CN109800995A - A kind of grid equipment fault recognition method and system - Google Patents
A kind of grid equipment fault recognition method and system Download PDFInfo
- Publication number
- CN109800995A CN109800995A CN201910089088.1A CN201910089088A CN109800995A CN 109800995 A CN109800995 A CN 109800995A CN 201910089088 A CN201910089088 A CN 201910089088A CN 109800995 A CN109800995 A CN 109800995A
- Authority
- CN
- China
- Prior art keywords
- data
- device data
- equipment
- sample
- major class
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of grid equipment fault recognition method and systems.The described method includes: collecting the major class device data of grid equipment；The major class device data includes operating status, equipment account, historical failure, equipment deficiency, overhaul of the equipments, facility environment and equipment safety；Each major class device data is classified as multiple group device datas by naive Bayesian machine learning model；Count the number of faults of each group device data in preset time range；The number of defects that future time equipment will occur is predicted by time series predicting model according to the number of faults.This method or system can predict the failure of grid equipment, make counter-measure in advance, avoid the generation of failure.
Description
Technical field
The present invention relates to grid equipment fault identification fields, more particularly to a kind of grid equipment fault recognition method and are
System.
Background technique
Power system device is very widely used in human society, and due to various reasons, power system device is frequent
Some system failures can occur and bring many troubles, also bring inconvenience to human lives.In this case, to power grid
The fault identification prediction of a equipment is just particularly important.Current techniques, only can be with there is no the technology for sufficiently using big data
After the failure occurs, failure cause may be obtained based on expert knowledge library deduction, it is indicated that fault point, and cannot accomplish pre- in advance
It is anti-.
Summary of the invention
The object of the present invention is to provide a kind of grid equipment fault recognition method and systems, to the failure to grid equipment
It is predicted, makes counter-measure in advance, avoid the generation of failure.
To achieve the above object, the present invention provides following schemes:
A kind of grid equipment fault recognition method, which comprises
Collect the major class device data of grid equipment；The major class device data includes operating status, equipment account, history
Failure, equipment deficiency, overhaul of the equipments, facility environment and equipment safety；
Each major class device data is classified as multiple group device datas by naive Bayesian machine learning model；
Count the number of faults of each group device data in preset time range；
The failure that future time equipment will occur is predicted by time series predicting model according to the number of faults
Number.
Optionally, it is described each major class device data is classified as by naive Bayesian machine learning model it is multiple small
Class device data, specifically includes:
Obtain sample data；The sample data includes each sample major class device data and corresponding sample group equipment
Data；
Pass through sample data training naive Bayesian machine learning model；
Each major class device data is classified as multiple groups by trained naive Bayesian machine learning model
Device data.
Optionally, described by sample data training naive Bayesian machine learning model, before further include:
The sample data is cleaned；
Sample data after cleaning is converted.
Optionally, described by sample data training naive Bayesian machine learning model, later further include:
The accuracy of naive Bayesian machine learning model after verifying training by the method for cross validation.
Optionally, described according to the number of faults, by time series predicting model, predict that future time equipment will
The number of defects of generation, specifically includes:
Count the number of faults of each sample group device data in preset time range；
Pass through the number of faults training time sequential forecasting models of each sample group device data in preset time range；
By trained time series predicting model, the number of defects that will occur in future time equipment is predicted.
The present invention also provides a kind of grid equipment fault finding system, the system comprises:
Data collection module, for collecting the major class device data of grid equipment；The major class device data includes operation
State, equipment account, historical failure, equipment deficiency, overhaul of the equipments, facility environment and equipment safety；
Categorization module, it is multiple for being classified as each major class device data by naive Bayesian machine learning model
Group device data；
Statistical module, for counting the number of faults of each group device data in preset time range；
Prediction module, for by time series predicting model, predicting that future time equipment will according to the number of faults
The number of defects to be occurred.
