CN207516464U - A kind of Portable metering automatization terminal trouble-shooter - Google Patents
A kind of Portable metering automatization terminal trouble-shooter Download PDFInfo
- Publication number
- CN207516464U CN207516464U CN201721548283.9U CN201721548283U CN207516464U CN 207516464 U CN207516464 U CN 207516464U CN 201721548283 U CN201721548283 U CN 201721548283U CN 207516464 U CN207516464 U CN 207516464U
- Authority
- CN
- China
- Prior art keywords
- module
- data
- human
- computer interaction
- detection sub
- 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.)
- Expired - Fee Related
Links
- 238000001514 detection method Methods 0.000 claims abstract description 30
- 230000003993 interaction Effects 0.000 claims abstract description 22
- 238000003745 diagnosis Methods 0.000 claims abstract description 9
- 230000001419 dependent effect Effects 0.000 claims abstract description 4
- 238000004891 communication Methods 0.000 claims description 39
- 238000005070 sampling Methods 0.000 claims description 18
- 238000012360 testing method Methods 0.000 claims description 4
- 238000007689 inspection Methods 0.000 claims 1
- 230000002159 abnormal effect Effects 0.000 abstract description 9
- 230000008014 freezing Effects 0.000 abstract description 4
- 238000007710 freezing Methods 0.000 abstract description 4
- 238000004458 analytical method Methods 0.000 abstract description 3
- 238000012631 diagnostic technique Methods 0.000 abstract description 3
- 238000012545 processing Methods 0.000 abstract description 3
- 230000011664 signaling Effects 0.000 abstract description 3
- 238000012549 training Methods 0.000 description 19
- 238000013528 artificial neural network Methods 0.000 description 14
- 239000013598 vector Substances 0.000 description 13
- 238000007726 management method Methods 0.000 description 10
- 238000012706 support-vector machine Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 241001269238 Data Species 0.000 description 2
- 230000004913 activation Effects 0.000 description 2
- 238000000034 method Methods 0.000 description 2
- 238000013024 troubleshooting Methods 0.000 description 2
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000013011 mating Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000004218 nerve net Anatomy 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Test And Diagnosis Of Digital Computers (AREA)
Abstract
The utility model is related to metering automation terminal abnormal diagnostic techniques, and in particular to a kind of Portable metering automatization terminal trouble-shooter specifically includes detection sub-module, power management module and human-computer interaction module；Detection sub-module is connect with human-computer interaction module；Power management module is connect respectively with detection sub-module and human-computer interaction module；Human-computer interaction module carries out configuration information for inputting dependent instruction, and detection sub-module is used for detection terminal failure；Power management module provides power supply for detection sub-module and human-computer interaction module；The utility model by detection sub-module acquire the control data of metering automation terminal device, remote signalling data, freezing data, pair when data, rate of qualified voltage data, and processing analysis is carried out to above-mentioned data, and then judge the failure of metering automation terminal, failure cause to terminal and guilty culprit point can be accurately positioned, improve the diagnosis efficiency of metering automation terminal device.
Description
Technical field
The utility model is related to metering automation terminal abnormal diagnostic techniques, and in particular to a kind of Portable metering automation
Terminal fault diagnostic device.
Background technology
Automated system is measured in China's grid company at present and is able to Large scale construction, while a large amount of mating meterings are certainly
Dynamicization terminal device also begins to be installed in metering node.With the continuous expansion of metering automation system scale, scene fortune
Metering automation terminal in row is continuously increased, and on-site terminal troubleshooting work is also continuously increased, this results in power grid public
Department needs to face larger maintenance cost pressure, so how fast, accurately to realize analysis to terminal fault and examine
Disconnected is that the major issue solved is badly in need of in current power enterprise.Portable movable detection terminal is safeguarded in metering automation system operation
Middle application, realize terminal blank screen, it is not online, can not the most common failures such as gathered data intelligent automation diagnosis, realize to terminal
The quick processing of failure improves the online rate of terminal, acquisition percentage of head rice, promotes O＆M efficiency, transported for situ metrology automatization terminal
Row is safeguarded and troubleshooting provides safeguard.
