration requirement of group of industrial enterprise.
Description
Technical field
The present invention relates to boiler for producing technical fields, it particularly relates to which a kind of equipment based on data-driven runs shape
State health degree analysis method and system.
Background technique
The unit exception monitoring method based on big data of comparative maturity mainly includes based on mechanism model, based on knowing at present
The method of knowledge and analysis model based on simple data, wherein
1) be based on mechanism model: this method is combined closely with control theory, passes through equipment operation mechanism founding mathematical models
Carry out Prediction System output, and by it compared with actual measured value, obtains residual error;Residual error is analyzed so that whether determination process is sent out
Further identification of defective type, wherein advantage: raw failure in conjunction with physical knowledge and monitoring system, is carried out by analysis residual error
Fault pre-alarming, it is readily understood;Disadvantage: most mechanism models are simplified linear system;It is often non-thread in actual industrial process
Property, freedom degree be higher, Multivariable Coupling system;Its using effect is not satisfactory.
2) Knowledge based engineering method: this method is based on the Heuristics of people, each list during qualitative or quantitative description
Connection relationship, fault propagation mode between member etc. were simulated after abnormal sign occurs in equipment by modes such as reasoning, deductions
Inferential capability of the Cheng expert in monitoring, to be automatically performed equipment fault early-warning and monitoring;Wherein, advantage: when monitored pair
As relatively simple, when process knowledge and knowhow are more sufficient, using effect is preferable;Disadvantage: expert is known in early warning accuracy
The height of the abundant degree and expertise level known has very strong dependence;Many experiences are difficult with a kind of reasonable form
Change expression way to be described.
3) it is analyzed based on simple data: fixed sum data screening foundation being set with equipment operating experience and artificial data weighting and is set
Standby mathematical model and expression process status realize that the fault pre-alarming of effective monitoring and the equipment operation of production process is based in this way
Mass data;Wherein, advantage: common simple fault pre-alarming can be monitored effectively, solved least a portion of fault pre-alarming and asked
Topic;Disadvantage: since data sample is very few, and equipment running status is extremely complex, the result is that failure accuracy is not high and early warning is imitated
Rate it is low.
For the problems in the relevant technologies, currently no effective solution has been proposed.
Summary of the invention
For above-mentioned technical problem in the related technology, the present invention proposes a kind of equipment running status based on data-driven
Health degree analysis method and system, are able to solve that industrial enterprise operates repeatedly or the dynamic unit exception or failure of periodic job are pre-
Police not in time, can not predict the time that will be broken down and position, it is difficult to equipment maintenance is targetedly carried out, and
The technical issues of being unable to satisfy the rank data integration requirement of group of industrial enterprise.
To realize the above-mentioned technical purpose, the technical scheme of the present invention is realized as follows:
A kind of equipment running status health degree analysis method based on data-driven, comprising the following steps:
S1: real-time watch device determines suitable device according to PHM principle P-F interval graph, and acquires suitable device measuring point
Data;
S2: the measuring point data of acquisition is handled;
S3: establishing device data correlation model, is set according to data characteristics model foundation in the device data correlation model
Received shipment row frame model;
S4: being monitored according to equipment operation frame model, and the exception or failure for monitoring carry out early warning;
S5: it is diagnosed and is handled according to early warning result.
Further, the measuring point data of acquisition processing is carried out in the step S2 to specifically include:
S21 analyzes measuring point data, and device data is classified by setting screening conditions;
Sorted measuring point data is normalized S22.
Further, the measuring point data of normalized is subjected to data normalization processing in the step S22.
It further, is σ by model result and measurement error meet mean value in data characteristics model in the step S3
Gaussian Profile, it is as follows as error function corrective data correlation degree model using quadratic sum:
In formula, x: the stochastic variable based on historical data;Y: another stochastic variable based on historical data;σ:
Standard deviation;μ: mathematic expectaion.
Further, in the step S3 equipment operation frame model using certainty factor algebra and support fitting algorithm into
Row optimization equipment operation frame model, obtains meeting the actual equipment operation frame of engineering.
Further, it is monitored and is specifically included according to operation frame model in the step S4:
Compared with S41 carries out measuring point data repeatedly with equipment operation frame model, is found using certainty factor algebra and meet work
The actual operation frame of journey;
S42 calculates all measuring point datas using Euclidean distance algorithm and meets the matching of the actual equipment operation frame of engineering
Situation;
S43 is compared according to device history moving law and current matching situation, the current health degree of analytical equipment, will
The equipment health degree of calculating is shown.
Another aspect of the present invention provides a kind of equipment running status health degree analysis system based on data-driven, packet
It includes:
Data acquisition module is used for real-time watch device, determines suitable device according to PHM principle P-F interval graph, and adopt
Collect suitable device measuring point data;
First processing module, for handling the measuring point data of acquisition;
EM equipment module is established, for establishing device data correlation model, according to data characteristics in device data correlation model
Model foundation equipment operation frame model;EM equipment module is established, for establishing device data correlation model, is closed according to device data
Data characteristics model foundation equipment operation frame model in gang mould type, by model result and measurement error in data characteristics model
Meet the Gaussian Profile that mean value is σ, as follows as error function corrective data correlation degree model using quadratic sum:
In formula, x: the stochastic variable based on historical data;Y: another stochastic variable based on historical data;σ:
Standard deviation;μ: mathematic expectaion;
Monitoring and warning module, for being monitored according to equipment operation frame model, the exception or failure for monitoring into
Row early warning;
Device diagnostic module, for it to be diagnosed and is handled according to early warning result.
Further, the data processing module includes:
Categorization module is classified device data by setting screening conditions for analyzing measuring point data;
Second processing module, for sorted measuring point data to be normalized.
Further, the EM equipment module of establishing includes computing module, and the computing module is used for equipment operation frame
Model optimizes equipment operation frame model using certainty factor algebra and support fitting algorithm, obtains meeting engineering actual
Equipment operation frame.