Optionally, the categorization module, specifically includes:
Sample data acquiring unit, for obtaining sample data；The sample data includes each sample major class device data
And corresponding sample group device data；
First training unit, for passing through sample data training naive Bayesian machine learning model；
Taxon, for being divided each major class device data by trained naive Bayesian machine learning model
Class is multiple group device datas.
Optionally, the categorization module, further includes:
Cleaning unit, for being cleaned to the sample data；
Converting unit, for being converted to the sample data after cleaning.
Optionally, the categorization module further include:
Authentication unit verifies the naive Bayesian machine learning model after training for the method by cross validation
Accuracy.
Optionally, the prediction module, specifically includes:
Statistic unit, for counting the number of faults of each sample group device data in preset time range；
Second training unit, for the number of faults training by each sample group device data in preset time range
Time series predicting model；
Predicting unit, for by trained time series predicting model, prediction will to occur in future time equipment
The number of defects.
Compared with prior art, the present invention has following technical effect that the present invention by big data machine learning method application
In grid equipment failure modes prediction, accomplishes the prediction that failure occurs first, make counter-measure in advance, avoid the hair of failure
It is raw, even failure has occurred, also can support staff the reason of positioning failure occurs in the shortest time as far as possible, failure occurs
Reason can carry out the Primary Location of failure, example by the Query mode of the higher failure modes of system failure early warning number ratio
Such as: if integral device broke down in the identical period in practical situations, the equipment fault of several subclassifications occur
Early warning can first go to inquire these exhaustive division equipment states, see whether failure really occur, so as to improve investigation efficiency.
It eliminates the hidden trouble, reduces the loss of failure bring to greatest extent.
Detailed description of the invention
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without any creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the flow chart of grid equipment of embodiment of the present invention fault recognition method；
Fig. 2 is the structural block diagram of grid equipment of embodiment of the present invention fault finding system.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The object of the present invention is to provide a kind of grid equipment fault recognition method and systems, to the failure to grid equipment
It is predicted, makes counter-measure in advance, avoid the generation of failure.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, with reference to the accompanying drawing and specific real
Applying mode, the present invention is described in further detail.
As shown in Figure 1, a kind of grid equipment fault recognition method includes:
Step 101: collecting the major class device data of grid equipment；The major class device data includes operating status, equipment
Account, historical failure, equipment deficiency, overhaul of the equipments, facility environment and equipment safety.
Step 102: each major class device data is classified as by multiple groups by naive Bayesian machine learning model
Device data.It is specific:
Obtain sample data；The sample data includes each sample major class device data and corresponding sample group equipment
Data；The sample data is cleaned；Sample data after cleaning is converted；
Pass through sample data training naive Bayesian machine learning model；
Each major class device data is classified as multiple groups by trained naive Bayesian machine learning model
Device data.
Step 103: the number of faults of each group device data of statistics in preset time range.
Step 104: according to the number of faults, by time series predicting model, predicting that future time equipment will be sent out
The raw number of defects.It is specific:
Count the number of faults of each sample group device data in preset time range；
Pass through the number of faults training time sequential forecasting models of each sample group device data in preset time range；
By trained time series predicting model, the number of defects that will occur in future time equipment is predicted.
Data cleansing is mainly to check the consistency and validity of data, to handle invalid value therein and incomplete value etc..
It is concentrated in notebook data, the important data item such as portions of log data title, system IP, failure key message, application message is sky
Value.Good result, and remaining data sample after deletion is not achieved because lacking important attribute data and will lead to mining mode
Amount can still meet model foundation, therefore such row containing null value is deleted.
Data transformation, which refers to, is changed into another form of expression from a kind of form of expression for data.By the way that data are carried out two-value
Change, discretization either continuous treatment, data value is treated as the form that data model is easier processing.
Model can be made to decrease the dimension of sample more effectively to reduce modeling by the selection to attributive character
Time & Space Complexity is effectively simplified learning model.Attributive character small with target signature correlation in data set is deleted,
Such as: system cpu occupied information, disk remaining space size information etc..In order to determine target signature, need to combine power equipment
The analysis and overall data distribution of industry live detection technical specification, expert to electrical equipment fault, to overall data into
Row correlation analysis obtains the target signature of basic feasible solution.