Metering automation terminal abnormal diagnostic techniques includes realizing that the wiring of metering automation terminal filed and interface fault are different
Often field diagnostic of diagnosis, terminal called channel abnormal and communication protocol compliance etc..And existing metering automation terminal is examined
Disconnected technology in face of complicated terminal system and in multifarious phenomenon of the failure, be difficult often basic reason to breaking down and
Trouble point accurately and rapidly judge and positioning.
Invention content
To solve the above-mentioned problems, the utility model provides a kind of Portable metering automatization terminal fault diagnosis dress
It puts, specific technical solution is as follows：
A kind of Portable metering automatization terminal trouble-shooter includes detection sub-module, power management module and man-machine
Interactive module；Detection sub-module is connect with human-computer interaction module；Power management module respectively with detection sub-module and human-computer interaction
Module connects；The human-computer interaction module carries out configuration information for inputting dependent instruction, and the detection sub-module is used to detect
Terminal fault；The power management module provides power supply for detection sub-module and human-computer interaction module；The detection sub-module packet
Include main control module, Local Communication Module, remote communication module, input and output module, AC sampling module and meter-reading module；It is described
Main control module respectively with Local Communication Module, remote communication module, input and output module, AC sampling module, meter-reading module and
Human-computer interaction module connects；
The main control module is used for control terminal fault detect, diagnosis and communication；
The Local Communication Module includes carrier communication module, micropower wireless communication module；
The data that the remote communication module acquires for upload；
The input and output module has the input of several way switch, the output of several way switch, is automated for test and measuring
Whether the input and output of terminal switch amount are normal；
The AC sampling module is for three-phase voltage data sampling, three-phase current data sampling；
The meter-reading module is for simulating ammeter, for providing the data of metering automation terminal copy reading ammeter.
Further, the three-phase current data sampling compatibility three-phase three-line system and three-phase four-wire system.
Further, the remote communication module uses GPRS 4G network communications.
The beneficial effects of the utility model are：The utility model acquires metering automation terminal device by detection sub-module
Control data, remote signalling data, freezing data, pair when data, rate of qualified voltage data, and to above-mentioned data carry out processing point
Analysis, and then judge the failure of metering automation terminal, the failure cause to terminal and guilty culprit point can be accurately positioned,
Improve the diagnosis efficiency of metering automation terminal device.
Description of the drawings
Fig. 1 is the structure diagram of the utility model；
Training step schematic diagrames of the Fig. 2 for Adaboost graders in the embodiment of the utility model；
Fig. 3 is the neural network structure figure containing autocoder in the embodiment of the utility model.
Specific embodiment
In order to be better understood from the utility model, the utility model is made in the following with reference to the drawings and specific embodiments further
Explanation：
As shown in Figure 1, a kind of Portable metering automatization terminal trouble-shooter includes detection sub-module, power management
Module and human-computer interaction module；Detection sub-module is connect with human-computer interaction module；Power management module respectively with detection sub-module
It is connected with human-computer interaction module；Human-computer interaction module carries out configuration information for inputting dependent instruction, and detection sub-module is used to examine
Survey terminal fault；Power management module provides power supply for detection sub-module and human-computer interaction module；Detection sub-module includes master control
Module, Local Communication Module, remote communication module, input and output module, AC sampling module and meter-reading module；Main control module point
Not with Local Communication Module, remote communication module, input and output module, AC sampling module, meter-reading module and human-computer interaction mould
Block connects.
Main control module is used for control terminal fault detect, diagnosis and communication；Specially three-phase voltage, three-phase current data
Acquisition, input and output buret reason：Such as 4 way switch amount I/O managements；Telecommunication management：As electric energy meter, concentrator communicate
(RS232, RS485 and infrared)；USB management, wireless communications management etc. and the functions such as local/remote communication module test are real
It is existing.
Local Communication Module includes carrier communication module, micropower wireless communication module.
Remote communication module is for uploading the data of acquisition, using GPRS 4G network communications.
Input and output module has the input of several way switch, the output of several way switch, for test and measuring automatization terminal
Whether On-off signal output is normal.
AC sampling module is for three-phase voltage data sampling, three-phase current data sampling；Wherein, three-phase current data are adopted
Sample is compatible with three-phase three-line system and three-phase four-wire system.