Further, the monitoring and warning module includes:
Comparison module, compared with carrying out measuring point data repeatedly with equipment operation frame model, using certainty factor algebra
It finds and meets the actual operation frame of engineering;
Matching module is run for calculating all measuring point datas using Euclidean distance algorithm and meeting the actual equipment of engineering
The match condition of frame;
Contrast module, for being compared according to device history moving law and current matching situation, analytical equipment is current
Health degree, the equipment health degree of calculating is shown.
Beneficial effects of the present invention:
1. unit exception or the timely early warning of failure can not only be made, the time that will be broken down and position are predicted, is had
It pointedly carries out equipment maintenance and meets the rank data integration requirement of group of industrial enterprise.
2. combining itself inherent characteristics of dynamic equipment operation, the slight change and deterioration of the operating status of detailed tracking equipment
Process;
3. having scientific, intuitive and comprehensive incipient fault warning function;
4. convenient, incipient fault efficiently and accurately diagnostic analysis method;
5. easy to operate, easy to use.
Detailed description of the invention
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without creative efforts, can also obtain according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the equipment running status health degree analysis method stream based on data-driven described according to embodiments of the present invention
Cheng Tu;
Fig. 2 is the CBM P-F curve synoptic diagram described according to embodiments of the present invention;
Fig. 3 is that the equipment running status health degree analysis system based on data-driven described according to embodiments of the present invention is shown
It is intended to;
Fig. 4 is the acquisition measuring point data described according to embodiments of the present invention;
Fig. 5 is the schematic diagram of described according to embodiments of the present invention device data correlation model and data characteristics model;
Fig. 6 is described based on data driven analysis model framework according to embodiments of the present invention;
Fig. 7 is the oxidation fan equipment health degree analysis tendency chart described according to embodiments of the present invention;
Fig. 8 is the oxidation fan equipment measuring point failure contribution degree schematic diagram described according to embodiments of the present invention;
Fig. 9 is described according to embodiments of the present invention oxidation fan equipment motor electric current measuring point actual value and predicted value trend
Figure;
Figure 10 is described according to embodiments of the present invention 1 measuring point actual value of oxidation fan equipment absorb the bottom of the tower liquid level and pre-
Measured value tendency chart.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art's every other embodiment obtained belong to what the present invention protected
Range.
As illustrated in fig. 1 and 2, a kind of equipment running status health based on data-driven according to embodiments of the present invention
Spend analysis method, comprising the following steps:
S1: real-time watch device determines suitable device according to PHM principle P-F interval graph, and acquires suitable device measuring point
Data;
Specifically, the common dynamic equipment of industry such as main monitoring rotation class equipment and reciprocating motion class equipment: such as power plant
And chemical plant, it includes that steam turbine, all kinds of pumps, all kinds of blowers, coal pulverizer and sky are pre- that wherein the equipment of real-time monitoring is capable of in power plant
Device etc.;The equipment that real-time monitoring is capable of in chemical plant includes air compressor, steam turbine, main pump and ventilation equipment etc., according to PHM original
Reason P-F interval graph determines suitable device, and acquires each monitoring point historical data of the equipment, is summarized storage;Equipment measuring point
Data are varied, by taking blower as an example, the equipment measuring point data of acquisition include blower motor electric current, blower motor coil temperature,
Blower medium inlet and outlet temperature, blower medium inlet and outlet pressure, blower rate-of flow, fan bearing temperature, fan vibration value and
Air door of fan aperture, wherein acquisition requires: it is time-centered, each measuring point data of equipment is acquired, the data of each time point are
One group of sampled value guarantees the time unification of each measuring point variable in every group of sampled value, meanwhile, all sampled values of equipment must be contained
Operating condition under lid Various Seasonal and different load, so, the data period of sampling is with 1-2;The time interval root of sampling
Depending on the difference rule of equipment operation;In general, the continually changing equipment of short term, sample frequency (i.e. sampling time interval)
With 1-5 seconds;The equipment that machine utilization will not frequently change in a short time, sample frequency was with 5-20 minutes.
S2: the measuring point data of acquisition is handled;
Specifically, the equipment running status health degree analysis technology based on data-driven based on data-driven needs to utilize
Big data analysis is carried out based on a large amount of historical datas of equipment self-operating, it is first before producing big data analysis to equipment
It first needs to carry out complex process to the historical data of equipment magnanimity, to meet the requirement of big data analysis.
S3: establishing device data correlation model, is set according to data characteristics model foundation in the device data correlation model
Received shipment row frame model;
As shown in figure 5, specifically, establishing data correlation model by dimensionality reduction, association and variance scheduling algorithm, rejecting is to result
Lesser variable is influenced, data characteristics model is advanced optimized, improves the efficiency and precision of data characteristics model;Model result with
Measurement error meets the Gaussian Profile that mean value is σ, using quadratic sum as error function corrective data correlation degree model
It is as follows:
In formula, x: the stochastic variable based on historical data;Y: another stochastic variable based on historical data;σ:
Standard deviation;μ: mathematic expectaion.
Device data related degree model can analyze the association of equipment normal operating condition and abnormal operating condition respectively
Degree, this analysis have been abandoned engineer in traditional analysis and have been run based on equipment based entirely on the moving law of equipment itself
The method of incidence relation between Analysis on Mechanism device parameter, passes through pure data analysing method, it is easier to excavate device parameter
Between complicated relationship, and then learn this incidence relation, discovering device earlier is abnormal.
On the basis of device data related degree model, multiple data characteristics models under different condition ultimately form equipment fortune
Row frame model, the equipment operation frame model include all operating status types of the equipment, form " health status " collection, table
Now for can qualitative analysis each data set model, establish the generation of operation frame model, operation frame carried out using K Mean Method
The coarse analysis of frame, is further exact computation results, uses certainty factor algebra and support based on big data analysis again
Fitting algorithm constantly optimizes operation frame, obtains meeting the actual equipment operation frame of engineering.