The process of naive Bayesian machine learning model training:
1) new training dataset is established using random sub- sampling and comprehensive oversampling technology, first with full doses such as infima species
Data start to be calculated.
2) discretization for carrying out attribute blurring and data to the sample set of class balance calculates, and constructs training sample set；
3) one initial network figure G of training sample set random configuration is utilized, prior probability and joint probability are calculated to node
And calculate the BIC1 score value of initial network figure G；
4) to G figure turn while or addition it is new while obtain new construction figure G ', to G ' calculating BIC2 score value.Compare BIC2 with
BIC1 size, then judge whether to delete side or carry out next calculating, it is maximum as optimal to eventually find BIC score value
Network G；
5) it is predicted using obtained maximum BIC scoring network structure G, and each class is calculated by confusion matrix
Predictablity rate and Average Accuracy.Terminate to calculate if having obtained promising result (generally 90%), otherwise, jump to
Step (1) re-starts data balancing sampling.
6) trained model is finally obtained, it, can be automatic according to its feature point by the grid equipment data of input system
It is divided into a classification, so that realizing a kind of pair of grid equipment carries out failure modes diagnosis prediction to Bayesian Classification Arithmetic mould
Type.Such as: from device history fault log information, we acquire 2560 datas, and are run according to algorithm, are divided
For the several main big classifications of hard disk failure, power environment system failure, line fault, switch fault, main transformer failure etc..
Here be for an example, be by operating status mentioned hereinbefore, equipment account, historical failure, equipment deficiency, equipment
The data of several aspects such as maintenance, facility environment, equipment safety select wherein device history fault data and carry out independent analysis, so
Device history failure is subjected to tiny category division afterwards.
This method further include: the naive Bayesian machine learning model after training is verified by the method for cross validation
Accuracy.Cross validation is the common method when model and verifying model parameter are established in machine learning.Cross validation, Gu Mingsi
Justice is exactly duplicate using data, obtained sample data is carried out cutting, group is combined into different training set and test set, uses
Training set carrys out training pattern, the quality predicted with test set come assessment models.The different instruction of available multiple groups on this basis
Practice collection and test set, certain sample in certain training set is likely to become the sample in test set in next time, i.e., so-called " intersection ".Root
Different according to the method for cutting, cross validation is divided into following three: the first is simple cross validation, so-called simple, be and its
His cross validation method is in contrast.Firstly, we it is random sample data is divided into two parts (such as: 70% training
Collection, 30% test set), model and parameter are then verified on test set come training pattern with training set.Then, Wo Menzai
Sample is upset, training set and test set are reselected, continues training data and testing model.Finally we select loss function
Assess optimal model and parameter.Second is S folding cross validation (S-Folder Cross Validation).And the first
Method is different, S folding cross validation can sample data it is random be divided into S part, it is each it is random select S-1 parts as training set,
Remaining 1 part is done test set.After the completion of this wheel, randomly chooses S-1 parts again and carry out training data.It is several to take turns and (be less than S) it
Afterwards, selection loss function assesses optimal model and parameter.The third is to stay a cross validation (Leave-one-out Cross
Validation), it is the special case of second situation, and S selects N-1 in this way for N number of sample equal to sample number N every time at this time
A sample carrys out training data, stays a sample to verify the quality of model prediction.It is considerably less that the method is mainly used for sample size
Situation, for example for common moderate problem, when N is less than 50, the present invention uses and stays a cross validation.The standard of model is improved with this
True property.
The training process of time series predicting model:
1) distributed computing of data counts merger
According to fault type sorting algorithm model, each daily record data is carried out to the division of main big type, then
It is marked as one data available of some type, carries out big data distributed storage.In some period granularity, by institute
After the data classification of processing in need is disposed, starting big data statistics calculation process presses the number of faults of a classification
According to the time cycle (such as: year, month, day, when etc.) carry out full dose data statistics, obtain in a period in several time cycles
The time series data of appearance.
