CN109800894A - One kind is based on deep learning discovery metering automation pipeline stall diagnostic method and diagnostic system - Google Patents
One kind is based on deep learning discovery metering automation pipeline stall diagnostic method and diagnostic system Download PDFInfo
- Publication number
- CN109800894A CN109800894A CN201910059363.5A CN201910059363A CN109800894A CN 109800894 A CN109800894 A CN 109800894A CN 201910059363 A CN201910059363 A CN 201910059363A CN 109800894 A CN109800894 A CN 109800894A
- Authority
- CN
- China
- Prior art keywords
- data
- deep learning
- module
- fault diagnosis
- platform
- 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
- 238000013135 deep learning Methods 0.000 title claims abstract description 30
- 238000002405 diagnostic procedure Methods 0.000 title claims abstract description 15
- 238000003745 diagnosis Methods 0.000 claims abstract description 35
- 238000010606 normalization Methods 0.000 claims abstract description 13
- 238000010801 machine learning Methods 0.000 claims abstract description 9
- 238000012795 verification Methods 0.000 claims abstract description 7
- 238000000605 extraction Methods 0.000 claims abstract description 4
- 238000002372 labelling Methods 0.000 claims abstract description 4
- 239000006185 dispersion Substances 0.000 claims description 12
- 238000003066 decision tree Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 claims description 5
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 5
- 230000008569 process Effects 0.000 abstract description 3
- 238000012631 diagnostic technique Methods 0.000 abstract description 2
- 238000005259 measurement Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The present invention relates to pipeline stall diagnostic techniques fields, and it discloses a kind of based on deep learning discovery metering automation pipeline stall diagnostic method, include the following steps, data acquisition, for acquiring the daily record data and external data of smart machine and sensor group feedback data, system operation;Data normalization is standardized the data in above-mentioned steps, and treated data by kafka stream process platform are sent to Hive data warehouse;Feature extraction, fuzzy logic and labeling are carried out to the data in above-mentioned steps based on Spark MLlib platform, machine learning is carried out, obtains fault diagnosis model;Real-time diagnosis is carried out to automatic assembly line with the fault diagnosis model in above-mentioned steps, early warning is carried out to failure.Metering automation pipeline stall diagnostic method should be found based on deep learning, can construct the algorithm model of automatic Verification failure by deep learning according to collected data, and diagnose to failure according to the exceptional value collected in data.
Description
Technical field
The present invention relates to pipeline stall diagnostic techniques fields, specially a kind of to find metering automation based on deep learning
Pipeline stall diagnostic method.
Background technique
It is intelligent with the development of China's power information to deepen continuously with smart grid, artificial intelligence technology
Measuring equipment demand increasingly increase, force the automation, standardization, procedure of measurement verification, and develop to intelligent direction
To improve calibrating quality and efficiency, integrated, the intelligentized research in management through quantification field just becomes the heat currently studied extensively
One of point project.With the construction of automatic calibration system, integrated level, the complexity of calibration equipment are higher and higher, by maintenance
The experience of personnel, which repairs maintenance, becomes more and more difficult.People are increasing to the dependence of calibrating assembly line, once calibrating
Assembly line breaks down, caused by calibration operation fault and loss be artificial several times even tens times.Therefore event how is carried out
Barrier early warning and quick positioning failure are the bases of measurement verification Artificial Intelligence Development, are that full-automatic and intelligence is realized in measurement verification
The guarantee of change.
These intelligence calibrating links of country various regions calibrating assembly line pass through the modes such as various sensor technologies, message at present
Collection system operation data, system carry out failure by the mode that hardware device reports an error by the data of the single hardware device of collection
Alarm produce mass data since assembly line more tends to be complicated, hardware pipeline manufacturer focuses mainly on hardware device,
The data of the history of assembly line are not saved, the mode of big data cannot be used to carry out big data to automatic assembly line
Analysis, cannot analyse the key point for influencing assembly line and the important feature that has an impact;In metering automation assembly line, respectively
A equipment is cooperated and is influenced each other, and existing equipment is reported an error by single hardware, not from entire assembly line and flowing water
The mutual influence of line is gone to consider and be analyzed, and holistic diagnosis effect is bad.
Summary of the invention
It is an object of the invention to overcome the deficiencies of the prior art and provide one kind to find metering automation based on deep learning
Pipeline stall diagnostic method.