The equipment operation frame model of foundation can be enabled, deactivated, failed, be deleted, the management of failure analysis model.
S4: being monitored according to equipment operation frame model, and the exception or failure for monitoring carry out early warning;
Measuring point data is real time data, real time data and equipment operation frame model repeatedly compared with, using certainty factor algebra
It finds and meets the actual equipment operation frame of engineering;Wherein, the actual equipment operation frame of engineering is best equipment operation frame,
Then Euclidean distance algorithm is used, the match condition of all measuring point datas Yu best equipment operation frame is calculated;Last basis
The device history moving law excavated is compared with current matching situation, is analyzed the current health degree of equipment, will be calculated
Equipment health degree out can notice curvilinear trend figure and show operator, grasp the healthy shape of equipment in real time for operator
State.
If monitor the equipment current health degree and history health degree generation larger difference that function calculating goes out on-line, system meeting
Sharp investigation reminds industrial enterprise's maintenance maintenance to these differences, and more than early warning is issued after certain health degree
Staff carries out the accurate maintenance of planned shutdown, adjustment production plan and production scheduling.
S5: it is diagnosed and is handled according to early warning result.
When issuing fault pre-alarming information, by big data analysis means, having found two kinds can be applied to engineering practice
Effective ways, be expert diagnosis or knowledge base diagnosis respectively, diagnostic result, which is inquired, counted and export, produces corresponding report
Table is used for enterprise work personal management or decision, is improved enterprise to equipment management ability and is reduced maintenance maintenance difficulty;Its
In,
Expert diagnosis
The current operating condition of equipment and meet the actual equipment operation frame matching degree of engineering it is not high when, equipment run frame
Frame model issue equipment fault early warning information, then using optimization correlation rule support algorithm and certainty factor algebra by
The each parameter of analytical equipment and the otherness for meeting the actual operation frame of engineering, finally find out and cause the bad of current early warning
Parameter, and it is prompted to equipment operation expert;Situation of change of the expert based on these current bad parameters is run, for many years in conjunction with itself
Operating experience, the reason of equipment fault early-warning can be analyzed, wherein how to analyze the bad parameter of current early warning, especially
It is stealthy and plausible parameter, is the key factor that can analyze equipment fault, by correlation rule calculation method,
These stealthy parameters can be found out, to help operations staff's analytical equipment failure cause as far as possible.
Knowledge base diagnosis
It, can be by this part Heuristics with knowledge after operations staff analyzes the bad parameter and failure cause of each failure
The form of rule is filled into fault knowledge library;When equipment issue fault pre-alarming information after, first in fault knowledge library therefore
Barrier is matched, and the higher failure of matching degree is searched, and informs the failure that operations staff is currently most likely to occur.
Early warning or troubleshooting
Planned maintenance is targetedly carried out to equipment according to diagnostic result, maintenance result feedback is eliminated to system platform
Early warning, to the operation frame model re -training of the equipment, guarantees analysis model essence if device structure or operating condition change
Truly have effect.
In one particular embodiment of the present invention, the measuring point data of acquisition is carried out to handle specific packet in the step S2
It includes:
S21 analyzes measuring point data, and device data is classified by setting screening conditions;
Sorted measuring point data is normalized S22.
Specifically, S21 carries out normal operating condition analysis, equipment stoppage in transit state analysis and unit exception shape to historical data
State analysis, by the way that screening conditions are artificially arranged, automatically by equipment normal operating condition data, equipment stoppage in transit status data and equipment
Abnormality data are classified, so that the later period establishes the data analysis that device model carries out different dimensions.
Sorted measuring point data is normalized S22.Data normalization (standardization) processing is data mining
An element task, different evaluation index often has different dimension and dimensional unit, and such situation influences whether to count
According to analysis as a result, in order to eliminate the dimension impact between index, need to carry out data normalization processing, to solve data target
Between comparativity problem;For initial data after data normalization is handled, each index is in the same order of magnitude, is appropriate for comprehensive
Comparative evaluation is closed, the method for normalizing used is common Min-Max Scaling, also known as Min-Max
Normalization, formula are as follows:
In formula, x: the stochastic variable based on historical data;Z: normalizing coefficient.
In one particular embodiment of the present invention, the measuring point data of normalized is counted in the step S22
According to standardization.
In one particular embodiment of the present invention, by model result and measurement in data characteristics model in the step S3
Value error meets the Gaussian Profile that mean value is σ, as follows as error function corrective data correlation degree model using quadratic sum:
In formula, x: the stochastic variable based on historical data;Y: another stochastic variable based on historical data;σ:
Standard deviation;μ: mathematic expectaion.
In one particular embodiment of the present invention, equipment operation frame model uses certainty factor algebra in the step S3
Equipment operation frame model is optimized with support fitting algorithm, obtains meeting the actual equipment operation frame of engineering.
In one particular embodiment of the present invention, specific packet is monitored according to operation frame model in the step S4
It includes:
Compared with S41 carries out measuring point data repeatedly with equipment operation frame model, is found using certainty factor algebra and meet work
The actual operation frame of journey;
S42 calculates all measuring point datas using Euclidean distance algorithm and meets the matching of the actual equipment operation frame of engineering
Situation;
S43 is compared according to device history moving law and current matching situation, the current health degree of analytical equipment, will
The equipment health degree of calculating is shown.
Different operating statuses is divided to equipment according to equipment health degree, when equipment health degree is higher than 75%, equipment fortune
Row state is green, represents equipment normal operation;When equipment health degree is lower than 75%, equipment running status becomes red, and holds
Continuous flashing, and meet staff and inquire the healthy irrelevance historical trend of all supervision equipments and single measuring point actual value, pre-
Measured value trend;Wherein: 1) health degree: the characterization by modeling all measuring points, to the evaluation result of equipment health status;2) single
The actual value of measuring point: from the collected actual measured value of field device signals point;3) predicted value of single measuring point: system is according to ginseng
Prediction of several incidence relations to single measuring point ideal traffic coverage.
