CN108694448A - PHM platforms - Google Patents
PHM platforms Download PDFInfo
- Publication number
- CN108694448A CN108694448A CN201810432744.9A CN201810432744A CN108694448A CN 108694448 A CN108694448 A CN 108694448A CN 201810432744 A CN201810432744 A CN 201810432744A CN 108694448 A CN108694448 A CN 108694448A
- Authority
- CN
- China
- Prior art keywords
- data
- stream
- management
- module
- user
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses PHM platforms, including modeling space unit:Including data flow structure module, stream management module, stream correction verification module, stream operation module and console output module, user obtains data using data flow structure module, the structure of data flow is completed in a manner of the easy graphical operation of dragging and line, then management operation and validity check are carried out to data stream by stream management module and stream correction verification module, user is assisted to complete model running and debugging, then feedback result is supplied to console output module, the execution feedback result of stream is carried out visualization to show, completes entire modeling;Data Management Unit, operational management unit, System Management Unit and page display unit.The present invention passes through above-mentioned principle, testing equipment monitoring running state, big data management and analysis mining, fault diagnosis and fault prediction are treated in realization, maintenance cost is low, abnormity diagnosis is simple, realizes real time monitoring, provides fundamental basis for military electronic product failure prediction capability.
Description
Technical field
The present invention relates to big data application fields, and in particular to PHM platforms.
Background technology
PHM is the abbreviation of Prognostic and Health Management, i.e. prognostic and health management.PHM
Major significance be that the time that will break down of precognition and position, predict the RUL (Remain of whole system in advance
Useful Life, remaining life), the operational reliability of system is improved, the maintenance cost of system is reduced and improves repair standard
True property realizes the CBM (Condition Based Maintainence, the repair based on state) of electronic system.Meanwhile PHM
System records and analyzes the health data of electronic system, and as management health, health control is carried out to whole system.With
High-tech continuous injection, the high efficiency of the highly integrated of Technology for Modern Equipment, high-intelligentization and analyzing processing problem are increasingly enhanced,
Fault diagnosis, maintenance support and the reliability of thing followed system are increasingly paid much attention to by people.Current high-end dress
Standby to be faced with some urgent problems to be solved, main includes that maintenance cost is high, abnormity diagnosis is difficult, can not monitor in real time, nothing
Method predictive maintenance maintenance etc..
Invention content
The technical problem to be solved by the present invention is to realize the abnormity diagnosis to Technology for Modern Equipment, real time monitoring, reduce and safeguard
Cost, and it is an object of the present invention to provide PHM platforms, realization fast and accurately treat testing equipment monitoring running state, big data management with
Analysis mining, fault diagnosis and fault prediction, maintenance cost bottom, abnormity diagnosis is simple, realization monitors in real time, being capable of predictive maintenance inspection
It repaiies, provides fundamental basis for military electronic product failure prediction capability.
The present invention is achieved through the following technical solutions:
PHM platforms, including
Modeling space unit:Including data flow structure module, stream management module, stream correction verification module, stream operation module and
Console output module, user obtains data using data flow structure module, to pull the easy graphical operation side with line
Formula completes the structure of data flow, then carries out management operation and validity to data stream by stream management module and stream correction verification module
Verification assists user to complete model running and debugging, feedback result is then supplied to console output module, passes through console
The execution feedback result of stream is carried out visualization and showed by output window, is finally completed entire modeling process;
Data Management Unit:External file import feature is provided, includes that the source data that modeling space unit is needed imports
Then data are uniformly managed by big data platform;
Operational management unit:For providing various data stream scheduling operation reserves, realizes the operation monitoring to data flow, transports
Line detail records and condition managing;
System Management Unit:Including user management, Role Management, rights management, unit module management, information department management
With log management, entire platform normal operation is ensured;
Page display unit:Display platform resource distribution situation, special time period operation data flow statistical analysis with
And platform operation entry is provided for each user, an independent personal space is provided for each user, is only opened up in personal space
Show the content of individual subscriber, including the model run recently, data and log.