2) it is based on Time Series AR IMA model algorithm prediction algorithm number of faults
According to ARIMA time series predicting model, input: certain one kind subdivision failure according to certain time granularity division
Time series data, such as: line fault is divided according to the time granularity of the moon, selectes 12 months historical time sequence weeks
Then issue uses ARIMA algorithm, predicts the fault data value for needing the predicted time period according to as input data, such as
Bimestrial line fault quantity under prediction.After data after above-mentioned statistics are normalized, running process predicts mould
Type process, several main other number of faults of major class are carried out subsequent quantity may time cycle sequence prediction.
3) equipment fault number is predicted based on Time Series AR IMA model algorithm
The number of faults of all kinds of fault types is obtained by Naive Bayes Classification method, then pre- by Time Series AR IMA model
Survey the number of faults in following a period of time.
Time Series AR IMA model prediction basic procedure:
Firstly, sample is standardized, such as seasonal difference (primary) and Z-score normalization；
Whether sample meets stationarity after test stoneization processing, prepares for ARIMA.That is: sample auto-correlation is calculated
And partial autocorrelation function, observation ending characteristic is (if can apparent cover half type truncation number really, so that it may jump to third step, directly
It connects to obtain ARIMA model)；
Model order number is carried out using BIC or AIC criterion.From ARMA (0,1), ARMA (1,0), ARMA (1,1), ARMA
(0,2), ARMA (2,0), ARMA (2,2), ARMA (1,2), ARMA (2,1) are middle to select the model of minimum AIC as final mask；
It is predicted using determining ARMA (p, q) model；
Predicted value carries out anti-standardization, restores initial value.
The specific embodiment provided according to the present invention, the invention discloses following technical effects: the present invention is by big data machine
Device learning method is applied in grid equipment failure modes prediction, accomplishes the prediction that failure occurs first, makes reply in advance and arrange
It applies, avoids the generation of failure, it, also can support staff's positioning failure generation in the shortest time as far as possible even failure has occurred
The reason of, fault occurrence reason can carry out event by the Query mode of the higher failure modes of system failure early warning number ratio
The Primary Location of barrier, such as: if integral device broke down in the identical period in practical situations, occur several small
The equipment fault early-warning of classification can first go to inquire these exhaustive division equipment states, see whether failure really occur, with this
Improve investigation efficiency.It eliminates the hidden trouble, reduces the loss of failure bring to greatest extent.
As shown in Fig. 2, the present invention also provides a kind of grid equipment fault finding system, the system comprises:
Data collection module 201, for collecting the major class device data of grid equipment；The major class device data includes fortune
Row state, equipment account, historical failure, equipment deficiency, overhaul of the equipments, facility environment and equipment safety.
Categorization module 202, for being classified as each major class device data by naive Bayesian machine learning model
Multiple group device datas.
The categorization module 202, specifically includes:
Sample data acquiring unit, for obtaining sample data；The sample data includes each sample major class device data
And corresponding sample group device data；
First training unit, for passing through sample data training naive Bayesian machine learning model；
Taxon, for being divided each major class device data by trained naive Bayesian machine learning model
Class is multiple group device datas；
Cleaning unit, for being cleaned to the sample data；
Converting unit, for being converted to the sample data after cleaning；
Authentication unit verifies the naive Bayesian machine learning model after training for the method by cross validation
Accuracy.
Statistical module 203, for counting the number of faults of each group device data in preset time range.
Prediction module 204, for by time series predicting model, predicting that future time is set according to the number of faults
The standby number of defects that will occur.