In order to solve the above technical problems, the technical solution adopted by the present invention is that, it mainly comprises the steps that
S1: data acquisition, for acquiring the daily record data of smart machine and sensor group feedback data, system operation and outer
Portion's data;
S2: data normalization is standardized the data acquired in step S1, and data pass through by treated
Kafka stream process platform is sent to Hive data warehouse;
S3: deep learning, based on Spark MLlib platform to after step S2 Plays data carry out feature extraction,
Fuzzy logic and labeling carry out machine learning, obtain fault diagnosis model;
S4: fault diagnosis, with the fault diagnosis model in S3 to automatic assembly line carry out real-time diagnosis, to failure into
Row early warning.
Preferably, in step s 2, data normalization is using useAlgorithm is handled, by a columns value tag
Value be scaled to mean value be 0, variance be 1 state.
Preferably, in step s3, Spark MLlib platform contain KMeans clustering algorithm, Decision Tree Algorithm and
Neural network deep learning algorithm, wherein the mathematics pattern of KMeans clustering algorithm are as follows:Decision tree point
The mathematics pattern of class algorithm are as follows:
The mathematics pattern of neural network deep learning algorithm are as follows:
Y=fθ(x)=s (Wx+b), x '=gθ' (y)=s (W ' y+b '), L (x, x ')=L (x, g (f (x))).
Preferably, in step s3, fault diagnosis model is assessed using Calinski-Harabaz index, for k cluster,
Calinski-Harabaz score s be as the ratio in dispersion average value between cluster and within-cluster groups between dispersion to
Out:
Wherein Bk is dispersion matrix between group, and Wk is by dispersion matrix in group defined below:
Wherein n is the points in data, cqFor the point set in cluster (cluster) q, nqFor the points in cluster (cluster) q.
Preferably, the smart machine in the step S1 is the sensing equipment of automatic Verification assembly line, includes at least machinery
Hand, measurement verification center single-phase electric energy meter, three-phase electric energy meter, low-voltage current mutual inductor and acquisition terminal, the sensor group is extremely
It include less temperature sensor, sensor noise and vibrating sensor, the external data includes external industrial big data data.
The present invention also provides one kind based on deep learning discovery metering automation pipeline stall diagnostic method and diagnosis system
System, including data source modules, AI platform and fault diagnosis module, the AI platform respectively with data source modules and fault diagnosis mould
Block signal connection, the data source modules include smart machine and the sensor for feedback data, system log and external number
According to the AI platform includes data normalization module, data transmission module, data memory module, data dynamic capacity-expanding/capacity reducing mould
Block, machine learning module and learning outcome memory module, the data normalization module is connect with data source modules signal, described
Learning outcome memory module is connect by RESTFul platform with fault diagnosis module signal.
Compared with prior art, the beneficial effects of the present invention are:
It is provided by the invention a kind of based on deep learning discovery metering automation pipeline stall diagnostic method, pass through acquisition
The data and external industrial big data number of related transducer equipment is fed back on automatic assembly line information data and system log
According to, and deep learning is carried out by machine learning algorithm to above data, fault diagnosis model is obtained, by the model to automatic
Change assembly line and carries out automatic fault diagnosis and early warning.
It is provided by the invention a kind of based on deep learning discovery metering automation pipeline stall diagnostic system, pass through setting
Data source modules, AI platform and fault diagnosis module intuitively carry out failure calibrating to automatic assembly line convenient for staff,
It is also convenient for getting information about automatic assembly line progress fault state.
Detailed description of the invention
Fig. 1 is a kind of stream that metering automation pipeline stall diagnostic method is found based on deep learning proposed by the present invention
Cheng Tu;
Fig. 2 is a kind of frame that metering automation pipeline stall diagnostic system is found based on deep learning proposed by the present invention
Composition.
Specific embodiment
The present invention is further illustrated With reference to embodiment.Wherein, attached drawing only for illustration,
What is indicated is only schematic diagram, rather than pictorial diagram, should not be understood as the limitation to this patent;Reality in order to better illustrate the present invention
Example is applied, the certain components of attached drawing have omission, zoom in or out, and do not represent the size of actual product;To those skilled in the art
For, the omitting of some known structures and their instructions in the attached drawings are understandable.
The same or similar label correspond to the same or similar components in the attached drawing of the embodiment of the present invention;It is retouched in of the invention
In stating, it is to be understood that if the orientation or positional relationship for having the instructions such as term " on ", "lower", "left", "right" is based on attached drawing
Shown in orientation or positional relationship, be merely for convenience of description of the present invention and simplification of the description, rather than indication or suggestion is signified
Device or element must have a particular orientation, be constructed and operated in a specific orientation, therefore positional relationship is described in attached drawing
Term only for illustration, should not be understood as the limitation to this patent, for the ordinary skill in the art, can
To understand the concrete meaning of above-mentioned term as the case may be.