Modeling space unit on PHM platforms realizes the establishment and maintenance of distributed high-performance algorithm, provides to magnanimity
The efficient process of data, Various types of data mining algorithm and data are read in and are write out, data prediction, feature selecting, mould by it
The necessary links during business models such as type assessment, are encapsulated as that form is consistent, configurable operator, and business model process
It is abstracted as orderly, the workflow that substep executes, user is allowed to complete analysis in a manner of the easy graphical operation with line to pull and dig
Pick process, and intuitively shown by page display unit, it is easy to use.This programme also sets up System Management Unit
Platform is managed collectively, improving O&M efficiency reduces O&M cost.This programme combination big data technology is realized quick and precisely
Treat testing equipment monitoring running state, big data management and analysis mining, fault diagnosis and fault prediction, it is maintenance cost bottom, different
Often diagnosis is simple, realize real time monitoring, being capable of predictive maintenance maintenance, moreover it is possible to be carried out according to equipment big data situation to be detected a
Propertyization is formulated, and is provided fundamental basis for military electronic product failure prediction capability.This programme also provides user collaboration space so that
Multiple users can light shared data and mining model.
Preferably, data flow structure module includes acquisition to data, storage, data prediction, the modeling of data
Algorithm and model is assessed.
Preferably, the modeling algorithm includes carrying out statistical model, mining algorithm to data stream according to data analysis tool
And signal processing, mining algorithm are exactly to find out to lie in information useful in data from a large amount of data, signal processing is
Be by the fast Fourier transform function in signal processing operator, by sensor acquire time-domain signal be converted to frequency domain and
Time and frequency zone signal, to extract time domain, frequency domain and the time and frequency zone feature of signal, to react the operating status of platform.
Preferably, it is found out from a large amount of data by modeling algorithm and lies in information useful in data and commented by model
Evaluation component on estimating assesses accuracy, feasibility and the applicability of information.
Preferably, the data flow structure module further includes miscellaneous function, and addition painting canvas shortcut key is supported.
Preferably, flow tube reason module includes the operational administrative to data flow, data flow classification management and data flow category
Property management, stream correction verification module include convection current node verification and stream logic verify, stream operation module by run stream carry out stream node
Between data processing and operating.
Preferably, the Data Management Unit includes the shared of the acquisition of data, the maintenance of data and data.
Preferably, the operational management unit includes the monitoring of the monitoring and operation details to log, data flow
Log each time runs the record of details and is managed to data flow operating status.
Preferably, the page display unit includes the note that showing resource configures, the activity of the user and user are recent
Record.
Compared with prior art, the present invention having the following advantages and advantages:
1, the present invention combines big data technology, realization fast and accurately to treat testing equipment monitoring running state, big data
Management and analysis mining, fault diagnosis and fault prediction, maintenance cost bottom, abnormity diagnosis is simple, realization monitors in real time, can be predictive
Maintenance and Repair, moreover it is possible to personalized formulation be carried out according to equipment big data situation to be detected, be military electronic product failure predication energy
Power is provided fundamental basis.
2, modeling space unit of the invention realizes the establishment and maintenance of distributed high-performance algorithm, provides to magnanimity number
According to efficient process, Various types of data mining algorithm and data are read in and are write out, data prediction, feature selecting, model by it
The necessary links during business models such as assessment, are encapsulated as that form is consistent, configurable operator, and business model process pumping
As to be orderly, the workflow that substep executes allows user to complete analysis mining in a manner of the easy graphical operation with line to pull
Process, and intuitively shown by page display unit, it is easy to use.
Description of the drawings
Attached drawing described herein is used for providing further understanding the embodiment of the present invention, constitutes one of the application
Point, do not constitute the restriction to the embodiment of the present invention.In the accompanying drawings:
Fig. 1 is schematic structural view of the invention.
Specific implementation mode
To make the objectives, technical solutions, and advantages of the present invention clearer, with reference to embodiment and attached drawing, to this
Invention is described in further detail, and exemplary embodiment of the invention and its explanation are only used for explaining the present invention, do not make
For limitation of the invention.
Embodiment 1:
As shown in Figure 1, the present invention includes PHM platforms, including
Modeling space unit:Including data flow structure module, stream management module, stream correction verification module, stream operation module and
Console output module, user obtains data using data flow structure module, to pull the easy graphical operation side with line
Formula completes the structure of data flow, then carries out management operation and validity to data stream by stream management module and stream correction verification module
Verification assists user to complete model running and debugging, feedback result is then supplied to console output module, passes through console
The execution feedback result of stream is carried out visualization and showed by output window, is finally completed entire modeling process.Console output module
For the feedback that executes of output stream, including situations such as the operating status of data flow, operation feedback, user is facilitated to check execution
Failure cause.Execute feedback refer to design interface click operation be immediately performed after, show stream execution state and if
Mistake has occurred, needs to show error message.To help user's Wrong localization.Execution each time can refresh in execution feedback
Hold.
