CN118091003A - Transformer oil chromatographic anomaly identification method based on random matrix - Google Patents
Transformer oil chromatographic anomaly identification method based on random matrix Download PDFInfo
- Publication number
- CN118091003A CN118091003A CN202410030269.8A CN202410030269A CN118091003A CN 118091003 A CN118091003 A CN 118091003A CN 202410030269 A CN202410030269 A CN 202410030269A CN 118091003 A CN118091003 A CN 118091003A
- Authority
- CN
- China
- Prior art keywords
- data
- random matrix
- transformer oil
- chromatographic
- oil chromatographic
- 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
- 238000000034 method Methods 0.000 title claims abstract description 40
- 239000011159 matrix material Substances 0.000 title claims abstract description 37
- 238000004587 chromatography analysis Methods 0.000 claims abstract description 26
- 230000002159 abnormal effect Effects 0.000 claims abstract description 8
- 238000011208 chromatographic data Methods 0.000 claims abstract description 8
- 230000009467 reduction Effects 0.000 claims abstract description 8
- 238000007405 data analysis Methods 0.000 claims abstract description 7
- 238000013528 artificial neural network Methods 0.000 claims description 13
- 238000012549 training Methods 0.000 claims description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 6
- 238000001228 spectrum Methods 0.000 claims description 4
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 claims description 3
- 230000002950 deficient Effects 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 8
- 238000001514 detection method Methods 0.000 abstract description 3
- 238000000605 extraction Methods 0.000 abstract description 3
- 230000000717 retained effect Effects 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 10
- 238000012360 testing method Methods 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 7
- 238000009826 distribution Methods 0.000 description 5
- 238000003745 diagnosis Methods 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 2
- 239000007789 gas Substances 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000008485 antagonism Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000001747 exhibiting effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
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
- Housings And Mounting Of Transformers (AREA)
Abstract
The invention relates to the technical field of transformers, in particular to a transformer oil chromatographic anomaly identification method based on a random matrix, which comprises the following steps: s1, reducing transformer oil chromatographic sample data; s2, simulating a negative sample of the oil chromatography data based on the generated model; s3, constructing an oil chromatography anomaly identification method by using a random matrix big data analysis tool. The invention adopts the rough intensive reduction oil chromatography data dimension, and compared with the method which directly takes all oil chromatography data as input data, the reduced data quantity is less and the dimension is lower; by adopting WGAN-GP to generate oil chromatography negative sample data, the number of the positive and negative samples can be balanced, and the detection effect on the transformer oil chromatography abnormal sample is improved; and the high-dimensional characteristics of the oil chromatographic data are obtained by utilizing the random matrix theory, and compared with the traditional characteristic extraction mode, the random matrix can retain the space-time correlation of the high-dimensional data, so that most of effective information of the data is retained, and the accuracy of anomaly identification is improved.
Description
Technical Field
The invention relates to the technical field of transformers, in particular to a transformer oil chromatographic anomaly identification method based on a random matrix.
Background
The power transformer is one of the most important electrical devices in the power system, and its operation state directly affects the safe and stable operation of the whole power system. The latent faults of the transformer are found early, and the safe operation of the transformer is guaranteed, so that the power supply reliability is improved, and the method is an important problem of general attention of a power system. Analysis of dissolved gases in oil is known worldwide as one of the most effective methods for monitoring and diagnosing early latent faults in oil-filled power transformers, as no power outage tests are required. Therefore, the transformer fault diagnosis technology is researched, the deep learning and the random matrix theory are combined, the oil chromatographic data is efficiently utilized to sense and monitor the state of the transformer substation equipment, the problem to be solved by the current further research is solved, the operation, maintenance and overhaul level of the transformer can be improved, and the transformer fault diagnosis method has important practical significance.
Accordingly, those skilled in the art have been working to develop a method for identifying chromatographic anomalies in transformer oil to improve the level of fault identification in transformers.
Disclosure of Invention
In view of the above-mentioned drawbacks of the prior art, the technical problem to be solved by the present invention is how to use oil chromatographic data to sense and monitor the state of substation equipment so as to improve the operation, maintenance and overhaul level of the transformer.
In order to solve the technical problems, the invention provides a transformer oil chromatographic anomaly identification method based on a random matrix, which comprises the following steps:
s1, reducing transformer oil chromatographic sample data;
S2, simulating a negative sample of the oil chromatography data based on the generated model;
S3, constructing an oil chromatography anomaly identification method by using a random matrix big data analysis tool.