Embodiment
Implement 1:
As shown in Figure 1, a kind of find metering automation pipeline stall diagnostic method, including following step based on deep learning
It is rapid:
S1: data acquisition, for acquiring the daily record data of smart machine and sensor group feedback data, system operation and outer
Portion's data;
S2: data normalization is standardized the data acquired in step S1, and data pass through by treated
Kafka stream process platform is sent to Hive data warehouse;
S3: deep learning, based on Spark MLlib platform to after step S2 Plays data carry out feature extraction,
Fuzzy logic and labeling carry out machine learning, obtain fault diagnosis model;
S4: fault diagnosis, with the fault diagnosis model in S3 to automatic assembly line carry out real-time diagnosis, to failure into
Row early warning.
Wherein, in step s 2, data normalization is using useAlgorithm is handled, by a columns value tag
It is 0 that value, which is scaled to mean value, the state that variance is 1.
In addition, in step s3, Spark MLlib platform contains KMeans clustering algorithm, Decision Tree Algorithm and mind
Through network depth learning algorithm, the wherein mathematics pattern of KMeans clustering algorithm are as follows:
The mathematics pattern of Decision Tree Algorithm are as follows:
The mathematics pattern of neural network deep learning algorithm are as follows:
Y=fθ(x)=s (Wx+b), x '=gθ' (y)=s (W ' y+b '), L (x, x ')=L (x, g (f (x))).
In addition, in step s3, fault diagnosis model is assessed using Calinski-Harabaz index, for k cluster,
Calinski-Harabaz score s be as the ratio in dispersion average value between cluster and within-cluster groups between dispersion to
Out:
Wherein Bk is dispersion matrix between group, and Wk is by dispersion matrix in group defined below:
Wherein n is the points in data, cqFor the point set in cluster (cluster) q, nqFor the points in cluster (cluster) q.
Wherein, smart machine in step sl is the sensing equipment of automatic Verification assembly line, includes at least manipulator, meter
Amount calibrating center single-phase electric energy meter, three-phase electric energy meter, low-voltage current mutual inductor and acquisition terminal, the sensor group include at least
Temperature sensor, sensor noise and vibrating sensor, external data include external industrial big data data.
Embodiment 2:
As shown in Fig. 2, a kind of find metering automation pipeline stall diagnostic system, including data source based on deep learning
Module 1, AI platform 2 and fault diagnosis module 3, AI platform 2 are connect with data source modules 1 and 3 signal of fault diagnosis module respectively,
Data source modules 1 include smart machine 11 and the sensor 12 for feedback data, system log 13 and external data 14, and AI is flat
Platform 2 includes data normalization module 21, data transmission module 22, data memory module 23, data dynamic capacity-expanding/capacity reducing module
24, machine learning module 25 and learning outcome memory module 26, data normalization module 21 are connect with 1 signal of data source modules,
Learning outcome memory module 26 is connect by RESTFul platform 4 with 3 signal of fault diagnosis module.
In the present embodiment, pass through the information data and system of related transducer equipment feedback on acquisition automatic assembly line
The data and external industrial big data data of log, and deep learning is carried out by machine learning algorithm to above data, it obtains
Fault diagnosis model carries out automatic fault diagnosis and early warning to automatic assembly line by the model, by the way that data source mould is arranged
Block 1, AI platform 2 and fault diagnosis module 3 intuitively carry out failure calibrating to automatic assembly line convenient for staff, also just
Fault state is carried out in getting information about automatic assembly line.
Obviously, the above embodiment of the present invention be only to clearly illustrate example of the present invention, and not be pair
The restriction of embodiments of the present invention.For those of ordinary skill in the art, may be used also on the basis of the above description
To make other variations or changes in different ways.There is no necessity and possibility to exhaust all the enbodiments.It is all this
Made any modifications, equivalent replacements, and improvements etc., should be included in the claims in the present invention within the spirit and principle of invention
Protection scope within.