The prediction module 204, specifically includes:
Statistic unit, for counting the number of faults of each sample group device data in preset time range；
Second training unit, for the number of faults training by each sample group device data in preset time range
Time series predicting model；
Predicting unit, for by trained time series predicting model, prediction will to occur in future time equipment
The number of defects.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with other
The difference of embodiment, the same or similar parts in each embodiment may refer to each other.For system disclosed in embodiment
For, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is said referring to method part
It is bright.
Used herein a specific example illustrates the principle and implementation of the invention, and above embodiments are said
It is bright to be merely used to help understand method and its core concept of the invention；At the same time, for those skilled in the art, foundation
Thought of the invention, there will be changes in the specific implementation manner and application range.In conclusion the content of the present specification is not
It is interpreted as limitation of the present invention.
Claims (10)
1. a kind of grid equipment fault recognition method, which is characterized in that the described method includes:
Collect the major class device data of grid equipment；The major class device data includes operating status, equipment account, history event
Barrier, equipment deficiency, overhaul of the equipments, facility environment and equipment safety；
Each major class device data is classified as multiple group device datas by naive Bayesian machine learning model；
Count the number of faults of each group device data in preset time range；
The number of defects that future time equipment will occur is predicted by time series predicting model according to the number of faults.
2. grid equipment fault recognition method according to claim 1, which is characterized in that described to pass through naive Bayesian machine
Each major class device data is classified as multiple group device datas by device learning model, is specifically included:
Obtain sample data；The sample data includes each sample major class device data and corresponding sample group number of devices
According to；
Pass through sample data training naive Bayesian machine learning model；
Each major class device data is classified as multiple group equipment by trained naive Bayesian machine learning model
Data.
3. grid equipment fault recognition method according to claim 2, which is characterized in that described to pass through the sample data
Training naive Bayesian machine learning model, before further include:
The sample data is cleaned；
Sample data after cleaning is converted.
4. grid equipment fault recognition method according to claim 2, which is characterized in that described to pass through the sample data
Training naive Bayesian machine learning model, later further include:
The accuracy of naive Bayesian machine learning model after verifying training by the method for cross validation.
5. grid equipment fault recognition method according to claim 2, which is characterized in that described according to the number of faults
Amount is predicted the number of defects that future time equipment will occur, is specifically included by time series predicting model:
Count the number of faults of each sample group device data in preset time range；
Pass through the number of faults training time sequential forecasting models of each sample group device data in preset time range；
By trained time series predicting model, the number of defects that will occur in future time equipment is predicted.
6. a kind of grid equipment fault finding system, which is characterized in that the system comprises:
Data collection module, for collecting the major class device data of grid equipment；The major class device data include operating status,
Equipment account, historical failure, equipment deficiency, overhaul of the equipments, facility environment and equipment safety；
Categorization module, for each major class device data to be classified as multiple groups by naive Bayesian machine learning model
Device data；
Statistical module, for counting the number of faults of each group device data in preset time range；
Prediction module, for by time series predicting model, predicting that future time equipment will be sent out according to the number of faults
The raw number of defects.
7. grid equipment fault finding system according to claim 6, which is characterized in that the categorization module is specific to wrap
It includes:
Sample data acquiring unit, for obtaining sample data；The sample data include each sample major class device data and
Corresponding sample group device data；
First training unit, for passing through sample data training naive Bayesian machine learning model；
Taxon, for being classified as each major class device data by trained naive Bayesian machine learning model
Multiple group device datas.
8. grid equipment fault finding system according to claim 7, which is characterized in that the categorization module, further includes:
Cleaning unit, for being cleaned to the sample data；
Converting unit, for being converted to the sample data after cleaning.
9. grid equipment fault finding system according to claim 7, which is characterized in that the categorization module further include:
Authentication unit verifies the accurate of the naive Bayesian machine learning model after training for the method by cross validation
Property.