Data Management Unit:External file import feature is provided, includes that the source data that modeling space unit is needed imports
Then data are uniformly managed by big data platform;
Operational management unit:For providing various data stream scheduling operation reserves, realizes the operation monitoring to data flow, transports
Line detail records and condition managing;
System Management Unit:Including user management, Role Management, rights management, unit module management, information department management
With log management, entire platform normal operation is ensured;User management provide the checking of user list (supporting search filtering) and
Management operation, including the newly-increased of user, modification, deletion etc. to user.User's inquiry can pass through department or user famous model
The filtering that paste matching carries out user is searched, and when not inquired, shows the list information of all users.
Role Management provides checking and (support search filtering) and being operated to the management of role for character list.Permission pipe
Manage for role or user provide operating resource/permissions list check and management function.Module management carries out pipe to module information
Reason, including check, add, removing module information, and to the enabling etc. of page module.Information department management provides organization
Management function can possess infinite stages, and staff can be affiliated under every level-one, similar multistage division management.Log management carries
Operation is checked and manages for log list.
Page display unit:Display platform resource distribution situation, special time period operation data flow statistical analysis with
And platform operation entry is provided for each user, an independent personal space is provided for each user, is only opened up in personal space
Show the content of individual subscriber, including the model run recently, data and log.
Page display unit includes the record that showing resource configures, the activity of the user and user are recent.Resource distribution
Cpu utilization rates in each time interval are illustrated, hdfs disk utilization rates, time interval can be from nearest 1 together for memory usage
Hour was to nearest 30 days.The activity of the user has counted the stream of the number of run top ten list in nearest 7 days or 30 days, takes most
3 long streams, create the quantity and associated data accounting of stream.The model that the recent record display user of user designs recently,
The data file uploaded recently, the record list of operation stream recently, and entry can be clicked directly on and enter corresponding build
Modular space, data management space and operational management space.
Modeling space unit on PHM platforms realizes the establishment and maintenance of distributed high-performance algorithm, provides to magnanimity
The efficient process of data, Various types of data mining algorithm and data are read in and are write out, data prediction, feature selecting, mould by it
The necessary links during business models such as type assessment, are encapsulated as that form is consistent, configurable operator, and business model process
It is abstracted as orderly, the workflow that substep executes, user is allowed to complete analysis in a manner of the easy graphical operation with line to pull and dig
Pick process, and intuitively shown by page display unit, it is easy to use.This programme also sets up System Management Unit
Platform is managed collectively, improving O&M efficiency reduces O&M cost.This programme combination big data technology is realized quick and precisely
Treat testing equipment monitoring running state, big data management and analysis mining, fault diagnosis and fault prediction, it is maintenance cost bottom, different
Often diagnosis is simple, realize real time monitoring, being capable of predictive maintenance maintenance, moreover it is possible to be carried out according to equipment big data situation to be detected a
Propertyization is formulated, and is provided fundamental basis for military electronic product failure prediction capability.This programme also provides user collaboration space so that
Multiple users can light shared data and mining model.
Embodiment 2:
The present embodiment is preferably as follows on the basis of embodiment 1:Data flow structure module includes acquisition to data, data
Storage, data prediction, modeling algorithm and model evaluation.The acquisition of data is provided by platform and is visited big data platform data
It asks, the function of data selection, user selects corresponding tables of data according to the needs of oneself.Data classification includes business data and use
Two parts of family personal data, and support the locating function to tables of data.After choosing tables of data, user can be to field
It carries out renaming or field is mapped to another from a source Nodes, and output field mistake is carried out to the tables of data of reading
The operation of filter.It supports to carry out data preview to the data field configured simultaneously.It is presented after data acquisition by visualizing, it can be fast
Speed grasps the essentials information can be using graphic assembly in order to support the ways of presentation to data simple, intuitive, realization data visualization
Change, including:Histogram, distribution map, scatter plot and multiple scatter plot.