Claims (6)
1. one kind finds metering automation pipeline stall diagnostic method based on deep learning, which is characterized in that including following step
It is rapid:
S1: data acquisition, for acquiring the daily record data and external number of smart machine and sensor group feedback data, system operation
According to;
S2: data normalization is standardized the data acquired in step S1, and data pass through kafka stream by treated
Processing platform is sent to Hive data warehouse;
S3: deep learning carries out feature extraction to the data after step S2 Plays based on SparkMLlib platform, obscures and patrol
It collects and labeling, progress machine learning obtains fault diagnosis model;
S4: fault diagnosis carries out real-time diagnosis to automatic assembly line with the fault diagnosis model in S3, carries out to failure pre-
It is alert.
2. a kind of deep learning that is based on according to claim 1 finds metering automation pipeline stall diagnostic method,
Be characterized in that: in step s 2, data normalization is using useAlgorithm is handled, and the value of a columns value tag is contracted
Putting into mean value is 0, the state that variance is 1.
3. a kind of deep learning that is based on according to claim 1 finds metering automation pipeline stall diagnostic method,
Be characterized in that: in step s3, SparkMLlib platform contains KMeans clustering algorithm, Decision Tree Algorithm and neural network
Deep learning algorithm, wherein the mathematics pattern of KMeans clustering algorithm are as follows:Decision Tree Algorithm
Mathematics pattern are as follows:
The mathematics pattern of neural network deep learning algorithm are as follows:
Y=fθ(x)=s (Wx+b), x '=gθ' (y)=s (W ' y+b '), L (x, x ')=L (x, g (f (x))).
4. a kind of deep learning that is based on according to claim 1 finds metering automation pipeline stall diagnostic method,
Be characterized in that: in step s3, fault diagnosis model is assessed using Calinski-Harabaz index, for k cluster,
Calinski-Harabaz score s be as the ratio in dispersion average value between cluster and within-cluster groups between dispersion to
Out:
Wherein Bk is dispersion matrix between group, and Wk is by dispersion matrix in group defined below:
Wherein n is the points in data, cqFor the point set in cluster (cluster) q, nqFor the points in cluster (cluster) q.
5. a kind of deep learning that is based on according to claim 1 finds metering automation pipeline stall diagnostic method,
Be characterized in that: the smart machine in the step S1 is the sensing equipment of automatic Verification assembly line, includes at least manipulator, metering
Calibrating center single-phase electric energy meter, three-phase electric energy meter, low-voltage current mutual inductor and acquisition terminal, the sensor group include at least temperature
Spending sensor, sensor noise and vibrating sensor, the external data includes external industrial big data data.
6. one kind finds that metering automation pipeline stall diagnostic system, including data source modules (1), AI are put down based on deep learning
Platform (2) and fault diagnosis module (3), it is characterised in that: the AI platform (2) respectively with data source modules (1) and fault diagnosis
Module (3) signal connection, the data source modules (1) include smart machine (11) and for feedback data sensor (12),
System log (13) and external data (14), the AI platform (2) include data normalization module (21), data transmission module
(22), data memory module (23), data dynamic capacity-expanding/capacity reducing module (24), machine learning module (25) and learning outcome are deposited
It stores up module (26), the data normalization module (21) connect with data source modules (1) signal, the learning outcome memory module
(26) it is connect by RESTFul platform (4) with fault diagnosis module (3) signal.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910059363.5A CN109800894A (en) | 2019-01-22 | 2019-01-22 | One kind is based on deep learning discovery metering automation pipeline stall diagnostic method and diagnostic system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910059363.5A CN109800894A (en) | 2019-01-22 | 2019-01-22 | One kind is based on deep learning discovery metering automation pipeline stall diagnostic method and diagnostic system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109800894A true CN109800894A (en) | 2019-05-24 |
Family
ID=66559975
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910059363.5A Pending CN109800894A (en) | 2019-01-22 | 2019-01-22 | One kind is based on deep learning discovery metering automation pipeline stall diagnostic method and diagnostic system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109800894A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111985558A (en) * | 2020-08-19 | 2020-11-24 | 安徽蓝杰鑫信息科技有限公司 | Electric energy meter abnormity diagnosis method and system |
CN113110385A (en) * | 2021-04-16 | 2021-07-13 | 广东电网有限责任公司计量中心 | Decision tree-based start-stop early warning method and device for metering automatic verification system |