Shown by unit exception or the On-line Fault exception that will test of monitoring or failure with list or other forms and
Early warning by exception or Analysis of Fault Diagnosis exception or failure cause, and provides diagnostic comments, root on the basis of early warning
According to diagnostic comments, corresponding alarm description, alarm cause analysis, processing mode are inserted after operator's in-situ processing, can be incited somebody to action
Early warning processing status becomes processed, completes exception or troubleshooting, realizes the closed loop of exception or troubleshooting process.
As shown in figure 3, another aspect of the present invention, provides a kind of equipment running status health degree based on data-driven point
Analysis system, comprising:
Data acquisition module is used for real-time watch device, determines suitable device according to PHM principle P-F interval graph, and adopt
Collect suitable device measuring point data;
It is set specifically, real-time watch device mainly monitors common move of the industry such as rotation class equipment and reciprocating motion class equipment
It is standby: such as power plant and chemical plant, wherein power plant be capable of the equipment of real-time monitoring include steam turbine, all kinds of pumps, all kinds of blowers,
Coal pulverizer and air preheater etc.;The equipment that real-time monitoring is capable of in chemical plant includes air compressor, steam turbine, main pump and ventilation equipment
Deng determining suitable device according to PHM principle P-F interval graph, and acquire each monitoring point historical data of the equipment, summarized and deposited
Storage;Equipment measuring point data is varied, and by taking blower as an example, the equipment measuring point data of acquisition includes blower motor electric current, blower electricity
Machine coil temperature, blower medium inlet and outlet temperature, blower medium inlet and outlet pressure, blower rate-of flow, fan bearing temperature, wind
Machine vibration value and air door of fan aperture, wherein acquisition requires: it is time-centered, acquire each measuring point data of equipment, each time
The data of point are one group of sampled value, guarantee the time unification of each measuring point variable in every group of sampled value, meanwhile, equipment is all to adopt
Sample value must cover the operating condition under Various Seasonal and different load, so, the data period of sampling is with 1-2;Sampling
Time interval is depending on the difference rule that equipment is run;In general, the continually changing equipment of short term, sample frequency (are adopted
Sample time interval) with 1-5 seconds;The equipment that machine utilization will not frequently change in a short time, sample frequency was with 5-20 minutes.
First processing module, for handling the measuring point data of acquisition;
Specifically, the equipment running status health degree analysis technology based on data-driven based on data-driven needs to utilize
Big data analysis is carried out based on a large amount of historical datas of equipment self-operating, it is first before producing big data analysis to equipment
It first needs to carry out complex process to the historical data of equipment magnanimity, to meet the requirement of big data analysis.
EM equipment module is established, for establishing device data correlation model, according to data characteristics in device data correlation model
Model foundation equipment operation frame model;
Specifically, establishing data correlation model by dimensionality reduction, association and variance scheduling algorithm, reject lesser on result influence
Variable advanced optimizes data characteristics model, improves the efficiency and precision of data characteristics model;Model result and measurement error
Meet the Gaussian Profile that mean value is σ, as follows as error function corrective data correlation degree model using quadratic sum:
In formula, x: the stochastic variable based on historical data;Y: another stochastic variable based on historical data;σ:
Standard deviation;μ: mathematic expectaion.
Device data related degree model can analyze the association of equipment normal operating condition and abnormal operating condition respectively
Degree, this analysis have been abandoned engineer in traditional analysis and have been run based on equipment based entirely on the moving law of equipment itself
The method of incidence relation between Analysis on Mechanism device parameter, passes through pure data analysing method, it is easier to excavate device parameter
Between complicated relationship, and then learn this incidence relation, discovering device earlier is abnormal.
On the basis of device data related degree model, multiple data characteristics models under different condition ultimately form equipment fortune
Row frame model, the equipment operation frame model include all operating status types of the equipment, form " health status " collection, table
Now for can qualitative analysis each data set model, establish the generation of operation frame model, operation frame carried out using K Mean Method
The coarse analysis of frame, is further exact computation results, uses certainty factor algebra and support based on big data analysis again
Fitting algorithm constantly optimizes operation frame, obtains meeting the actual equipment operation frame of engineering.
The equipment operation frame model of foundation can be enabled, deactivated, failed, be deleted, the management of failure analysis model.
Monitoring and warning module, for being monitored according to equipment operation frame model, the exception or failure for monitoring into
Row early warning;
Specifically, measuring point data is real time data, real time data and equipment operation frame model repeatedly compared with, using confidence
Degree algorithm, which is found, meets the actual equipment operation frame of engineering;Wherein, the actual equipment operation frame of engineering is best equipment fortune
Then row frame uses Euclidean distance algorithm, calculates the match condition of all measuring point datas Yu best equipment operation frame;Most
It is compared afterwards according to the device history moving law excavated with current matching situation, analyzes the current health degree of equipment,
Calculated equipment health degree can be noticed into curvilinear trend figure and show operator, grasp equipment in real time for operator
Health status.
If monitor the equipment current health degree and history health degree generation larger difference that function calculating goes out on-line, system meeting
Sharp investigation reminds industrial enterprise's maintenance maintenance to these differences, and more than early warning is issued after certain health degree
Staff carries out the accurate maintenance of planned shutdown, adjustment production plan and production scheduling.
Device diagnostic module, for it to be diagnosed and is handled according to early warning result.
In one particular embodiment of the present invention, the data processing module includes:
Categorization module is classified device data by setting screening conditions for analyzing measuring point data;
Second processing module, for sorted measuring point data to be normalized.