In a preferred embodiment of the present invention, in step S1, the reduction of the transformer oil chromatographic sample data includes: and selecting transformer substation oil chromatograph sampling data, establishing a decision table by utilizing the sampling data, performing attribute reduction on the decision table based on a neighborhood rough set algorithm to obtain a final minimum fault feature set and related importance, and taking the first p features of key feature index sequencing as key features.
Preferably, the p is 80% of the final minimum fault feature lumped feature number.
In another preferred embodiment of the present invention, in step S2, the generating model-based simulated oil chromatography data negative sample includes: after the key features are obtained, sample generation is carried out based on WGAN-GP, a neural network structure of the discriminator and the generator is determined, the optimization solver selects Adam, and the learning rate and training times of the generator and the discriminator are set.
In still another preferred embodiment of the present invention, in step S3, the constructing a method for identifying an anomaly of oil chromatography using a random matrix big data analysis tool includes: and comprehensively analyzing the original partial discharge statistics by using a random matrix method, wherein the average spectrum radius MSR of the defective equipment and part of normal equipment is determined, and when the MSR exceeds a preset value in a normal threshold range, the equipment is abnormal.
Preferably, the generator learning rate is set to 0.001 and the training times are 2000.
Preferably, the learning rate of the discriminator is set to 0.001, and the training times are 2000 times.
Preferably, the discriminator adopts a 5-layer neural network structure.
Preferably, the generator adopts a 6-layer neural network structure.
Preferably, the predetermined value is 20% of the normal threshold range.
The invention has the beneficial effects that:
1. The rough set reduction oil chromatography data dimension is adopted, and compared with the method that all oil chromatography data are directly used as input data, the reduced data quantity is less, and the dimension is lower;
2. The WGAN-GP is adopted to generate oil chromatography negative sample data, so that the number of positive and negative samples can be balanced, the preference of an identification algorithm on a plurality of types of samples can be eliminated through WGAN-GP balanced data, and the detection effect on the transformer oil chromatography abnormal samples is improved;
3. and the high-dimensional characteristics of the oil chromatographic data are obtained by utilizing the random matrix theory, and compared with the traditional characteristic extraction mode, the random matrix can retain the space-time correlation of the high-dimensional data, so that most of effective information of the data is retained, and the accuracy of anomaly identification is improved.
The conception, specific structure, and technical effects of the present invention will be further described with reference to the accompanying drawings to fully understand the objects, features, and effects of the present invention.
Drawings
FIG. 1 is a flow chart of a method for identifying chromatographic anomalies in transformer oil based on a random matrix according to an embodiment of the present invention;
FIG. 2 is a WGAN-GP schematic diagram in an embodiment of the invention.
Detailed Description
The following description of the preferred embodiments of the present invention refers to the accompanying drawings, which make the technical contents thereof more clear and easy to understand. The present invention may be embodied in many different forms of embodiments and the scope of the present invention is not limited to only the embodiments described herein.
Referring to fig. 1 to 2, in one implementation of the present invention, a method for identifying chromatographic anomalies of transformer oil based on a random matrix is provided, including the following steps:
s1, reducing transformer oil chromatographic sample data;
S2, simulating a negative sample of the oil chromatography data based on the generated model;
S3, constructing an oil chromatography anomaly identification method by using a random matrix big data analysis tool.
In a preferred embodiment of the present invention, in step S1, the reduction of the transformer oil chromatographic sample data includes: and selecting oil chromatogram sampling data of a certain transformer substation, establishing a decision table by utilizing the sampling data, performing attribute reduction on the decision table based on a neighborhood rough set algorithm to obtain a final minimum fault feature set and related importance, and taking the first p features of key feature index sequencing as key features. Preferably, the p is 80% of the final minimum fault feature lumped feature number.
