CN107037278A - A kind of substandard intelligent substation method for diagnosing faults of IEC61850 - Google Patents
A kind of substandard intelligent substation method for diagnosing faults of IEC61850 Download PDFInfo
- Publication number
- CN107037278A CN107037278A CN201610963005.3A CN201610963005A CN107037278A CN 107037278 A CN107037278 A CN 107037278A CN 201610963005 A CN201610963005 A CN 201610963005A CN 107037278 A CN107037278 A CN 107037278A
- Authority
- CN
- China
- Prior art keywords
- intelligent substation
- fault diagnosis
- information
- hidden
- diagnosis result
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
Abstract
The invention discloses a kind of substandard intelligent substation method for diagnosing faults of IEC61850, implementation steps include obtaining the static information of intelligent substation by the functional configuration model of intelligent substation, pass through the multidate information of the Receive message intelligent substation of intelligent substation, by the static information of acquisition, in the fault diagnosis model based on deep learning network that multidate information input is trained, export the fault diagnosis result of intelligent substation, detect fault diagnosis result, if fault diagnosis result breaks down for intelligent substation, failure wave-recording is then carried out according to fault type classification, and output control information repairs the malfunction of intelligent substation.Learning ability of the present invention is strong, the fault detect degree of accuracy is high, noiseproof feature is good, detection speed is fast, and be particularly suitable for use in nonlinear thermal gradient, is had an enormous advantage for pattern-recognition and classification.
Description
Technical field
The present invention relates to the fault diagnosis of intelligent substation and assessment technology, and in particular to a kind of IEC61850 is substandard
Intelligent substation method for diagnosing faults.
Background technology
The scale of modern power systems is increasing, and the voltage class more and more higher of transmission line of electricity, length is increasingly longer, defeated
Environment residing for electric line is also more complicated, therefore the possibility and number of times that break down also inevitably roll up.Therefore,
The research of power failure detection algorithm has very important theory significance and wide application prospect.In addition, in order to adapt to numeral
Change transformer station and future substation automated system demand for development, IEC61850 communication standards realized in fault diagnosis system,
Interoperability that can be in very big lifting transformer station between intelligent electronic device.
After intelligent substation, revolutionary variation occurs compared with conventional substation for its electrical secondary system structure and form,
The communication network bearing function logical signal of physics is shown as, conventional secondary circuit is changed into the connection between communication network, signal
It is changed into virtual terminal and virtual circuit.No longer exist one by one between network physical topology and function information and the input and output of signal
Corresponding relation, traditional fault diagnosis and appraisal procedure are fault detect and analysis based on secondary electric loop mostly, it is impossible to
Applied to intelligent substation, cause to be difficult to carry out to the business such as intelligent substation fault detect and analysis.Meanwhile, existing power transformation
Station failure diagnostic method merely with the running status of primary equipment in transformer station in itself and warning information or utilizes open circuit mostly
Device acts situation and Trouble Report to carry out fault diagnosis, provides the probability of equipment fault.Secondary system of intelligent substation structure
Change and its management and the automation of maintenance service and intelligent requirements are improved, make traditional fault diagnosis and appraisal procedure
The demand of intelligent substationization operation can not have been met in diagnosis depth and diagnostic method.Current fault diagnosis algorithm is a lot,
But there are still two problems：Although 1) common algorithm calculates simple, its poor robustness is easily affected by noise；2) intelligence
The bad adaptability of energy algorithm, lacks self-learning capability.In order to solve these problems, it is necessary to design a strong adaptability, robust
The high method of property.
The content of the invention
The technical problem to be solved in the present invention：For the above mentioned problem of prior art, there is provided a kind of learning ability is strong, failure
Accuracy in detection is high, noiseproof feature is good, detection speed is fast, and be particularly suitable for use in nonlinear thermal gradient, for pattern-recognition and divides
The substandard intelligent substation method for diagnosing faults of IEC61850 that class has an enormous advantage.
In order to solve the above-mentioned technical problem, the technical solution adopted by the present invention is：
A kind of substandard intelligent substation method for diagnosing faults of IEC61850, implementation steps include：
1) static information of intelligent substation, the static information bag are obtained by the functional configuration model of intelligent substation
The physics and logic association information, communication network for including intelligent substation main electrical scheme topology, primary equipment and secondary device are believed substantially
Breath；
2) by the multidate information of the Receive message intelligent substation of intelligent substation, the multidate information includes once setting
Standby operation information, the position of breaker and disconnecting switch, action message, the network traffic information of protection and control device；
3) fault diagnosis model based on deep learning network for training the static information of acquisition, multidate information input
In, export training package in the fault diagnosis result of intelligent substation, the fault diagnosis model based on deep learning network and contain
Classification map relation between the fault diagnosis result of static information, multidate information and intelligent substation；
4) fault diagnosis result is detected, if fault diagnosis result breaks down for intelligent substation, according to failure classes
Type classification carries out failure wave-recording, and output control information repairs the malfunction of intelligent substation.
Preferably, step 3) in the fault diagnosis model based on deep learning network be limited Boltzmann machine model, institute
State limited Boltzmann machine model to be made up of visual layers v and hidden layer h, visual layers v includes m visual layer unit v1~vm, hidden layer h
Include n Hidden unit h1~hn, visual layers v bias vector a { a1~an, the bias vector b { b of Hidden unit1~bm, can
Depending on layer v and hidden layer h with not connected between layer unit node, the weight W on the side between visual layer unit and Hidden unit, and pin
Limited Boltzmann machine model is used sdpecific dispersion method is trained.
The substandard intelligent substation method for diagnosing faults tools of IEC61850 of the present invention have the advantage that：
1st, the present invention is strong using deep learning network characterization learning ability, and the fault detect degree of accuracy is high, and noiseproof feature is good etc.
Advantage, designs the substandard intelligent substation method for diagnosing faults of IEC61850, by the intelligence of collection on the basis of the model
Substation information is as deep learning network inputs, so as to conveniently and efficiently complete fault diagnosis, with learning ability is strong, failure
Accuracy in detection is high, the advantage that noiseproof feature is good, detection speed is fast.
2nd, the present invention has the machine learning model of many hidden layers and the training number of magnanimity based on deep learning network struction
According to learn more useful feature, so that the finally accuracy of lifting classification or prediction, easily obtains globally optimal solution, it is adaptable to
Nonlinear thermal gradient, has an enormous advantage for pattern-recognition and classification.
Brief description of the drawings
Fig. 1 is the basic procedure schematic diagram of present invention method.
Fig. 2 is the limited Boltzmann machine model schematic of present invention method.
Fig. 3 is the deep learning network iterativecurve of present invention method.
Fig. 4 is the effect comparison schematic diagram of present invention method.
Embodiment
As shown in figure 1, the implementation steps bag of the substandard intelligent substation method for diagnosing faults of the present embodiment IEC61850
Include：
1) static information of intelligent substation is obtained by the functional configuration model of intelligent substation, static information includes intelligence
Physics and logic association information, the communication network essential information of energy Substation Bus Arrangement topology, primary equipment and secondary device；
2) by the multidate information of the Receive message intelligent substation of intelligent substation, multidate information includes primary equipment
Operation information, the position of breaker and disconnecting switch, action message, the network traffic information of protection and control device；
3) fault diagnosis model based on deep learning network for training the static information of acquisition, multidate information input
In, the fault diagnosis result of intelligent substation is exported, training contains quiet in the fault diagnosis model based on deep learning network
Classification map relation between the fault diagnosis result of state information, multidate information and intelligent substation；
4) fault diagnosis result is detected, if fault diagnosis result breaks down for intelligent substation, according to failure classes
Type classification carries out failure wave-recording, and output control information repairs the malfunction of intelligent substation.
The present embodiment step 3) in the fault diagnosis model based on deep learning network be limited Boltzmann machine model, by
Limit Boltzmann machine model is made up of visual layers v and hidden layer h, and visual layers v includes m visual layer unit v1~vm, hidden layer h include n
Individual Hidden unit h1~hn, visual layers v bias vector a { a1~an, the bias vector b { b of Hidden unit1~bm, visual layers v
And hidden layer h between layer unit node with not connecting, the weight W on the side between visual layer unit and Hidden unit, and for by
Limit Boltzmann machine model is used to be trained to sdpecific dispersion method.As shown in Fig. 2 the limited Boltzmann machine model of the present embodiment
By visual layers v (v1~vm) and hidden layer h (h1~hn) composition, visual layers v1~vmBetween do not connect, hidden layer h1~hnBetween do not connect
Connect, the bias vector a (a of visual layer unit1~an), the bias vector b (b of Hidden unit1~bm), visual layers vm, hidden layer hnIt
Between side weight wnm。
For limited Boltzmann machine model, when inputting v, hidden layer h is obtained according to p (h | v)；Obtain hidden layer h it
Afterwards, visual layers can be obtained by p (v | h) again.Limited Boltzmann machine model (RBM) is a kind of based on energy model, is limited
Boltzmann machine model can apparent variable v and hidden variable h joint configuration energy such as formula (1) shown in；
In formula (1), E (v, h) represent can apparent variable v and hidden variable h joint configuration energy, aiRepresent i-th can
Apparent variable viOffset parameter, viRepresent i-th can apparent variable, bjRepresent j-th of hidden variable hjOffset parameter, hjRepresent the
J hidden variable, wijRepresent i-th can apparent variable vi, j-th of hidden variable hjBetween side weight.
Generally, the parameter for being limited Boltzmann machine model is represented by θ={ W, a, b }, and wherein W is visible element and hidden
The weight on the side between unit, a and b are respectively the bias vector of visible element and hidden unit.From visual layers v and hidden layer h
The energy of joint configuration can obtain shown in v and h joint probability such as formula (2)；
In formula (2), Pθ(v, h) represents v and h joint probability, and Z (θ) represents the regularization coefficient of parameter θ, E (v, h；θ)
Represent parameter θ under can apparent variable v and hidden variable h joint configuration energy.Joined by the limited Boltzmann machine model of training
Number, can make limited Boltzmann machine model obtain optimal performance.
Parallel tempering sampling is a kind of very efficient way to limited Boltzmann machine model training.In training process
In, each temperature one gibbs chain of correspondence is simultaneously sampled using the method being tempered parallel.Every gibbs chain one difference of correspondence
Temperature ti, tiMeet 1=t1<……<ti<……<tM-1<tM, whether handed over according to certain conditional decision between different temperatures chain
Change sampled value.According to formula (2), at different temperature, parallel limited Boltzmann machine model joint probability such as formula (3) institute of tempering
Show；
In formula (3), Pr(v, h) represents model joint probability, Z (ti) represent parameter tiRegularization coefficient, tiRepresent temperature
Degree, E (v, h；θ) represent parameter θ under can apparent variable v and hidden variable h joint configuration energy, M represent gibbs chain
Number.
The parallel tempering Monte Carlo EGS4 method of limited Boltzmann machine model includes two stages：
1st, Metropolis-Hastings sample phases：The next of Current Temperatures is calculated according to existing sampled value to adopt
Sampling point, shown in basic sampling calculation formula such as formula (4)；
In formula (4), xi+1Represent i+1 sampled point, xiRepresent ith sample point, Metropolis-Hastings tables
Show sampling function,Represent that average is that 0, variance isNormal distyribution function, tkRepresent temperature.
2nd, switching phase：After sampling is completed, two adjacent temperature (t in temperature collection are calculatedrAnd tr-1) under it is aobvious hidden
Node layer (vr,hr) and (vr-1,hr-1) whether meet the condition of exchange, the exchange bar of the parallel limited Boltzmann machine model of tempering
Shown in part such as formula (5)；
In formula (5), trAnd tr-1Represent two adjacent temperature in temperature collection, E (vr,hr) represent can apparent variable vrWith it is hidden
Hide variable hrJoint configuration energy, E (vr-1,hr-1) represent can apparent variable vr-1With hidden variable hr-1Joint configuration energy
Amount.If meeting condition shown in formula (5), just the sampled point under adjacent temperature chain is exchanged, otherwise not exchanged.By repeatedly following
Ring sampling, exchange, at last most t1Sampled value at a temperature of=1 is used to be limited Boltzmann machine model pre-training model parameter θ, adopts
The target sample value obtained with parallel tempering can make limited Boltzmann machine model training obtain preferable application effect.By inciting somebody to action
The parameter θ of the limited Boltzmann machine model of formula (1)RBMAobvious node layer in={ W, a, b } and the connection weight between the node of hidden layer
Value W is multiplied by temperature ss, and the parameter of whole model is changed into θRBM-PT={ β W, a, b }, does not change for biasing weights a and b.Now,
Parallel tempering algorithm can be organically combined with limited Boltzmann machine, improve training effectiveness.
In order to improve deep learning Network Recognition effect, first to the substandard intelligent substations of the present embodiment IEC61850
The fault diagnosis model (limited Boltzmann machine model) based on deep learning network of method for diagnosing faults is trained.Herein
Randomly select training sample 300 and test sample 100.Bring the training sample randomly selected into limited Boltzmann machine mould
Type is trained, and maximum frequency of training is 5000 times, and training objective error is 10-5.Fig. 3 is deep learning network iterativecurve,
From the figure 3, it may be seen that after 612 times are trained, the mean square error of limited Boltzmann machine model converges to error expected requirement.Take survey
Sample this limited Boltzmann machine model trained is verified, obtain Fault Identification model.
In order to verify the robustness of the substandard intelligent substation method for diagnosing faults of the present embodiment IEC61850, by this reality
The substandard intelligent substation method for diagnosing faults of an IEC61850 (being referred to as proposition method in figure) is applied to be respectively compared in addition
The fault detect performance of two kinds of algorithms (wavelet coefficient method, neural network).5 kinds of signal to noise ratio of each measuring and calculation, the letter of signal
Make an uproar than arriving 10dB for -10dB.Fig. 4 is fault detection algorithm effect comparison schematic diagram, as can be seen from Figure 4, with the reduction of signal to noise ratio,
The fault detect rate of the substandard intelligent substation method for diagnosing faults of the present embodiment IEC61850 is constantly reduced.From algorithm angle
Compare, the fault detect rate highest of the substandard intelligent substation method for diagnosing faults of the present embodiment IEC61850, neutral net
Algorithm takes second place, and wavelet coefficient method verification and measurement ratio is minimum.Contrast failure recall rate, the substandard intelligent power transformations of the present embodiment IEC61850
The fault detect rate of station failure diagnostic method is higher than neural network algorithm by 17.8%, higher than wavelet coefficient method by 10.4%.
In summary, the substandard intelligent substation method for diagnosing faults of the present embodiment IEC61850 builds IEC61850 marks
The fault diagnosis model (limited Boltzmann machine model) based on deep learning network under accurate, is set by gathering intelligent substation
The static state such as standby voltage, electric current, frequency and dynamic parameter, the failure based on deep learning network is inputted by static and dynamic parameter
In diagnostic model (limited Boltzmann machine model), fault type is exported；According to fault type, output control information repairs power transformation
Stand state, and carry out failure wave-recording of classifying.The substandard intelligent substation method for diagnosing faults of the present embodiment IEC61850 has
Diagnosis process is clear, diagnostic method self-perfection, the accurate reliable advantage of diagnostic result, so as to improve electric network security.
Described above is only the preferred embodiment of the present invention, and protection scope of the present invention is not limited merely to above-mentioned implementation
Example, all technical schemes belonged under thinking of the present invention belong to protection scope of the present invention.It should be pointed out that for the art
Those of ordinary skill for, some improvements and modifications without departing from the principles of the present invention, these improvements and modifications
It should be regarded as protection scope of the present invention.
Claims (2)
1. a kind of substandard intelligent substation method for diagnosing faults of IEC61850, it is characterised in that implementation steps include：
1) static information of intelligent substation is obtained by the functional configuration model of intelligent substation, the static information includes intelligence
Physics and logic association information, the communication network essential information of energy Substation Bus Arrangement topology, primary equipment and secondary device；
2) by the multidate information of the Receive message intelligent substation of intelligent substation, the multidate information includes primary equipment
Operation information, the position of breaker and disconnecting switch, action message, the network traffic information of protection and control device；
3) in the fault diagnosis model based on deep learning network for training the static information of acquisition, multidate information input,
Training contains quiet in the fault diagnosis result of output intelligent substation, the fault diagnosis model based on deep learning network
Classification map relation between the fault diagnosis result of state information, multidate information and intelligent substation；
4) fault diagnosis result is detected, if fault diagnosis result breaks down for intelligent substation, according to fault type point
Class carries out failure wave-recording, and output control information repairs the malfunction of intelligent substation.
2. the substandard intelligent substation method for diagnosing faults of IEC61850 according to claim 1, it is characterised in that step
It is rapid 3) in the fault diagnosis model based on deep learning network be limited Boltzmann machine model, the limited Boltzmann machine mould
Type is made up of visual layers v and hidden layer h, and visual layers v includes m visual layer unit v1~vm, hidden layer h includes n Hidden unit h1~
hn, visual layers v bias vector a { a1~an, the bias vector b { b of Hidden unit1~bm, visual layers v and hidden layer h are with layer list
Do not connected between first node, the weight W on the side between visual layer unit and Hidden unit, and for limited Boltzmann machine mould
Type uses and sdpecific dispersion method is trained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610963005.3A CN107037278A (en) | 2016-11-04 | 2016-11-04 | A kind of substandard intelligent substation method for diagnosing faults of IEC61850 |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610963005.3A CN107037278A (en) | 2016-11-04 | 2016-11-04 | A kind of substandard intelligent substation method for diagnosing faults of IEC61850 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107037278A true CN107037278A (en) | 2017-08-11 |
Family
ID=59530325
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610963005.3A Pending CN107037278A (en) | 2016-11-04 | 2016-11-04 | A kind of substandard intelligent substation method for diagnosing faults of IEC61850 |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107037278A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108520472A (en) * | 2018-02-28 | 2018-09-11 | 北京邮电大学 | A kind of method, apparatus and electronic equipment of processing electric power system data |
| CN109672175A (en) * | 2018-12-30 | 2019-04-23 | 国网北京市电力公司 | Power grid control method and device |
| CN109902373A (en) * | 2019-02-21 | 2019-06-18 | 国网山东省电力公司临沂供电公司 | A kind of area under one's jurisdiction Fault Diagnosis for Substation, localization method and system |
| WO2020007372A1 (en) * | 2018-07-06 | 2020-01-09 | 东莞市李群自动化技术有限公司 | Control method and device employing industrial ethernet |
| CN112510699A (en) * | 2020-11-25 | 2021-03-16 | 国网湖北省电力有限公司咸宁供电公司 | Transformer substation secondary equipment state analysis method and device based on big data |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102570599A (en) * | 2010-12-30 | 2012-07-11 | 郑州平高自动化有限公司 | Communication method of low-voltage-grade transformer substation |
| CN103901298A (en) * | 2014-03-13 | 2014-07-02 | 广东电网公司电力科学研究院 | Method and system for detecting operating states of substation equipment |
| CN104794534A (en) * | 2015-04-16 | 2015-07-22 | 国网山东省电力公司临沂供电公司 | Power grid security situation predicting method based on improved deep learning model |
| US20160004964A1 (en) * | 2014-07-07 | 2016-01-07 | Gwangju Institute Of Science And Technology | Neuromorphic system and method for operating the same |
-
2016
- 2016-11-04 CN CN201610963005.3A patent/CN107037278A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102570599A (en) * | 2010-12-30 | 2012-07-11 | 郑州平高自动化有限公司 | Communication method of low-voltage-grade transformer substation |