Specifically, S21 carries out normal operating condition analysis, equipment stoppage in transit state analysis and unit exception shape to historical data
State analysis, by the way that screening conditions are artificially arranged, automatically by equipment normal operating condition data, equipment stoppage in transit status data and equipment
Abnormality data are classified, so that the later period establishes the data analysis that device model carries out different dimensions.
Sorted measuring point data is normalized S22.Data normalization (standardization) processing is data mining
An element task, different evaluation index often has different dimension and dimensional unit, and such situation influences whether to count
According to analysis as a result, in order to eliminate the dimension impact between index, need to carry out data normalization processing, to solve data target
Between comparativity problem;For initial data after data normalization is handled, each index is in the same order of magnitude, is appropriate for comprehensive
Comparative evaluation is closed, the method for normalizing used is common Min-Max Scaling, also known as Min-Max
Normalization, formula are as follows:
In formula, x: the stochastic variable based on historical data;Z: normalizing coefficient.
In one particular embodiment of the present invention, the EM equipment module of establishing includes computing module, the computing module
For equipment operation frame model to be optimized equipment operation frame model using certainty factor algebra and support fitting algorithm,
It obtains meeting the actual equipment operation frame of engineering.
In one particular embodiment of the present invention, the monitoring and warning module includes:
Comparison module, compared with carrying out measuring point data repeatedly with equipment operation frame model, using certainty factor algebra
It finds and meets the actual operation frame of engineering;
Matching module is run for calculating all measuring point datas using Euclidean distance algorithm and meeting the actual equipment of engineering
The match condition of frame;
Contrast module, for being compared according to device history moving law and current matching situation, analytical equipment is current
Health degree, the equipment health degree of calculating is shown.
Specifically, different operating statuses is divided to equipment according to equipment health degree, when equipment health degree is higher than 75%,
Equipment running status is green, represents equipment normal operation;When equipment health degree is lower than 75%, equipment running status becomes red
Color, and continue to flash, and meet healthy irrelevance historical trend and single measuring point reality that staff inquires all supervision equipments
Actual value, predicted value trend;Wherein: 1) health degree: the characterization by modeling all measuring points, to the evaluation knot of equipment health status
Fruit;2) actual value of single measuring point: from the collected actual measured value of field device signals point;3) predicted value of single measuring point:
Prediction of the system according to the incidence relation of parameter to single measuring point ideal traffic coverage.
Shown by unit exception or the On-line Fault exception that will test of monitoring or failure with list or other forms and
Early warning by exception or Analysis of Fault Diagnosis exception or failure cause, and provides diagnostic comments, root on the basis of early warning
According to diagnostic comments, corresponding alarm description, alarm cause analysis, processing mode are inserted after operator's in-situ processing, can be incited somebody to action
Early warning processing status becomes processed, completes exception or troubleshooting, realizes the closed loop of exception or troubleshooting process.
In order to facilitate understanding above-mentioned technical proposal of the invention, below by way of in specifically used mode to of the invention above-mentioned
Technical solution is described in detail.
When specifically used, the equipment running status health degree analysis side according to the present invention based on data-driven
Method, by taking certain power plant as an example, which, which builds together, erects standby 20, running state analysis model, No. 2 machines of real time on-line monitoring
Group equipment running status.
There is early warning on November 23rd, 2016, No. 1 oxidation fan model of the Power Plant No.2 unit, and the following are detailed early warning
Analytic process.
As shown in figure 4, the historical data of the 1. acquisition following measuring point 3-12 of the oxidation fan months:
No. 1 pressure fan outlet pressure;
No. 1 pressure fan current of electric;
No. 1 pressure fan damper positions;
No. 1 pressure fan bearing temperature;
No. 1 pressure fan coil temperature;
No. 1 pressure fan left hand inlet port flow;
No. 1 pressure fan right hand inlet port flow;
No. 1 air blower inlet temperature;
No. 1 pressure fan outlet temperature;
As shown in figure 5,2. on the basis of historical data, each measuring point characteristic model of the oxidation fan and correlation model are established,
As shown in Figure 5.
As shown in fig. 6,3. in each measuring point characteristic model of the oxidation fan and on the basis of correlation model, the oxidation fan is established
Operation frame model;
As shown in fig. 7,4. acquisition each equipment real time datas of the oxidation fan, real time data carry out in operation frame model
Big data analysis is unsatisfactory for the requirement of equipment operation frame and then carries out health degree early warning, when equipment health irrelevance is higher than 15 and full
When sufficient early warning regulation engine, system issues red early warning information, it can be seen from the figure that November 21 started No. 1 oxidation wind
Machine model begins to send out red early warning information, exception or fault management module and issues early warning and alarming information to staff.
As shown in FIG. 8 and 9,5. early warning diagnostic analysis and troubleshooting, steps are as follows:
(1), the measuring point for leading to current device model prediction occurring is found out by network analysis tool, it can from Fig. 8
Out, the measuring point for causing current alerts is oxidation fan current of electric and absorption tower oxidation air pressure;
(2), check that (deviation is that measuring point is practical for the deviation of oxidation fan electric current and absorption tower oxidation air pressure-measuring-point
Value and the difference of measuring point predicted value, predicted value are according to current device operating status calculated equipment normal operating condition automatically
Value), the deviation figure of oxidation fan electric current, this deviation figure as seen from Figure 9, the deviation of absorption tower oxidation air pressure with
Oxidation fan electric current is similar, is learnt by deviation, and oxidation air pressure and oxidation fan electric current are above normal value.
The possible cause that analysis causes oxidation fan model to alarm has:
Oxidation fan export branch road has individual pipeline blockages
Oxidation fan inlet filter needs to clean
The non-standard-sized sheet of oxidation fan outlet portal or blocking
Absorbing tower liquid-level is excessively high
Absorption tower density is excessively high
Oxidation fan outlet noise silencer may have sundries blocking
(3), pass through above-mentioned analysis, exclusive segment possibility, further to lock early warning reason.