Calculating and storage resource in this platform can be extending transversely, and used model can be with major section in modeling algorithm
Institutes's cooperation is ground, new Professional Model is irregularly extended.And the model of profession can be with preset a large amount of distributed data statistics, diggings
Dig model algorithm;Preset a large amount of high-end equipment Professional Models;With the fields PHM scientific research institution, colleges and universities and business tie-up, in practice
It continues to optimize and sophisticated model;Support export, the transplanting of User Defined algorithm and model.In addition the modeling algorithm of this programme is also
It can find some incidence relations between data automatically by machine learning, auxiliary is provided for regularity exploring, trend prediction etc..
Since data are generally all imperfect, inconsistent dirty datas in the real world of acquisition, can not directly carry out
Data mining or Result are barely satisfactory.Quality this programme in order to improve data mining devises data prediction.We
The data preprocessing method that case uses includes data scrubbing, data integration, data transformation, data regularization etc. to the data of acquisition into
Data mining is carried out by modeling algorithm again after line number Data preprocess.These data processing techniques use before data mining,
The quality of data mining pattern is substantially increased, the actual excavation required time is reduced.In addition, carrying out data prediction
When, also there are record processing function and field processing function to preprocessed data.Record processing function refers in data record level
On it is adjusted and is converted, be used for the data preparation of modelling phase, and user's particular traffic requirements can be met to data
It is cut out accordingly.Including selection, sample, differentiation and balance etc..Field processing prepare analyze during selection,
Remove or construct data.Field operations include many nodes for being suitable for this conversion and preparation, including type, filling,
Subregion etc..
Modeling algorithm includes being carried out at statistical model, mining algorithm and signal to data stream according to data analysis tool
Reason, mining algorithm is exactly to find out to lie in information useful in data from a large amount of data, and signal processing is to pass through signal
The fast Fourier transform function in operator is handled, the time-domain signal that sensor acquires is converted into frequency domain and time and frequency zone signal,
To extract time domain, frequency domain and the time and frequency zone feature of signal, to react the operating status of platform.
Statistical model provides the data analysis tool based on statistical model based on platform, and tool includes:Arithmetical operation, correlation
Coefficient, hypothesis testing carry out data statistic analysis.Arithmetical operation includes basic statistical calculation, i.e.,:Counting, average value, conjunction
Meter, maximum value, minimum value etc..Related coefficient is used to reflect the statistical indicator of linear relationship between variable two-by-two.In reality
It is generally used for being fitted two groups of data and similarity degree carries out quantitative analysis.Hypothesis testing is basis in mathematical statistics
Certain assumed condition is inferred a kind of overall method by sample.It is main at present to provide it is assumed hereinafter that check system:Chi-square Test with
And KS is examined.
Mining algorithm is also referred to as the Knowledge Discovery in database.Briefly, it is exactly found out from a large amount of data implicit
The useful information in data.With the arrival in big data epoch, data mining shows more vigorous vitality.If will
The technology of data mining is combined with the fault data of army's weapon system, by suitable Algorithm Analysis, is provided to these data
Source is effectively utilized, and the corresponding model that weapon system failure can be effectively predicted is established, to the weapon system that may occur
Failure is forecast in advance.So platform provides a series of machine learning algorithms, including:Decision tree, support vector machines, with
Machine forest etc..
In the process of running, the signals such as vibration, temperature, pressure of generation respond the operation that can reflect equipment to equipment
State.By signal processing, acquisition signal can reaction equipment operating status.
It is found out from a large amount of data by modeling algorithm and lies in information useful in data and pass through in model evaluation
Evaluation component assesses accuracy, feasibility and the applicability of information.
The data flow structure module further includes miscellaneous function, and the Experience Degree for improving user adds painting canvas shortcut key
It supports, such as the deletion of stream, the revocation of stream, the preservation reformed, flowed flowed, the verification of stream.
In order to which the quality of evaluation model, platform are commented using model evaluation offer after having been trained by modeling algorithm implementation model
Estimate component and analysis and assessment are carried out to it.User can be come the accuracy of assessment models, feasibility and is applicable in using evaluation component
Property etc..The assessment that this programme is realized is divided into three major class:Cluster Evaluation, classification assessment and recurrence assessment.
Data flow, which builds module, can also use data storage component, and data are carried out to the storage of persistence.It can be by number
According to being stored as following kind of data.Including the storage of big data platform table, flat file storage and generate EXCEL.
The process is realized to abnormal quick diagnosis, and carries out real-time display, diagnosis letter by intuitive graphical interfaces
Just, maintenance cost is low and reliability is high for folk prescription.
Embodiment 3:
The present embodiment is preferably as follows on the basis of embodiment 1:Stream management module includes operational administrative, the number to data flow
According to stream Classification Management and data stream property management, stream correction verification module includes the verification of convection current node and stream logic verify, stream operation mould
Block carries out data processing and operating between stream node by running stream.It ensure that the accurate and reliable of testing result.
Data stream management includes the newly-increased of convection current, opening, deletion, modification, preservation etc..It should be noted source data node
It is the beginning of data flow, input terminal can only be used as;Output node is only output end, and cannot reconnect other nodes.
Data flow classification management provides data flow classification management function, and support flows into industry business Classification Management behaviour to data
Make.User can establish classification tree, and different stream is returned and is placed under different class nodes.
The display format and data preview line number of data in User Defined data flow are supported in data stream property management
Limitation.Realize Personalized service.
It includes that the input and output to all components in model and parameter setting etc. carry out inspection verification to flow node verification, inspection
Have in looking into:Whether node has configured input;Node must fill out whether parameter is empty;Whether node parameter Value Types match;Ginseng
Numerical check etc..In the process, if verification does not pass through, there will be corresponding miscue information (to occur after text box red
Prompt), and then user can be arranged according to prompt message adjusting parameter.
Stream logic verify includes that the logical design of convection current is verified.As two components to repel each other cannot function as upstream and downstream pass
System is attached;Data source component can only have output, so other assemblies is not supported to be input to this component;Graphic assembly and storage
Component can only have input, so not supporting to output data to other assemblies.One component definition is that only there are one input and one
A output, then just not supporting multi input and multi output etc..Realize that predictive maintenance overhauls situation by verification.
Operation module is flowed after stream debugging is completed, and data processing and fortune between stream node can be carried out by operation stream
Turn.The mode of stream operation includes three kinds:It can click to be immediately performed in design interface and be run immediately, can also be managed in operation
Reason interface is clicked to be immediately performed to be run immediately, and the timing operation of stream can also be set by periodic duty.It provides and terminates work(
Stream in capable of being used for suspending or stopping running.
The Data Management Unit includes the shared of the acquisition of data, the maintenance of data and data.Data acquisition is using text
File can be saved as big data automatically and put down by the mode that part uploads, including flat file, txt file, csv file types, system
Number of units needs specified file data delimiter and logarithm according to file standard format, user during carrying out file upload
It is described according to field.Realize effective management to big data.
Data maintenance includes checking details, deleting the operations such as data and downloading data.Check details:Check data
Essential information, storage information of data etc..Delete data:Support the delete operation to data.Downloading data:Allow former format
The locally downloading file system of data.
Platform provides data sharing management, and user can freely distribute the data of oneself and be used for other user.
The operational management unit includes the monitoring of the monitoring and operation details to log, the fortune each time of data flow
Row record, runs the record of details and is managed to data flow operating status.Operation on operational management unit monitors work(
It can include the monitoring of the monitoring and operation details to log.Log include instance name, the time started, at the end of
Between, state, operation duration etc..
Running details includes:One, the overall process of stream example operation is presented in the complete stream design drawing of displaying on stream design drawing
Information, and component difference operating status is indicated by different colors;Two, the process logs details list of displaying example operation,
Error message is searched by checking the event in daily record details to running report an error;Three, the operation letter of all nodes is checked
Breath.Four, the historical data of convection current log collection is checked.
Condition managing refers to that convection current operating status is managed, including operation, termination, failure and completes, and can choose certain
The operations such as a stream carries out periodic duty or convection current is run, terminated.If stream has been configured with periodic duty plan,
It can click and check plan details.Allow simultaneously to planning into the operations such as edlin or deletion.
After the completion of dataflow design, stream is prepared state, and operation is clicked under the state, and the state of stream is changed in operation;
For running stream, the operation for terminating and can interrupting the stream is clicked, the state of stream is changed to terminate;Nothing in operational process is flowed to appoint
What error message, after the completion of operation, the state of stream is changed to be completed.
Above-described specific implementation mode has carried out further the purpose of the present invention, technical solution and advantageous effect
It is described in detail, it should be understood that the foregoing is merely the specific implementation mode of the present invention, is not intended to limit the present invention
Protection domain, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should all include
Within protection scope of the present invention.