10. grid equipment fault finding system according to claim 7, which is characterized in that the prediction module is specific to wrap
It includes:
Statistic unit, for counting the number of faults of each sample group device data in preset time range；
Second training unit, for the number of faults training time by each sample group device data in preset time range
Sequential forecasting models；
Predicting unit, for predicting the event that will occur in future time equipment by trained time series predicting model
Hinder number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910089088.1A CN109800995A (en) | 2019-01-30 | 2019-01-30 | A kind of grid equipment fault recognition method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910089088.1A CN109800995A (en) | 2019-01-30 | 2019-01-30 | A kind of grid equipment fault recognition method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109800995A true CN109800995A (en) | 2019-05-24 |
Family
ID=66559162
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910089088.1A Pending CN109800995A (en) | 2019-01-30 | 2019-01-30 | A kind of grid equipment fault recognition method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109800995A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110349289A (en) * | 2019-06-21 | 2019-10-18 | 东软集团股份有限公司 | Failure prediction method, device, storage medium and electronic equipment |
| CN110413601A (en) * | 2019-07-04 | 2019-11-05 | 东南大学 | A kind of generating set Identification Data screening technique combined based on Gauss Naive Bayes Classifier and Predict error method |
| CN111325410A (en) * | 2020-03-13 | 2020-06-23 | 安图实验仪器(郑州)有限公司 | General fault early warning system based on sample distribution and early warning method thereof |
| CN113127804A (en) * | 2021-03-10 | 2021-07-16 | 广州亚美信息科技有限公司 | Method and device for determining number of vehicle faults, computer equipment and storage medium |
| CN113377595A (en) * | 2021-06-11 | 2021-09-10 | 上海壁仞智能科技有限公司 | Fault diagnosis method, fault diagnosis device, electronic equipment and storage medium |
| CN113807462A (en) * | 2021-09-28 | 2021-12-17 | 中电福富信息科技有限公司 | AI-based network equipment fault reason positioning method and system |
| US11451053B2 (en) * | 2019-06-13 | 2022-09-20 | Siemens Aktiengesellschaft | Method and arrangement for estimating a grid state of a power distribution grid |
| CN115460647A (en) * | 2022-10-21 | 2022-12-09 | 北京中电飞华通信有限公司 | Internet of things fault positioning method and system based on eSIM card and 5G base station |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012009804A1 (en) * | 2010-07-23 | 2012-01-26 | Corporation De L'ecole Polytechnique | Tool and method for fault detection of devices by condition based maintenance |
| CN103116531A (en) * | 2013-01-25 | 2013-05-22 | 浪潮(北京)电子信息产业有限公司 | Storage system failure predicting method and storage system failure predicting device |
| CN104504525A (en) * | 2014-12-26 | 2015-04-08 | 国家电网公司 | Method for realizing power-grid equipment failure prewarning through big data mining technology |
| CN104866926A (en) * | 2015-05-30 | 2015-08-26 | 国网上海市电力公司 | Fault amount predicting method of power distribution network based on meteorological factors and time sequence analysis |
-
2019
- 2019-01-30 CN CN201910089088.1A patent/CN109800995A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012009804A1 (en) * | 2010-07-23 | 2012-01-26 | Corporation De L'ecole Polytechnique | Tool and method for fault detection of devices by condition based maintenance |
| CN103116531A (en) * | 2013-01-25 | 2013-05-22 | 浪潮(北京)电子信息产业有限公司 | Storage system failure predicting method and storage system failure predicting device |
| CN104504525A (en) * | 2014-12-26 | 2015-04-08 | 国家电网公司 | Method for realizing power-grid equipment failure prewarning through big data mining technology |
| CN104866926A (en) * | 2015-05-30 | 2015-08-26 | 国网上海市电力公司 | Fault amount predicting method of power distribution network based on meteorological factors and time sequence analysis |
Non-Patent Citations (1)
| Title |
|---|
| 杨海滨: ""基于贝叶斯网络的电网设备故障诊断", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11451053B2 (en) * | 2019-06-13 | 2022-09-20 | Siemens Aktiengesellschaft | Method and arrangement for estimating a grid state of a power distribution grid |