1) the past period absorbing tower liquid-level curvilinear trend is checked, as shown in Figure 10: from October, absorbing tower liquid-level is equal
Keep higher level, and equipment operation frame model early warning since 11 the end of month are just, absorbing tower liquid-level is excessively high causes model for exclusion
The possibility of early warning.
2) check that the recent density in absorption tower is compared with before known to the past period absorption tower density curve, in lower
Level, therefore exclude the excessively high possibility for causing model to be alarmed of absorption tower density.
3) failure is locked, when equipment is rotated at regular intervals, No. 2 oxidation fan operations, No. 1 oxidation fan stops, but No. 2 oxidations
The reason of blower model generates same warning information, shows 1, No. 2 oxidation fan model pre-warning is common part failure, therefore is locked
Determining early warning reason is that oxidation fan export branch road has individual pipeline blockages.
6. troubleshooting
In the morning on December 4th, 2016, by system early warning information, it is oxidation fan that desulfurization personnel on site, which overhauls feedback result,
Export branch road pipeline blockage, it is consistent with prediction.To the morning on December 5th, 2016, in-situ processing is finished, and equipment restores normal fortune
Row operating condition, the alarm of equipment operation frame model disappear, and processing information is submitted in exception or fault management module, is completed at failure
Manage process closed loop.
In conclusion unit exception or failure can not only be made pre- in time by means of above-mentioned technical proposal of the invention
It is alert, the time that will be broken down and position are predicted, equipment maintenance is targetedly carried out and meets industrial enterprise's collection
Regimental other data integration requirement.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Within mind and principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of equipment running status health degree analysis method based on data-driven, which comprises the following steps:
S1: real-time watch device, according toPHMPrincipleP-FInterval graph determines suitable device, and acquires suitable device measuring point number
According to;
S2: the measuring point data of acquisition is handled;
S3: establishing device data correlation model, is transported according to data characteristics model foundation equipment in the device data correlation model
Row frame model;
S4: being monitored according to equipment operation frame model, and the exception or failure for monitoring carry out early warning;
S5: it is diagnosed and is handled according to early warning result.
2. the equipment running status health degree analysis method according to claim 1 based on data-driven, which is characterized in that
The measuring point data of acquisition processing is carried out in the step S2 to specifically include:
S21 analyzes measuring point data, and device data is classified by setting screening conditions;
Sorted measuring point data is normalized S22.
3. the equipment running status health degree analysis method according to claim 2 based on data-driven, which is characterized in that
The measuring point data of normalized is subjected to data normalization processing in the step S22.
4. the equipment running status health degree analysis method according to claim 1 based on data-driven, which is characterized in that
By model result and measurement error meet mean value for the Gaussian Profile of σ, using flat in data characteristics model in the step S3
It is square and as follows as error function corrective data correlation degree model:
In formula,x: the stochastic variable based on historical data;y: another stochastic variable based on historical data;σ: standard
Difference;μ: mathematic expectaion.
5. the equipment running status health degree analysis method according to claim 1 based on data-driven, which is characterized in that
Equipment operation frame model optimizes equipment operation frame using certainty factor algebra and support fitting algorithm in the step S3
Frame model obtains meeting the actual equipment operation frame of engineering.
6. the equipment running status health degree analysis method according to claim 5 based on data-driven, which is characterized in that
It is monitored and is specifically included according to operation frame model in the step S4:
Compared with S41 carries out measuring point data repeatedly with equipment operation frame model, is found using certainty factor algebra and meet engineering reality
The operation frame on border;
S42 calculates all measuring point datas using Euclidean distance algorithm and meets the matching feelings of the actual equipment operation frame of engineering
Condition;
S43 is compared according to device history moving law and current matching situation, and the current health degree of analytical equipment will be counted
The equipment health degree of calculation is shown.
7. a kind of equipment running status health degree analysis system based on data-driven characterized by comprising
Data acquisition module is used for real-time watch device, according toPHMPrincipleP-FInterval graph determines suitable device, and acquires suitable
With equipment measuring point data;
First processing module, for handling the measuring point data of acquisition;
EM equipment module is established, for establishing device data correlation model, according to data characteristics model in device data correlation model
Equipment operation frame model is established, model result in data characteristics model and measurement error are met into the Gauss point that mean value is σ
Cloth, as follows as error function corrective data correlation degree model using quadratic sum:
In formula,x: the stochastic variable based on historical data;y: another stochastic variable based on historical data;σ: standard
Difference;μ: mathematic expectaion;
Monitoring and warning module, for being monitored according to equipment operation frame model, exception or failure for monitoring carry out pre-
It is alert;
Device diagnostic module, for it to be diagnosed and is handled according to early warning result.
8. the equipment running status health degree analysis system according to claim 7 based on data-driven, which is characterized in that
The data processing module includes:
Categorization module is classified device data by setting screening conditions for analyzing measuring point data;
Second processing module, for sorted measuring point data to be normalized.
9. the equipment running status health degree analysis system according to claim 7 based on data-driven, which is characterized in that
The EM equipment module of establishing includes computing module, and the computing module is used to equipment operation frame model using certainty factor algebra
Equipment operation frame model is optimized with support fitting algorithm, obtains meeting the actual equipment operation frame of engineering.