Claims (9)
1.PHM platforms, which is characterized in that including
Modeling space unit:Including data flow structure module, stream management module, stream correction verification module, stream operation module and control
Platform output module, user obtains data using data flow structure module, complete in a manner of the easy graphical operation of dragging and line
At the structure of data flow, management operation and validity school are then carried out to data stream by stream management module and stream correction verification module
It tests, assists user to complete model running and debugging, feedback result is then supplied to console output module, it is defeated by console
Go out window by the execution feedback result of stream carry out visualization show, be finally completed entire modeling process;
Data Management Unit:External file import feature is provided, includes that the source data that modeling space unit is needed imports big number
According to platform, then data are uniformly managed;
Operational management unit:For providing various data stream scheduling operation reserves, realize that operation monitoring, the operation to data flow are detailed
Feelings record and condition managing;
System Management Unit:Including user management, Role Management, rights management, unit module management, information department management and day
Will management ensures entire platform normal operation;
Page display unit:Display platform resource distribution situation, special time period operation data flow statistical analysis and be
Each user provides platform operation entry, provides an independent personal space for each user, only shows and use in personal space
The content of family individual, including the model that runs recently, data and log.
2. PHM platforms according to claim 1, which is characterized in that the data flow structure module includes being obtained to data
Take, the storage of data, data prediction, modeling algorithm and model evaluation.
3. PHM platforms according to claim 2, which is characterized in that the modeling algorithm includes according to data analysis tool
Statistical model, mining algorithm and signal processing are carried out to data stream, mining algorithm is exactly found out from a large amount of data implicit
The useful information in data, signal processing are that will be sensed by the fast Fourier transform function in signal processing operator
The time-domain signal of device acquisition is converted to frequency domain and time and frequency zone signal, to extract time domain, frequency domain and the time and frequency zone feature of signal, from
And react the operating status of platform.
4. PHM platforms according to claim 3, which is characterized in that found out from a large amount of data by modeling algorithm hidden
Information useful in data is contained in by the evaluation component in model evaluation to assess the accuracy of information, feasibility, Yi Jishi
The property used.
5. PHM platforms according to claim 2, which is characterized in that the data flow structure module further includes miscellaneous function,
Painting canvas shortcut key is added to support.
6. PHM platforms according to claim 1, which is characterized in that flow tube reason module includes the operation to data flow
Management, data flow classification management and data stream property management, stream correction verification module include the verification of convection current node and stream logic verify, stream
Operation module carries out data processing and operating between stream node by running stream.
7. PHM platforms according to claim 1, which is characterized in that the Data Management Unit includes the acquisition of data, number
According to maintenance and data it is shared.
8. PHM platforms according to claim 1, which is characterized in that the operational management unit includes to log
The monitoring of monitoring and operation details, the log each time of data flow run the record of details and run shape to data flow
State is managed.
9. PHM platforms according to claim 1, which is characterized in that the page display unit include showing resource configuration,
The activity of the user and the recent record of user.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810432744.9A CN108694448A (en) | 2018-05-08 | 2018-05-08 | PHM platforms |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810432744.9A CN108694448A (en) | 2018-05-08 | 2018-05-08 | PHM platforms |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108694448A true CN108694448A (en) | 2018-10-23 |
Family
ID=63847287
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810432744.9A Pending CN108694448A (en) | 2018-05-08 | 2018-05-08 | PHM platforms |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108694448A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109886328A (en) * | 2019-02-14 | 2019-06-14 | 国网浙江省电力有限公司电力科学研究院 | A kind of electric car electrically-charging equipment failure prediction method and system |
CN110009117A (en) * | 2019-03-28 | 2019-07-12 | 中车青岛四方机车车辆股份有限公司 | Mechanism model generation method and device based on train status data |
CN110188040A (en) * | 2019-05-21 | 2019-08-30 | 江苏锐天信息科技有限公司 | A kind of software platform for software systems fault detection and health state evaluation |