In transformer oil chromatographic anomaly identification, if the dimension of input data for generating a data model is too high, more time and resource space are spent, and the embodiment of the invention adopts the rough set to reduce the dimension of the oil chromatographic data, and compared with the method of directly taking all the oil chromatographic data as the input data, the reduced data quantity is smaller and the dimension is lower. The rough set theory (Rough Theory) was originally proposed by the z.pawlak teaching of poland in the 80 s of the 20 th century, and is a mathematical method for studying incomplete data, the expression of uncertain knowledge, learning and induction. The method provides an effective tool for researching analysis and reasoning of inaccurate data and mining the relation among the data and finding potential knowledge. Not all eigenvalues are necessary for fault classification, some eigenvalues are redundant for fault classification, and removal of these eigenvalues does not affect the classification effect of the fault. In fault diagnosis, the rough set theory is mostly combined with the fuzzy neural network, and the respective characteristics of the rough set theory and the fuzzy neural network are fully utilized. In order to apply the rough set theory in the field of transformer anomaly identification, the oil chromatographic dataset needs to be described as a form of decision system D s = < U, a=c U D, V, f.
Where u= { X 1,X2,…Xn } is the set of all samples X n (g=1, 2, …, N) in the transformer fault dataset; a is a feature set of a sample, a= { a 1,a2,…am}∪{d};C＝{a1,a2,…am } is a conditional feature set, and d= { D } is a decision feature set; v is the type of transformer fault representing the sample, and is the label attribute of the sample, namely v= a∈AVa,Va represents the value range of all samples under the characteristic a; f is an information function used to specify the value of the sample under the corresponding feature attribute, and the value of the sample X g under the feature a is recorded as f (X, a) ∈V a.
For the followingThe relationship of U under B is expressed as:/>And U/R B＝{[x]B |x ε U, where x B＝{y∈U|(x,y)∈RB. Based on the relation R, for/>The upper and lower approximation sets of X with respect to R are expressed as: /(I)
The rough set theory can effectively reduce sample data, remove redundant attributes and simplify rule sets, so that the fuzzy neural network established based on the simplified rules has a better topological structure on the premise of ensuring classification capacity, and the learning speed is greatly improved, the judgment is accurate, the fault tolerance capacity is strong, and the fuzzy neural network has higher practical value. Fault diagnosis based on rough set theory has the following characteristics: (1) Compared with other fuzzy set theory and probability statistical methods, the rough set analysis method only utilizes information provided by the data itself, and does not need any priori knowledge; (2) The rough set analysis method can reduce the data on the premise of keeping the key information, and obtain the minimum expression of knowledge; (3) The rough set theory has stronger fault tolerance capability, and provides a more effective analysis method for a system with incomplete information and inaccurate data.
By adopting the rough set reduction oil chromatography data dimension, the reduced data quantity is less, the dimension is lower, and the calculation pressure of a subsequent anomaly identification algorithm is reduced.
In another embodiment of the present invention, in step S2, the generating model-based simulated oil chromatography data negative sample includes: after the key features are obtained, generating an antagonism network (WASSERSTEIN GENERATIVE ADVERSARIAL network WITH GRADIENT PENALTY, WGAN-GP) based on Wasserstein distance and gradient penalty carries out sample generation, determining a neural network structure of a discriminator and a generator, and setting the learning rate and training times of the generator and the discriminator by using an adaptive moment estimation (adaptive moment estimation, adam) as an optimization solver.
In a preferred embodiment, the generator and arbiter learning rate is set to 0.001 and the number of exercises is 2000.
The discriminator adopts a 5-layer neural network structure, and the generator adopts a 6-layer neural network structure.
The generation of the countermeasure network (WASSERSTEIN GENERATIVE ADVERSARIAL network, WGAN) based on the waserstein distance is mainly improved from the point of view of a loss function, the improvement can obtain good performance results even in a fully connected layer, the loss function cross entropy (JS divergence) of the GAN is not suitable for measuring the distance between distributions with disjoint parts, the WASSERTEIN distance is used for measuring the distance between the data distribution and the real data distribution, the problem of unstable training is solved theoretically, and the training of a balance generator and a discriminator is not needed. Meanwhile, the problem of mode collapse is solved, so that the generation results of the generator are more various. However WGAN directly uses WEIGHT CLIPPING to cut the weight to a certain range of-0.01, so that too simple and rough a storm can lead to weakening of modeling capability of a model, gradient disappearance or explosion. WGAN-GP is an improved version of WGAN, mainly introducing a gradient penalty GRADIENT PENALTY, improving Lipschitz continuity constraints. The design logic of the gradient penalty (GRADIENT PENALTY, GP) term is: a microtranspirable function satisfies the 1-Lipschitz condition if and only if its gradient norms do not exceed 1 anywhere.