Meter-reading module is for simulating ammeter, for providing the data of metering automation terminal copy reading ammeter.
The operation principle of the utility model is：The control data, distant of detection sub-module acquisition metering automation terminal device
Letter data, freezing data, pair when data, rate of qualified voltage data, and these data are normalized, using neural network and
Support vector machines technology is combined, first with trained Ababoost graders and neural network to metering automation terminal
Various types of data classified and extract feature, support vector machines is recycled to find optimal classification for the data characteristics extracted
Face, failure cause and guilty culprit point to terminal are accurately positioned, and improve the diagnosis effect of metering automation terminal device
Rate.Concrete methods of realizing is as follows：
1st, gathered data：Detection sub-module acquires the Various types of data of metering automation terminal device, specifically includes control number
According to, remote signalling data, freezing data, pair when data, rate of qualified voltage data.
2nd, information normalized：Detection sub-module carries out the Various types of data of collected metering automation terminal device
The data sample of Various types of data is converted into the numerical value in the range of [0,1] by standard normalized, using as all kinds of Adaboost
The input variable of Weak Classifier；Wherein the Various types of data of collected metering automation terminal device is carried out at standard normalization
Reason, is converted into the numerical value in the range of [0,1] by the data sample of Various types of data, specifically includes following steps：
Equipped with K evaluation data target and the gathered data of M times, xijWhat the jth time data for the i-th item data index acquired
Value carries out it value r of jth time data acquisition of the i-th item data index after standard normalized must normalizeij：
WhereinThe maxima and minima of the value of jth time data acquisition respectively in K item datas index.
3rd, the training of all kinds of Adaboost graders weights：Training data is initialized to N number of data sample of Various types of data
Weight, each training sample start to be endowed identical weights：1/N passes through the collected abnormal data of fuzzy inputing method
It with the different faults type of output, is calculated by successive ignition, the updated set of data samples of weight is used to train next number
According to sample set, and each Adaboost Weak Classifiers that training is obtained are combined into Adaboost strong classifiers, and increase classification
The weight of the small Adaboost Weak Classifiers of error reduces the weight of the big Adaboost Weak Classifiers of error in classification rate.
Wherein, abnormal data includes voltage data, current data, Local Communication Module read-write data flow, telecommunication mould
Block data flow, On-off signal output state data.It is different that fault type includes voltage and current wiring error, Local Communication Module
Often, remote communication module is abnormal, output switch parameter output module is abnormal.
As shown in Fig. 2, the training of all kinds of Adaboost graders weights specifically includes following steps：
1) training data weight is initialized to N number of data sample of Various types of data, each training sample starts to be endowed
Identical weights：1/N；D1 is the weight matrix of training sample：
2) successive ignition is carried out, iterations are represented with m, in entire iterative process, for collected different abnormal
Different faults type caused by data is learnt using the set of data samples being distributed with weights, obtains basic classification device Gm (x)：
3) error in classification rate es of the Gm (x) on learning data sample set is calculatedm, i.e., by Gm(x) weight of misclassification sample
The sum of：
Wherein wmiWeight for each training sample of the m times iteration.
4) coefficient of Gm (x), wherein α are calculatedmRepresent that Gm (x) is analyzing the Adaboost graders of final failure cause
Significance level：
5) distribution of trained values weight is updated, is updated for the weight of next round, wherein Zm is weight standardizing factor：
Dm+1=(wm+1,1,wm+1,2...wm+1,i...,wm+1,N)；
6) after the completion of iteration, each Weak Classifier is combined so as to obtain final strong classifier：
4th, neural network is trained on the basis of the Adaboost strong classifiers：By metering automation terminal
Various failure causes and likelihood of failure pass through all kinds of numbers for the acquisition that A weighting daboost strong classifiers obtain as input quantity
According to as output；During training, using all kinds of failure causes as input variable, the historical datas of all kinds of failures as output variable,
Feature extraction and training are carried out using neural network.