| CN103901298A (en) * | 2014-03-13 | 2014-07-02 | 广东电网公司电力科学研究院 | Method and system for detecting operating states of substation equipment |
| US20160004964A1 (en) * | 2014-07-07 | 2016-01-07 | Gwangju Institute Of Science And Technology | Neuromorphic system and method for operating the same |
| CN104794534A (en) * | 2015-04-16 | 2015-07-22 | 国网山东省电力公司临沂供电公司 | Power grid security situation predicting method based on improved deep learning model |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108520472A (en) * | 2018-02-28 | 2018-09-11 | 北京邮电大学 | A kind of method, apparatus and electronic equipment of processing electric power system data |
| WO2020007372A1 (en) * | 2018-07-06 | 2020-01-09 | 东莞市李群自动化技术有限公司 | Control method and device employing industrial ethernet |
| CN109672175A (en) * | 2018-12-30 | 2019-04-23 | 国网北京市电力公司 | Power grid control method and device |
| CN109672175B (en) * | 2018-12-30 | 2023-07-14 | 国网北京市电力公司 | Power grid control method and device |
| CN109902373A (en) * | 2019-02-21 | 2019-06-18 | 国网山东省电力公司临沂供电公司 | A kind of area under one's jurisdiction Fault Diagnosis for Substation, localization method and system |
| CN112510699A (en) * | 2020-11-25 | 2021-03-16 | 国网湖北省电力有限公司咸宁供电公司 | Transformer substation secondary equipment state analysis method and device based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107037278A (en) | A kind of substandard intelligent substation method for diagnosing faults of IEC61850 | |
| CN106777984B (en) | A method of photovoltaic array Working state analysis and fault diagnosis are realized based on density clustering algorithm | |
| US20160124031A1 (en) | Smart multi-purpose monitoring system using wavelet design and machine learning for smart grid applications | |
| Hassani et al. | Unsupervised concrete feature selection based on mutual information for diagnosing faults and cyber-attacks in power systems | |
| Gao et al. | Fault line detection using waveform fusion and one-dimensional convolutional neural network in resonant grounding distribution systems | |
| CN111999591B (en) | Method for identifying abnormal state of primary equipment of power distribution network | |
| Poudel et al. | Circuit topology estimation in an adaptive protection system | |
| CN113610119B (en) | Method for identifying power transmission line development faults based on convolutional neural network | |
| CN107462810A (en) | A kind of fault section location method suitable for active power distribution network | |
| CN106019043A (en) | Grid fault diagnosis method based on fuzzy integral multi-source information fusion theory | |
| Bhuiyan et al. | A deep learning through DBN enabled transmission line fault transient classification framework for multimachine microgrid systems | |
| Alabbawi et al. | Faults detection, location, and classification of the elements in the power system using intelligent algorithm | |
| Afrasiabi et al. | Machine Learning-Based Condition Monitoring of Solar Photovoltaic Systems: A Review | |
| Chang et al. | Fault location identification in power transmission networks: Using novel nonintrusive fault-monitoring systems | |
| Kanwal et al. | Artificial Intelligence based Faults Identification, Classification, and Localization Techniques in Transmission Lines-A Review | |
| Wang et al. | Free Probability Theory Based Event Detection for Power Grids using IoT-Enabled Measurements | |
| Zhong et al. | Fault Diagnosis of PV Array Based on Time Series and Support Vector Machine | |
| He et al. | Application and Comparative Analysis of Traditional Machine Learning and Deep Learning in Transmission Line Fault Classification | |
| Wu et al. | A deep recognition network of capacitor voltage transformer based on dilated convolution and Bi-LSTM | |
| Zhou et al. | Detecting and classifying multiple class events in a power distribution system based on an eigenvalue fluctuation model | |
| Gong et al. | State detection method of secondary equipment in smart substation based on deep belief network and trend prediction | |
| Intriago et al. | Real-time power system event detection: A novel instance selection approach | |
| Shan et al. | Power system fault diagnosis based on logistic regression deep neural network | |
| Ojetola et al. | Time Series Classification for Detecting Fault Location in a DC Microgrid | |
| Zhang et al. | Line Fault Detection of DC Distribution Networks Using the Artificial Neural 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: 20170811 |