CN113156529A (en) * | 2021-05-07 | 2021-07-23 | 广东电网有限责任公司计量中心 | Start-stop control method, system, terminal and storage medium of metrological verification assembly line |
CN117233540A (en) * | 2023-11-15 | 2023-12-15 | 广东电网有限责任公司江门供电局 | Metering pipeline fault detection method and system based on deep learning |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105974356A (en) * | 2016-07-22 | 2016-09-28 | 国网浙江省电力公司电力科学研究院 | Fault diagnosis method for electric power metering automatic verification assembly line |
CN106019195A (en) * | 2016-07-22 | 2016-10-12 | 国网浙江省电力公司电力科学研究院 | Electric power measurement automation verification assembly line fault diagnosis system |
CN107817787A (en) * | 2017-11-29 | 2018-03-20 | 华南理工大学 | A kind of intelligent producing line robotic failure diagnostic method based on machine learning |
-
2019
- 2019-01-22 CN CN201910059363.5A patent/CN109800894A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105974356A (en) * | 2016-07-22 | 2016-09-28 | 国网浙江省电力公司电力科学研究院 | Fault diagnosis method for electric power metering automatic verification assembly line |
CN106019195A (en) * | 2016-07-22 | 2016-10-12 | 国网浙江省电力公司电力科学研究院 | Electric power measurement automation verification assembly line fault diagnosis system |
CN107817787A (en) * | 2017-11-29 | 2018-03-20 | 华南理工大学 | A kind of intelligent producing line robotic failure diagnostic method based on machine learning |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111985558A (en) * | 2020-08-19 | 2020-11-24 | 安徽蓝杰鑫信息科技有限公司 | Electric energy meter abnormity diagnosis method and system |
CN113110385A (en) * | 2021-04-16 | 2021-07-13 | 广东电网有限责任公司计量中心 | Decision tree-based start-stop early warning method and device for metering automatic verification system |
CN113156529A (en) * | 2021-05-07 | 2021-07-23 | 广东电网有限责任公司计量中心 | Start-stop control method, system, terminal and storage medium of metrological verification assembly line |
CN117233540A (en) * | 2023-11-15 | 2023-12-15 | 广东电网有限责任公司江门供电局 | Metering pipeline fault detection method and system based on deep learning |
CN117233540B (en) * | 2023-11-15 | 2024-02-20 | 广东电网有限责任公司江门供电局 | Metering pipeline fault detection method and system based on deep learning |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109800894A (en) | One kind is based on deep learning discovery metering automation pipeline stall diagnostic method and diagnostic system | |
CN106199276B (en) | The intelligent diagnosis system and method for exception information in a kind of power information acquisition system | |
CN102361014B (en) | State monitoring and fault diagnosis method for large-scale semiconductor manufacture process | |
CN110320892A (en) | The sewage disposal device fault diagnosis system and method returned based on Lasso | |
CN104808585B (en) | A kind of quick inspection method of lathe health status | |
CN104267668B (en) | Bayes-method-based spaceflight valve part fault diagnosis method in machining process | |
CN109813544B (en) | A kind of rotating machinery Incipient Fault Diagnosis method and system based on on-line monitoring | |
CN106655522A (en) | Master station system suitable for operation and maintenance management of secondary equipment of power grid | |
CN108376298A (en) | A kind of Wind turbines generator-temperature detection fault pre-alarming diagnostic method | |
CN107273924B (en) | Multi-data fusion power plant fault diagnosis method based on fuzzy clustering analysis | |
CN109634233A (en) | Industrial big data intellectual analysis decision-making technique, readable storage medium storing program for executing and terminal | |
CN107679089A (en) | A kind of cleaning method for electric power sensing data, device and system | |
CN108764305A (en) | A kind of improved colony intelligence machine learning fault diagnosis system | |
CN106997513A (en) | The distribution net equipment state evaluation system analyzed based on big data | |
CN110375983A (en) | Failsafe valve real-time diagnosis system and diagnostic method based on time series analysis | |
CN109614451A (en) | Industrial big data intellectual analysis decision making device | |
CN112036087A (en) | Multi-strategy fused nuclear power key equipment fault diagnosis method and system | |
CN116185757A (en) | Intelligent monitoring system for energy consumption of machine room | |
CN115614292A (en) | Vibration monitoring device and method for vertical water pump unit | |
CN109324264B (en) | Identification method and device for abnormal value of power distribution network line impedance data | |
CN108416386A (en) | A kind of method and system judged extremely for Hydropower Unit bearing temperature | |
CN117521498A (en) | Charging pile guide type fault diagnosis prediction method and system | |
CN116308293A (en) | Intelligent agricultural equipment management system and method based on digital platform | |
CN116032003A (en) | Power grid inspection method and system based on power private network data processing | |
CN114442543A (en) | Computer monitoring method suitable for early warning of hydropower station fault |
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 |
Application publication date: 20190524 |
|
RJ01 | Rejection of invention patent application after publication |