Fig. 3 is the neural network structure schematic diagram containing autocoder, wherein carrying out feature extraction using neural network
The specific steps are：
1) a metering automation terminal fault sample set X={ x without label is giveni| 1≤i≤L }, xiRepresent failure sample
I-th of sample of this concentration, sample length m；The fault vectors of input are mapped by autocoder, output vector
Collection is combined into Y={ yi| 1≤i≤L }, hiRepresent the corresponding feature vector of i-th of fault sample, then H=f(W,b)(X)=sf(WX+
B), W is the weight matrix of neural network input layer and hidden layer, bias matrixes of the b between input layer and hidden layer, sfTo compile
The activation primitive of code device part；
2) the hidden layer output variable that encoder obtains is reconstructed into and is originally inputted variable：Output vector collection is combined intoThe length of output vector is identical with decoding prior fault vector length, and the analytic equation of decoder is：sgNeuron activation functions for decoder section；
3) by constantly minimizing the reconstructed error between output vector and input vectorRealize that extraction is special
The purpose of sign, reconstructed error areGradient descent method is recycled constantly to adjust a nerve net
The weight matrix and bias matrix of network input layer and hidden layer make reconstructed error reach minimum, and specific implementation formula is as follows：
Wherein, o is the learning rate of neural network；
4) reconstructed error between the output vector for above-mentioned formula calculating gained and input vector, is reversely passed with error
It broadcasts algorithm and constantly adjusts weights and biasing { W1,b1,W1',b1' so that structure error is minimum, completes the training of the neural first order；
Then, retain the encoder section of this grade, input vector of the characteristic layer output vector as next stage neural network input layer；
5) second level neural network is trained according to similary step 1) to step 4), repeats step 1) training to step 4)
Journey, to the training for completing afterbody neural network, when completion front is all trained, the hidden layer output of last layer is
Final feature vector.
5th, the Various types of data acquired using trained Ababoost graders and neural network to metering automation terminal
Classify, and feature is extracted for received signal, found using support vector machines for the feature extracted optimal
Classifying face, i.e., most accurate fault type classification.In the N kind fault types for metering automation terminal fault, for example, it is common
Type：Remote communication module failure, meter-reading module failure, AC sampling module failure, input/output module failure etc., for
Training sample (the x after feature is extracted using neural networki,yi), i=1,2 ..., N, yiFor category label, yi∈(1,-1)。
6th, using the support vector machine classifier of two classification, a N class grader can be constructed through the following steps：
1) the support vector machines classifier rules of N number of two classification are constructed：Construct the classification function f of training samplej(x),
J=1,2...N separates jth class sample and the training sample of other classifications (if training sample xiBelong to jth class, then sgn [fj
(xi)]=1, otherwise sgn [fj(xi)]=- 1)；
2) pass through Selection of Function fj(x), the classification F (x in j=1,2...N in N kinds classification corresponding to maximum valuei)=
argmax{f1(xi).,..fN(xi), a N class grader is constructed, which can realize will be per a kind of and remaining N-1 class
Fault sample separates, and is achieved in the purpose of metering automation terminal fault diagnostic classification.
The utility model is not limited to above-described specific embodiment, and the foregoing is merely the preferable of the utility model
Case study on implementation is not intended to limit the utility model, made within the spirit and principles of the present invention any
Modifications, equivalent substitutions and improvements etc., should be included within the scope of protection of this utility model.
Claims (3)
1. a kind of Portable metering automatization terminal trouble-shooter, it is characterised in that：Including detection sub-module, power management
Module and human-computer interaction module；Detection sub-module is connect with human-computer interaction module；Power management module respectively with detection sub-module
It is connected with human-computer interaction module；The human-computer interaction module carries out configuration information, the detection submodule for inputting dependent instruction
Block is used for detection terminal failure；The power management module provides power supply for detection sub-module and human-computer interaction module；The inspection
Submodule is surveyed to include main control module, Local Communication Module, remote communication module, input and output module, AC sampling module and copy
Table module；The main control module respectively with Local Communication Module, remote communication module, input and output module, AC sampling module,
Meter-reading module is connected with human-computer interaction module；
The main control module is used for control terminal fault detect, diagnosis and communication；
The Local Communication Module includes carrier communication module, micropower wireless communication module；
The electric energy data that the remote communication module acquires for upload；
The input and output module has the input of several way switch, the output of several way switch, for test and measuring automatization terminal
Whether On-off signal output is normal；
The AC sampling module is for three-phase voltage data sampling, three-phase current data sampling；
The meter-reading module is for simulating ammeter, for providing the data of metering automation terminal copy reading ammeter.