10. the equipment running status health degree analysis system according to claim 9 based on data-driven, feature exist
In the monitoring and warning module includes:
Comparison module is found compared with carrying out measuring point data repeatedly with equipment operation frame model using certainty factor algebra
Meet the actual operation frame of engineering;
Matching module, for calculating all measuring point datas using Euclidean distance algorithm and meeting the actual equipment operation frame of engineering
Match condition;
Contrast module, for being compared according to device history moving law and current matching situation, analytical equipment is current to be good for
The equipment health degree of calculating is shown by Kang Du.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811258951.3A CN109240244B (en) | 2018-10-26 | 2018-10-26 | Data-driven equipment running state health degree analysis method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811258951.3A CN109240244B (en) | 2018-10-26 | 2018-10-26 | Data-driven equipment running state health degree analysis method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109240244A true CN109240244A (en) | 2019-01-18 |
CN109240244B CN109240244B (en) | 2020-11-20 |
Family
ID=65078509
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811258951.3A Expired - Fee Related CN109240244B (en) | 2018-10-26 | 2018-10-26 | Data-driven equipment running state health degree analysis method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109240244B (en) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109886430A (en) * | 2019-01-24 | 2019-06-14 | 同济大学 | A kind of equipment health state evaluation and prediction technique based on industrial big data |
CN110259433A (en) * | 2019-06-28 | 2019-09-20 | 宝鸡石油机械有限责任公司 | A kind of entity drilling machine digital monitoring method |
CN110794799A (en) * | 2019-11-28 | 2020-02-14 | 桂林电子科技大学 | Big data system with fault diagnosis function applied to industrial production |
CN110807245A (en) * | 2019-09-26 | 2020-02-18 | 上海长庚信息技术股份有限公司 | Automatic modeling method and system for equipment fault early warning |
CN111224468A (en) * | 2019-12-30 | 2020-06-02 | 宁夏百川电力股份有限公司 | Photovoltaic equipment safe operation and maintenance platform based on cloud computing |
CN111753396A (en) * | 2020-05-22 | 2020-10-09 | 安徽科技学院 | Desulfurization and denitrification circulating pump fault diagnosis method based on mechanism-data driving model |
CN112529036A (en) * | 2020-11-06 | 2021-03-19 | 上海发电设备成套设计研究院有限责任公司 | Fault early warning method, device, equipment and storage medium |
CN112580993A (en) * | 2020-12-23 | 2021-03-30 | 云南电网有限责任公司昆明供电局 | Power grid equipment fault probability analysis method |
CN112817280A (en) * | 2020-12-04 | 2021-05-18 | 华能国际电力股份有限公司玉环电厂 | Implementation method for intelligent monitoring alarm system of thermal power plant |
CN113125150A (en) * | 2021-04-20 | 2021-07-16 | 武汉信达天成物联网技术有限公司 | Method and system for monitoring health state of drive chain on line |
CN113806969A (en) * | 2021-10-26 | 2021-12-17 | 国家石油天然气管网集团有限公司 | Compressor unit health prediction method based on time domain data correlation modeling |
CN113810214A (en) * | 2020-10-09 | 2021-12-17 | 广东毓秀科技有限公司 | Data-driven prediction maintenance method for FSK communication equipment |
CN113963456A (en) * | 2020-07-20 | 2022-01-21 | 核工业理化工程研究院 | Method and system for analyzing operation data of multiple high-speed rotating devices |
CN114595271A (en) * | 2022-02-24 | 2022-06-07 | 贵州具京网络科技有限公司 | Big data mining method and system |
CN114757380A (en) * | 2022-04-29 | 2022-07-15 | 西安热工研究院有限公司 | Thermal power plant fault early warning system and method, electronic equipment and storage medium |
CN117291582A (en) * | 2023-11-27 | 2023-12-26 | 合肥宝康自动化系统有限公司 | Industrial production interconnection monitoring system based on data analysis |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102270271A (en) * | 2011-05-03 | 2011-12-07 | 北京中瑞泰科技有限公司 | Equipment failure early warning and optimizing method and system based on similarity curve |
CN102765643A (en) * | 2012-05-31 | 2012-11-07 | 天津大学 | Elevator fault diagnosis and early-warning method based on data drive |
CN102809965A (en) * | 2012-07-30 | 2012-12-05 | 燕山大学 | Fault early warning method for hydraulic equipment based on fault frequent pattern |
US20160071004A1 (en) * | 2015-10-23 | 2016-03-10 | Sarkhoon and Qeshm LLC | Method and system for predictive maintenance of control valves |
CN106762452A (en) * | 2016-12-08 | 2017-05-31 | 东北大学 | Fan master control system fault diagnosis and on-line monitoring method based on data-driven |
-
2018
- 2018-10-26 CN CN201811258951.3A patent/CN109240244B/en not_active Expired - Fee Related
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102270271A (en) * | 2011-05-03 | 2011-12-07 | 北京中瑞泰科技有限公司 | Equipment failure early warning and optimizing method and system based on similarity curve |
CN102765643A (en) * | 2012-05-31 | 2012-11-07 | 天津大学 | Elevator fault diagnosis and early-warning method based on data drive |
CN102809965A (en) * | 2012-07-30 | 2012-12-05 | 燕山大学 | Fault early warning method for hydraulic equipment based on fault frequent pattern |
US20160071004A1 (en) * | 2015-10-23 | 2016-03-10 | Sarkhoon and Qeshm LLC | Method and system for predictive maintenance of control valves |
CN106762452A (en) * | 2016-12-08 | 2017-05-31 | 东北大学 | Fan master control system fault diagnosis and on-line monitoring method based on data-driven |
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109886430A (en) * | 2019-01-24 | 2019-06-14 | 同济大学 | A kind of equipment health state evaluation and prediction technique based on industrial big data |
CN109886430B (en) * | 2019-01-24 | 2023-07-07 | 同济大学 | Equipment health state assessment and prediction method based on industrial big data |
CN110259433A (en) * | 2019-06-28 | 2019-09-20 | 宝鸡石油机械有限责任公司 | A kind of entity drilling machine digital monitoring method |