| CN110349289A (en) * | 2019-06-21 | 2019-10-18 | 东软集团股份有限公司 | Failure prediction method, device, storage medium and electronic equipment |
| CN110413601A (en) * | 2019-07-04 | 2019-11-05 | 东南大学 | A kind of generating set Identification Data screening technique combined based on Gauss Naive Bayes Classifier and Predict error method |
| CN110413601B (en) * | 2019-07-04 | 2021-10-19 | 东南大学 | Generator data screening method based on Gauss naive Bayes and prediction error method |
| CN111325410A (en) * | 2020-03-13 | 2020-06-23 | 安图实验仪器(郑州)有限公司 | General fault early warning system based on sample distribution and early warning method thereof |
| CN111325410B (en) * | 2020-03-13 | 2023-10-10 | 安图实验仪器(郑州)有限公司 | Universal fault early warning system based on sample distribution and early warning method thereof |
| CN113127804A (en) * | 2021-03-10 | 2021-07-16 | 广州亚美信息科技有限公司 | Method and device for determining number of vehicle faults, computer equipment and storage medium |
| CN113377595A (en) * | 2021-06-11 | 2021-09-10 | 上海壁仞智能科技有限公司 | Fault diagnosis method, fault diagnosis device, electronic equipment and storage medium |
| CN113807462A (en) * | 2021-09-28 | 2021-12-17 | 中电福富信息科技有限公司 | AI-based network equipment fault reason positioning method and system |
| CN113807462B (en) * | 2021-09-28 | 2023-07-04 | 中电福富信息科技有限公司 | Network equipment fault cause positioning method and system based on AI |
| CN115460647A (en) * | 2022-10-21 | 2022-12-09 | 北京中电飞华通信有限公司 | Internet of things fault positioning method and system based on eSIM card and 5G base station |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109800995A (en) | A kind of grid equipment fault recognition method and system | |
| CN106951984B (en) | Dynamic analysis and prediction method and device for system health degree | |
| JP4846923B2 (en) | How to predict the timing of future service events for a product | |
| Zio | Reliability engineering: Old problems and new challenges | |
| Strachan et al. | Providing decision support for the condition-based maintenance of circuit breakers through data mining of trip coil current signatures | |
| CN110162445A (en) | The host health assessment method and device of Intrusion Detection based on host log and performance indicator | |
| JP2013538543A (en) | Machine learning for power grids | |
| CN115221982B (en) | Traction power supply operation and maintenance method and device, terminal and storage medium | |
| CN116862081B (en) | Operation and maintenance method and system for pollution treatment equipment | |
| CN110992204A (en) | Emergency diesel generator set reliability evaluation method and system based on big data technology | |
| CN113239087A (en) | Anti-electricity-stealing inspection monitoring method and system | |
| CN116720324A (en) | Traction substation key equipment fault early warning method and system based on prediction model | |
| CN115204536A (en) | Building equipment fault prediction method, device, equipment and storage medium | |
| CN110361609A (en) | Extra-high voltage equipment monitors system and method | |
| CN114997741A (en) | Maintenance plan risk assessment method, system and equipment | |
| CN117436846B (en) | Equipment predictive maintenance method and system based on neural network | |
| Weiss | Predicting telecommunication equipment failures from sequences of network alarms | |
| CN113835947B (en) | Method and system for determining abnormality cause based on abnormality recognition result | |
| Gu et al. | Investigation on the quality assurance procedure and evaluation methodology of machine learning building energy model systems | |
| CN115936663A (en) | Maintenance method and device for power system | |
| EP4033421A1 (en) | Method and system for predicting a failure of a monitored entity | |
| CN116108376A (en) | Monitoring system and method for preventing electricity stealing, electronic equipment and medium | |
| CN115864644A (en) | Relay protection device state evaluation method, system, equipment and medium | |
| Scheinert et al. | The shape of watershed governance: Locating the boundaries of multiplex networks | |
| Wang et al. | LSTM-based alarm prediction in the mobile communication network |
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: 20190524 |