CN112486475A (en) * | 2020-12-03 | 2021-03-12 | 成都大数据产业技术研究院有限公司 | Visual business modeling and model management system based on big data platform |
CN112685380A (en) * | 2020-12-03 | 2021-04-20 | 成都大数据产业技术研究院有限公司 | Big data value discovery and application innovation platform system |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020184178A1 (en) * | 2001-06-04 | 2002-12-05 | Honeywell International, Inc. | Adaptive knowledge management system for vehicle trend monitoring, health management and preventive maintenance |
CN103853821A (en) * | 2014-02-21 | 2014-06-11 | 河海大学 | Method for constructing multiuser collaboration oriented data mining platform |
CN106020154A (en) * | 2016-07-12 | 2016-10-12 | 中国石油化工股份有限公司 | Safe dynamic health assessment method and assessment system for ethylene production |
CN107526600A (en) * | 2017-09-05 | 2017-12-29 | 成都优易数据有限公司 | A kind of visual numeric simulation analysis platform and its data cleaning method based on hadoop and spark |
CN107697107A (en) * | 2017-09-12 | 2018-02-16 | 中国中车股份有限公司 | A kind of train groups prognostic and health management ground intelligent processing system and method |
-
2018
- 2018-05-08 CN CN201810432744.9A patent/CN108694448A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020184178A1 (en) * | 2001-06-04 | 2002-12-05 | Honeywell International, Inc. | Adaptive knowledge management system for vehicle trend monitoring, health management and preventive maintenance |
CN103853821A (en) * | 2014-02-21 | 2014-06-11 | 河海大学 | Method for constructing multiuser collaboration oriented data mining platform |
CN106020154A (en) * | 2016-07-12 | 2016-10-12 | 中国石油化工股份有限公司 | Safe dynamic health assessment method and assessment system for ethylene production |
CN107526600A (en) * | 2017-09-05 | 2017-12-29 | 成都优易数据有限公司 | A kind of visual numeric simulation analysis platform and its data cleaning method based on hadoop and spark |
CN107697107A (en) * | 2017-09-12 | 2018-02-16 | 中国中车股份有限公司 | A kind of train groups prognostic and health management ground intelligent processing system and method |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109886328A (en) * | 2019-02-14 | 2019-06-14 | 国网浙江省电力有限公司电力科学研究院 | A kind of electric car electrically-charging equipment failure prediction method and system |
CN110009117A (en) * | 2019-03-28 | 2019-07-12 | 中车青岛四方机车车辆股份有限公司 | Mechanism model generation method and device based on train status data |
CN110188040A (en) * | 2019-05-21 | 2019-08-30 | 江苏锐天信息科技有限公司 | A kind of software platform for software systems fault detection and health state evaluation |
CN112486475A (en) * | 2020-12-03 | 2021-03-12 | 成都大数据产业技术研究院有限公司 | Visual business modeling and model management system based on big data platform |
CN112685380A (en) * | 2020-12-03 | 2021-04-20 | 成都大数据产业技术研究院有限公司 | Big data value discovery and application innovation platform system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108694448A (en) | PHM platforms | |
Yang et al. | A system architecture for manufacturing process analysis based on big data and process mining techniques | |
CN112396404A (en) | Data center system | |
CN111008197A (en) | Data center design method for power marketing service system | |
US20060184410A1 (en) | System and method for capture of user actions and use of capture data in business processes | |
CN110990391A (en) | Integration method and system of multi-source heterogeneous data, computer equipment and storage medium | |
CN106295983A (en) | Power marketing data visualization statistical analysis technique and system | |
CN112181960B (en) | Intelligent operation and maintenance framework system based on AIOps | |
CN107066500B (en) | Power grid mass data quality verification method based on PMS model | |
CN107832567A (en) | A kind of System Effectiveness Cooperative Assessment System and appraisal procedure | |
CN106293891A (en) | Multidimensional investment target measure of supervision | |
CA2871942A1 (en) | Business process analytics | |
CN109639791A (en) | Cloud workflow schedule method and system under a kind of container environment | |
CN112148260A (en) | Decision engine implementation method, device, equipment and storage medium | |
CN107301498A (en) | Generalized information management system for the military service monitoring of high-speed railway wheel rail relation | |
CN111190814B (en) | Method and device for generating software test case, storage medium and terminal | |
CN103164774A (en) | Automobile complete vehicle development system based on workflow | |
CN111538720A (en) | Method and system for cleaning basic data in power industry | |
CN103473305A (en) | Method and system for performing decision-making process show in statistic analysis | |
CN107544895A (en) | Hadoop big datas platform testing system and method | |
CN113506098A (en) | Power plant metadata management system and method based on multi-source data | |
Kropp et al. | An approach to data driven process discovery in the cost estimation process of a construction company | |
CN117493412A (en) | Digital base operation management system | |
CN112686773A (en) | Method for constructing power metering full-link key service abnormity positioning model based on fusion service topology | |
Wahyudi et al. | Improving company performance by the correctness of management decision through implementation dashboard using Power BI tools (case study at company Y) |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181023 |