CN110807245A (en) * | 2019-09-26 | 2020-02-18 | 上海长庚信息技术股份有限公司 | Automatic modeling method and system for equipment fault early warning |
CN110807245B (en) * | 2019-09-26 | 2023-05-23 | 上海长庚信息技术股份有限公司 | Automatic modeling method and system for equipment fault early warning |
CN110794799A (en) * | 2019-11-28 | 2020-02-14 | 桂林电子科技大学 | Big data system with fault diagnosis function applied to industrial production |
CN111224468B (en) * | 2019-12-30 | 2020-10-20 | 宁夏百川电力股份有限公司 | Photovoltaic equipment safe operation and maintenance platform based on cloud computing |
CN111224468A (en) * | 2019-12-30 | 2020-06-02 | 宁夏百川电力股份有限公司 | Photovoltaic equipment safe operation and maintenance platform based on cloud computing |
CN111753396A (en) * | 2020-05-22 | 2020-10-09 | 安徽科技学院 | Desulfurization and denitrification circulating pump fault diagnosis method based on mechanism-data driving model |
CN113963456A (en) * | 2020-07-20 | 2022-01-21 | 核工业理化工程研究院 | Method and system for analyzing operation data of multiple high-speed rotating devices |
CN113810214A (en) * | 2020-10-09 | 2021-12-17 | 广东毓秀科技有限公司 | Data-driven prediction maintenance method for FSK communication equipment |
CN112529036A (en) * | 2020-11-06 | 2021-03-19 | 上海发电设备成套设计研究院有限责任公司 | Fault early warning method, device, equipment and storage medium |
CN112529036B (en) * | 2020-11-06 | 2023-06-13 | 上海发电设备成套设计研究院有限责任公司 | Fault early warning method, device, equipment and storage medium |
CN112817280A (en) * | 2020-12-04 | 2021-05-18 | 华能国际电力股份有限公司玉环电厂 | Implementation method for intelligent monitoring alarm system of thermal power plant |
CN112580993A (en) * | 2020-12-23 | 2021-03-30 | 云南电网有限责任公司昆明供电局 | Power grid equipment fault probability analysis method |
CN112580993B (en) * | 2020-12-23 | 2022-08-26 | 云南电网有限责任公司昆明供电局 | Power grid equipment fault probability analysis method |
CN113125150A (en) * | 2021-04-20 | 2021-07-16 | 武汉信达天成物联网技术有限公司 | Method and system for monitoring health state of drive chain on line |
CN113806969A (en) * | 2021-10-26 | 2021-12-17 | 国家石油天然气管网集团有限公司 | Compressor unit health prediction method based on time domain data correlation modeling |
CN114595271B (en) * | 2022-02-24 | 2022-12-13 | 北京中数睿智科技有限公司 | Big data mining method and system |
CN114595271A (en) * | 2022-02-24 | 2022-06-07 | 贵州具京网络科技有限公司 | Big data mining method and system |
CN114757380A (en) * | 2022-04-29 | 2022-07-15 | 西安热工研究院有限公司 | Thermal power plant fault early warning system and method, electronic equipment and storage medium |
CN114757380B (en) * | 2022-04-29 | 2024-02-20 | 西安热工研究院有限公司 | Fault early warning system and method for thermal power plant, electronic equipment and storage medium |
CN117291582A (en) * | 2023-11-27 | 2023-12-26 | 合肥宝康自动化系统有限公司 | Industrial production interconnection monitoring system based on data analysis |
CN117291582B (en) * | 2023-11-27 | 2024-03-29 | 合肥宝康自动化系统有限公司 | Industrial production interconnection monitoring system based on data analysis |
Also Published As
Publication number | Publication date |
---|---|
CN109240244B (en) | 2020-11-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109240244A (en) | Equipment running status health degree analysis method and system based on data-driven | |
US7827006B2 (en) | Heat exchanger fouling detection | |
CN113705924B (en) | Intelligent diagnosis method and system for thermal control equipment | |
US8606544B2 (en) | Methods and systems for detecting deviation of a process variable from expected values | |
CA2803114C (en) | System, method, and apparatus for oilfield equipment prognostics and health management | |
US7526405B2 (en) | Statistical signatures used with multivariate statistical analysis for fault detection and isolation and abnormal condition prevention in a process | |
EP1435023B1 (en) | Method and system for analysing control circuit performance in industrial process | |
DE60203458T2 (en) | Integrated control and diagnostics for a motorized system using vibration, pressure, temperature, velocity and / or current analysis | |
US8032340B2 (en) | Method and system for modeling a process variable in a process plant | |
CN108763729A (en) | Process industry Mechatronic Systems couple state appraisal procedure based on network structure entropy | |
CN109118384A (en) | A kind of Wind turbines healthy early warning method | |
KR20140130540A (en) | Method and system for real-time performance recovery advisory for centrifugal compressors | |
US8145358B2 (en) | Method and system for detecting abnormal operation of a level regulatory control loop | |
Rodrigues et al. | Short and long forecast to implement predictive maintenance in a pulp industry | |
CN115267646A (en) | Method for detecting state of electric energy meter and related equipment | |
Sanayha et al. | Fault detection for circulating water pump using time series forecasting and outlier detection | |
Wang | Key techniques in intelligent predictive maintenance (IPdM)–a framework of intelligent faults diagnosis and prognosis system (IFDaPS) | |
Chesterman et al. | Condition monitoring of wind turbines and extraction of healthy training data using an ensemble of advanced statistical anomaly detection models | |
KR102140532B1 (en) | Method for Failure Detecting in Air Compressor based on big data | |
Omitaomu et al. | Online support vector regression approach for the monitoring of motor shaft misalignment and feedwater flow rate | |
CN116483054A (en) | Industrial robot running state monitoring and early warning system and method | |
Zhang | Comparison of data-driven and model-based methodologies of wind turbine fault detection with SCADA data | |
Sonthipo et al. | Fault detection and identification for centrifugal compressor by ensemble model | |
Staino et al. | A Monte-Carlo approach for prognostics of clogging process in HVAC filters using a hybrid strategy a real case study in railway systems | |
Alimohammadi et al. | Predict the remaining useful life in HVAC filters using a hybrid strategy |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20201120 Termination date: 20211026 |