According to the embodiment of the invention, by adopting WGAN-GP to generate the oil chromatography negative sample data, the number of positive and negative samples can be balanced, compared with the method of directly adopting the original data to identify, the data balanced by WGAN-GP can eliminate the preference of an identification algorithm on a plurality of types of samples, and the detection effect on the transformer oil chromatography abnormal samples is improved. The method solves the problem that the model is easy to generate preference to most samples (normal state samples) due to less abnormal data.
In still another embodiment of the present invention, in step S3, the constructing the oil chromatography anomaly identification method using a random matrix big data analysis tool includes: and comprehensively analyzing the original partial discharge statistics by using a random matrix method, wherein the average spectrum radius MSR of the defective equipment and part of normal equipment is determined, and when the MSR exceeds a preset value in a normal threshold range, the equipment is abnormal. Preferably said predetermined value is 20% of said normal threshold range.
The linear eigenvalue statistic can be further than the single loop law, exhibiting the statistical properties of the random matrix from a variety of angles. The linear eigenvalue statistic of the random matrix X is defined as:
Wherein: lambda i (i=1, 2,., n) is the eigenvalue of the random matrix X, The representation is a test function. Selecting different test functions/>Such as the MSR values used in embodiments of the present invention/>Will result in differentValues and different test effects. Different test functions may be selected to achieve significant statistical effects depending on the particular system.
The invention mainly selects the average spectrum radius-MSR value as the linear characteristic value statistic. As shown in the formula:
Wherein: (i=1, 2, … N) is/> Characteristic value of/>For/>Distribution radius on complex plane.
The MSR value is related to the single loop law, and is the average value of the radius values of all eigenvalue points in the single loop law map. When the system is not in the event of a failure,The eigenvalues of (2) are approximately distributed between the two rings, and when the system is in an event, part of the eigenvalues enter the inner ring. /(I)The eigenvalue radius distribution of (c) decreases as events occur, deviating from normal values, so that the MSR value κ MSR can be used as a high-dimensional parameter to detect whether there is an event in the system. The minimum limit of the MSR can be calculated from historical data for a particular system, i.e., below the threshold, an abnormal event occurs in the system.
How to select the appropriate test function values is a key step in the application of random matrix theory analysis. The selection of the test function is required according to the historical data condition of the system, and then the real-time state of the system is researched by the selected test function.
According to the embodiment of the invention, the high-dimensional characteristics of the oil chromatographic data are obtained by utilizing the random matrix theory, so that most of effective information of the data can be reserved, and the accuracy of anomaly identification is improved. In addition, the oil chromatography anomaly identification method based on the random matrix can analyze and process the oil chromatography data under the condition of no priori knowledge and fault labels so as to sense the state of transformer equipment, reduce the problem of oil chromatography data burrs caused by instantaneous gas generation in the anomaly process, improve the anomaly identification robustness and convenience, and solve the problem of space-time correlation of the characteristics which are difficult to retain in the traditional characteristic extraction mode.
The foregoing describes in detail preferred embodiments of the present invention. It should be understood that numerous modifications and variations can be made in accordance with the concepts of the invention without requiring creative effort by one of ordinary skill in the art. Therefore, all technical solutions which can be obtained by logic analysis, reasoning or limited experiments based on the prior art by the person skilled in the art according to the inventive concept shall be within the scope of protection defined by the claims.
Claims (10)
1. A transformer oil chromatographic anomaly identification method based on a random matrix is characterized by comprising the following steps:
s1, reducing transformer oil chromatographic sample data;
S2, simulating a negative sample of the oil chromatography data based on the generated model;
S3, constructing an oil chromatography anomaly identification method by using a random matrix big data analysis tool.
2. The method for identifying transformer oil chromatographic anomalies based on a random matrix according to claim 1, wherein in step S1, the reducing transformer oil chromatographic sample data comprises: and selecting transformer substation oil chromatograph sampling data, establishing a decision table by utilizing the sampling data, performing attribute reduction on the decision table based on a neighborhood rough set algorithm to obtain a final minimum fault feature set and related importance, and taking the first p features of key feature index sequencing as key features.
3. The method for identifying transformer oil chromatographic anomalies based on a random matrix according to claim 2, wherein p is 80% of the final minimum fault feature lumped feature number.
4. The method for identifying transformer oil chromatographic anomalies based on a random matrix according to claim 2, wherein in step S2, the generating model-based simulated oil chromatographic data negative sample comprises: after the key features are obtained, sample generation is carried out based on WGAN-GP, a neural network structure of the discriminator and the generator is determined, the optimization solver selects Adam, and the learning rate and training times of the generator and the discriminator are set.
5. The method for identifying transformer oil chromatographic anomalies based on random matrix according to claim 4, wherein in step S3, the constructing the method for identifying transformer oil chromatographic anomalies using a random matrix big data analysis tool comprises: and comprehensively analyzing the original partial discharge statistics by using a random matrix method, wherein the average spectrum radius MSR of the defective equipment and part of normal equipment is determined, and when the MSR exceeds a preset value in a normal threshold range, the equipment is abnormal.
6. The method for identifying transformer oil chromatographic anomalies based on a random matrix according to claim 4, wherein the generator learning rate is set to 0.001 and the training times are 2000.
7. The method for identifying transformer oil chromatographic anomalies based on a random matrix according to claim 6, wherein the learning rate of the discriminator is set to 0.001 and the training times are 2000.
8. The method for identifying transformer oil chromatographic anomalies based on a random matrix according to claim 7, wherein the discriminator adopts a 5-layer neural network structure.
9. The method for identifying the chromatographic anomalies of the transformer oil based on the random matrix according to claim 8, wherein the generator adopts a 6-layer neural network structure.
10. The method of claim 5, wherein the predetermined value is 20% of the normal threshold range.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410030269.8A CN118091003A (en) | 2024-01-09 | 2024-01-09 | Transformer oil chromatographic anomaly identification method based on random matrix |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410030269.8A CN118091003A (en) | 2024-01-09 | 2024-01-09 | Transformer oil chromatographic anomaly identification method based on random matrix |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118091003A true CN118091003A (en) | 2024-05-28 |
Family
ID=91144711
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410030269.8A Pending CN118091003A (en) | 2024-01-09 | 2024-01-09 | Transformer oil chromatographic anomaly identification method based on random matrix |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118091003A (en) |
-
2024
- 2024-01-09 CN CN202410030269.8A patent/CN118091003A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111768082A (en) | Power equipment state evaluation method based on big data analysis | |
| Tong et al. | Detection and classification of transmission line transient faults based on graph convolutional neural network | |
| CN108304567B (en) | Method and system for identifying working condition mode and classifying data of high-voltage transformer | |
| CN114386537B (en) | Lithium battery fault diagnosis method and device based on Catboost and electronic equipment | |
| CN110297207A (en) | Method for diagnosing faults, system and the electronic device of intelligent electric meter | |
| CN114091549A (en) | Equipment fault diagnosis method based on deep residual error network | |
| CN110348683A (en) | The main genetic analysis method, apparatus equipment of electrical energy power quality disturbance event and storage medium | |
| CN113985156A (en) | Intelligent fault identification method based on transformer voiceprint big data | |
| CN115345392A (en) | Abnormal situation studying and judging and key index screening method for power distribution network | |
| CN111737993B (en) | Method for extracting equipment health state from fault defect text of power distribution network equipment | |
| CN111487563B (en) | Transformer state knowledge acquisition method and device based on genetic algorithm and attribute support degree | |
| CN116070140B (en) | Power distribution substation safe operation state monitoring system and method | |
| CN117560300A (en) | Intelligent internet of things flow prediction and optimization system | |
| CN112949201A (en) | Wind speed prediction method and device, electronic equipment and storage medium | |
| CN110807014B (en) | Cross validation based station data anomaly discrimination method and device | |
| CN117031201A (en) | Multi-scene topology anomaly identification method and system for power distribution network | |
| CN115729761B (en) | Hard disk fault prediction method, system, equipment and medium | |
| CN116522138A (en) | Multi-element time sequence anomaly detection method and system based on weak supervision integration | |
| CN118091003A (en) | Transformer oil chromatographic anomaly identification method based on random matrix | |
| Qi et al. | Rapid transformer health state recognition through canopy cluster-merging of dissolved gas data in high-dimensional space | |
| Alawode et al. | Fuzzy logic and decision tree models for dissolved gas analysis in power transformer fault diagnosis | |
| CN113886592A (en) | Quality detection method for operation and maintenance data of power information communication system | |
| Jingyu et al. | Statistical analysis of distribution network fault information based on multi-source heterogeneous data mining | |
| CN113191556A (en) | Nuclear power Loca event fault prediction and diagnosis method | |
| CN111723851A (en) | Production line fault detection method |
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 |