2. a kind of Portable metering automatization terminal trouble-shooter according to claim 1, it is characterised in that：It is described
Three-phase current data sampling is compatible with three-phase three-line system and three-phase four-wire system.
3. a kind of Portable metering automatization terminal trouble-shooter according to claim 1, it is characterised in that：It is described
Remote communication module uses GPRS 4G network communications.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201721548283.9U CN207516464U (en) | 2017-11-18 | 2017-11-18 | A kind of Portable metering automatization terminal trouble-shooter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201721548283.9U CN207516464U (en) | 2017-11-18 | 2017-11-18 | A kind of Portable metering automatization terminal trouble-shooter |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN207516464U true CN207516464U (en) | 2018-06-19 |
Family
ID=62539228
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201721548283.9U Expired - Fee Related CN207516464U (en) | 2017-11-18 | 2017-11-18 | A kind of Portable metering automatization terminal trouble-shooter |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN207516464U (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107817404A (en) * | 2017-11-18 | 2018-03-20 | 广西电网有限责任公司电力科学研究院 | A kind of Portable metering automatization terminal trouble-shooter and its diagnostic method |
-
2017
- 2017-11-18 CN CN201721548283.9U patent/CN207516464U/en not_active Expired - Fee Related
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107817404A (en) * | 2017-11-18 | 2018-03-20 | 广西电网有限责任公司电力科学研究院 | A kind of Portable metering automatization terminal trouble-shooter and its diagnostic method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107817404A (en) | A kind of Portable metering automatization terminal trouble-shooter and its diagnostic method | |
| CN110223196B (en) | Anti-electricity-stealing analysis method based on typical industry feature library and anti-electricity-stealing sample library | |
| CN110108914B (en) | Intelligent decision-making method, system, equipment and medium for preventing electricity stealing | |
| CN110097297A (en) | A kind of various dimensions stealing situation Intellisense method, system, equipment and medium | |
| CN110458230A (en) | A kind of distribution transforming based on the fusion of more criterions is with adopting data exception discriminating method | |
| CN110837866A (en) | XGboost-based electric power secondary equipment defect degree evaluation method | |
| CN105022021B (en) | A kind of state identification method of the Electric Energy Tariff Point Metering Device based on multiple agent | |
| CN110634080A (en) | Abnormal electricity utilization detection method, device, equipment and computer readable storage medium | |
| CN101738998B (en) | System and method for monitoring industrial process based on local discriminatory analysis | |
| CN109784276A (en) | A kind of voltage dip feature extraction based on DBN and temporarily drop source discrimination method | |
| CN109767054A (en) | Efficiency cloud appraisal procedure and edge efficiency gateway based on deep neural network algorithm | |
| CN107480698A (en) | Method of quality control based on multiple monitoring indexes | |
| CN108761377A (en) | A kind of electric energy metering device method for detecting abnormality based on long memory models in short-term | |
| CN111949535B (en) | Software defect prediction device and method based on open source community knowledge | |
| CN111060838A (en) | Medical electronic equipment switching power supply fault diagnosis method based on multi-dimensional feature fusion | |
| CN110309884A (en) | Electricity consumption data anomalous identification system based on ubiquitous electric power Internet of Things net system | |
| CN115511367B (en) | Intelligent quality management system of production line | |
| CN110379036A (en) | Intelligent substation patrol recognition methods, system, device and storage medium | |
| CN109214464A (en) | A kind of doubtful stealing customer identification device and recognition methods based on big data | |
| CN111551888A (en) | Electric energy meter metering data fault analysis method based on improved AdaBoost algorithm | |
| CN110837532A (en) | Method for detecting electricity stealing behavior of charging pile based on big data platform | |
| CN111143835B (en) | Non-invasive protection method for business logic of electric power metering system based on machine learning | |
| CN112507479A (en) | Oil drilling machine health state assessment method based on manifold learning and softmax | |
| CN115017828A (en) | Power cable fault identification method and system based on bidirectional long-short-time memory network | |
| CN115617784A (en) | Data processing system and processing method for informationized power distribution |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180619 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |