CN112085108B - Photovoltaic power station fault diagnosis algorithm based on automatic encoder and K-means clustering - Google Patents
Photovoltaic power station fault diagnosis algorithm based on automatic encoder and K-means clustering Download PDFInfo
- Publication number
- CN112085108B CN112085108B CN202010952632.3A CN202010952632A CN112085108B CN 112085108 B CN112085108 B CN 112085108B CN 202010952632 A CN202010952632 A CN 202010952632A CN 112085108 B CN112085108 B CN 112085108B
- Authority
- CN
- China
- Prior art keywords
- data
- photovoltaic power
- fault diagnosis
- matrix
- diagnosis algorithm
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Probability & Statistics with Applications (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Photovoltaic Devices (AREA)
Abstract
The invention relates to the technical field of photovoltaic power generation, in particular to a photovoltaic power station fault diagnosis algorithm based on an automatic encoder and K-means clustering, which comprises the steps of S1, screening real-time production data of a photovoltaic module, wherein M production data with output power larger than 1w in the production data are screened, and output power items are deleted to obtain a matrix of M7; s2, carrying out normalization processing on the matrix obtained in the step S1 to obtain a normalized matrix; s3, inputting the normalized matrix elements into the trained AE neural network model to obtain the dimension reduction representation of the data; s4, taking 2-dimensional data output by a third hidden layer of the AE neural network model, and classifying by using the trained k-means to obtain a model prediction state. According to the invention, the real-time production data of the photovoltaic module is subjected to dimension reduction by using AE, and a scatter diagram is drawn for the dimension reduced data, so that the dimension reduced data is used as the real-time state of the photovoltaic module. The diagnosis accuracy is high; the diagnosis result is easy to understand; the dimension reducing effect is better.
Description
Technical Field
The invention relates to the technical field of photovoltaic power generation, in particular to a photovoltaic power station fault diagnosis algorithm based on an automatic encoder and K-means clustering.
Background
Photovoltaic power generation is used as a renewable energy source, and has been rapidly developed in China in recent years due to the advantages of good regional adaptability, short construction period and the like. The photovoltaic module is unqualified in the process of manufacturing or is influenced by the external severe meteorological environment conditions, so that the faults such as battery aging, breakage, hot spot shielding and the like are easy to occur, the power generation of a photovoltaic power station is reduced, and the service life of the module is shortened. Therefore, the fault diagnosis of the photovoltaic module has practical significance in safe and efficient operation of the photovoltaic power station.
The maintenance mode that photovoltaic power plant adopted generally adopts artifical inspection, because photovoltaic module generally distributes extensively, and the position is mostly high or the place that the topography is complicated, consequently manual inspection often wastes time and energy, and real-time is relatively poor. Intelligent detection of photovoltaic modules is a hotspot in industry research. At present, most of fault diagnosis of a photovoltaic module is performed by collecting or simulating data in a test environment to generate enough fault data sample training models, and real fault samples of a photovoltaic power station are often fewer, so that the prediction accuracy is low.
Disclosure of Invention
The invention aims to provide a photovoltaic power station fault diagnosis algorithm based on an automatic encoder and K-means clustering so as to solve the problems in the background technology.
In order to achieve the above purpose, the present invention provides the following technical solutions: the photovoltaic power station fault diagnosis algorithm based on the automatic encoder and the K-means clustering comprises the following steps:
s1, screening real-time production data of a photovoltaic module, screening M production data with output power larger than 1w in the production data, deleting output power items, and obtaining a matrix with M being 7;
s2, carrying out normalization processing on the matrix obtained in the step S1 to obtain a normalized matrix;
s3, inputting the normalized matrix elements into the trained AE neural network model, encoding and decoding the data, and obtaining the dimension reduction representation of the data by using a layer with the minimum number of neurons in a hidden layer of the AE neural network model;
s4, taking 2-dimensional data output by a third hidden layer of the AE neural network model, and classifying the 2-dimensional data by using trained k-means to obtain the category, namely the model prediction state.
Preferably, the normalization calculation formula in the step S2 is as follows:
wherein: i, j are the indices of the rows and columns, respectively, in the original data matrix, x ij Matrix element, v, representing normalized i, j positions ij Matrix elements representing the i, j positions before normalization,for normalizing the smallest matrix element in the first j columns,/->To normalize the largest matrix element in the first j columns.
Preferably, the AE neural network model in the step S3 is composed of an input layer, a hidden layer, and an output layer, where the input layer is used for inputting data, the hidden layer is used for encoding and decoding the input data, that is, compressing and recovering information, and the output layer is used for outputting the result of the model.
Parameters to be set in the AE neural network model are as follows:
the number of neurons of the input layer is 7, and the number of columns after normalization matrix is corresponding;
number of neurons of the hidden layers first through fifth layers: 5,3,2,3,5 respectively;
the number of neurons of the output layer is 7, which is equal to the number of neurons of the input layer.
Preferably, the training manner of the AE neural network model in the step S3 is as follows:
preferably, the photovoltaic module production data of the previous year is subjected to the screening of the step S1 and the normalization of the step S2, each row is used as one piece of data, every 1024 pieces of data are used as 1 batch, namely, batch-size is 1024, the number of training wheels epoch is set to 200, and AE is trained according to the setting.
Preferably, the normalized matrix is input one by one, the trained AE is input, the 2-dimensional data output by the third hidden layer is taken, M pieces of 2-dimensional data are obtained, and a scatter diagram is drawn for the M pieces of 2-dimensional data and used for observing the distribution of the data.
Preferably, the obtained M pieces of 2-dimensional data are clustered by using k-means, fault records of the previous year are searched, and the categories of the data corresponding to different fault moments are used as corresponding fault states.
Preferably, production data of one month are selected for testing, trained k-means are used for classifying, the classification, namely the model prediction state, is obtained, the model prediction state is compared with the real state corresponding to the data, the accuracy of model prediction for normal states and various fault states is calculated, and if the prediction accuracy is lower than 90%, k-means adjustment parameters are retrained.
Preferably, the photovoltaic module production data of 1 month in the current year is subjected to screening in the step S1 and normalization in the step S2, N faults occur in 1 month in the current year, 10 pieces of normalized matrix data in a corresponding time period are taken for each fault, 10 pieces of data at normal moment are randomly taken, total test data (N+1) are input into the trained AE, 2-dimensional data output by a third hidden layer are taken, N+1) 10 pieces of 2-dimensional data are obtained, and then the 2-dimensional data are classified by using the trained k-means.
Compared with the prior art, the invention has the beneficial effects that: according to the invention, the real-time production data of the photovoltaic module is subjected to dimension reduction by using AE, a scatter diagram is drawn for the dimension reduced data, and the scatter diagram is observed to obtain the types of the data which can be divided, so that the real-time state of the photovoltaic module is taken as the real-time state of the photovoltaic module.
The following advantages can be achieved by the diagnosis of the invention:
1. the diagnosis accuracy is high. The working state of the photovoltaic module is judged based on a data driving mode, and faults are accurately diagnosed after the model is online.
2. The diagnostic result is easy to understand. The real-time production data of the photovoltaic module is subjected to dimension reduction, and a scatter diagram is drawn, so that the category of the data can be intuitively seen.
3. The dimension reducing effect is better. Compared with the common PCA dimension reduction, AE can learn more complex relation, and the dimension reduction result can reflect the real distribution of data.
Drawings
FIG. 1 is an AE model structure of the present invention;
FIG. 2 is a graph showing the change of the loss function during the AE model training process of the present invention;
fig. 3 is a diagram of training results of 2-dimensional data.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1-3, the present invention provides a technical solution: the photovoltaic power station fault diagnosis algorithm based on the automatic encoder and the K-means clustering comprises the following steps:
s1, screening real-time production data of a photovoltaic module, screening M production data with output power larger than 1w in the production data, deleting output power items, and obtaining a matrix with M being 7;
s2, carrying out normalization processing on the matrix obtained in the step S1 to obtain a normalized matrix; the normalized calculation formula is:
wherein: i, j are the indices of the rows and columns, respectively, in the original data matrix, x ij Matrix element, v, representing normalized i, j positions ij Matrix elements representing the i, j positions before normalization,to normalize the smallest matrix element in the first j columns,to normalize the largest matrix element in the first j columns.
S3, inputting the normalized matrix elements into the trained AE neural network model, encoding and decoding the data, and obtaining the dimension reduction representation of the data by using a layer with the minimum number of neurons in a hidden layer of the AE neural network model;
the AE neural network model consists of an input layer, a hidden layer and an output layer, wherein the input layer is used for inputting data, the hidden layer is used for encoding and decoding the input data, namely, compressing and recovering information, and the output layer is used for outputting a result of the model.
Parameters to be set in the AE neural network model are as follows:
the number of neurons of the input layer is 7, and the number of columns after normalization matrix is corresponding;
number of neurons of the hidden layers first through fifth layers: 5,3,2,3,5 respectively;
the number of neurons of the output layer is 7, which is equal to the number of neurons of the input layer.
The AE neural network model is trained as follows:
the photovoltaic module production data of the previous year (namely 2019, data of other years can be taken) is subjected to screening in the step S1 and normalization in the step S2, each row is used as one piece of data, each 1024 pieces of data are used as 1 batch, namely batch-size is 1024, the number of training wheels epoch is set to 200, and AE is trained according to the setting.
And inputting the normalized matrix one by one, namely inputting the trained AE, taking the 2-dimensional data output by the third hidden layer, obtaining M pieces of 2-dimensional data, and drawing a scatter diagram for the M pieces of 2-dimensional data for observing the distribution of the data.
The obtained M pieces of 2-dimensional data are clustered by using k-means, fault records of the previous year (namely 2019, data of other years can be taken) are searched, and the categories of the data corresponding to different fault moments are used as corresponding fault states.
In addition, production data testing of one month (data of 1 month in 2020 and data of other months except 2019 are selected), classification is carried out by using trained k-means to obtain the category, namely the model prediction state, the model prediction state is compared with the real state corresponding to the data, the accuracy rate of model prediction on the normal state and various fault states is calculated, and if the prediction accuracy rate is lower than 90%, k-means adjustment parameters are retrained.
And (3) screening the photovoltaic module production data of month 1 in 2020, carrying out normalization processing of step S1 and step S2, wherein N faults occur in month 1 in 2020, each fault takes 10 pieces of normalized matrix data in a corresponding time period, randomly takes 10 pieces of data at normal time, inputs 10 pieces of common test data (N+1) into a trained AE, takes 2-dimensional data output by a third hidden layer, obtains 10 pieces of 2-dimensional data (N+1), and classifies the 2-dimensional data by using trained k-means.
S4, taking 2-dimensional data output by a third hidden layer of the AE neural network model, and classifying the 2-dimensional data by using trained k-means to obtain the category, namely the model prediction state.
Examples:
and (3) training an AE-KMeas algorithm by using production data from 1 month in 2019 to 1 month in 2020, wherein 13:12 minutes are started on 15 days in 2 months in 2020, the model continuously judges the component with the number of N1267 as a shielding state, a maintainer confirms on site, a plastic bag is found on the surface of the component, and the model is used for judging the failure of the component normally.
The term learning according to the invention: in a machine learning task, a certain evaluation criterion (accuracy) is accumulated with continuous experience, and the evaluation index in the task is improved.
The foregoing is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art, who is within the scope of the present invention, should make equivalent substitutions or modifications according to the technical scheme of the present invention and the inventive concept thereof, and should be covered by the scope of the present invention.
Claims (9)
1. The photovoltaic power station fault diagnosis algorithm based on the automatic encoder and the K-means clustering is characterized in that: the method comprises the following steps:
s1, screening real-time production data of a photovoltaic module, screening M production data with output power larger than 1w in the production data, deleting output power items, and obtaining a matrix with M being 7;
s2, carrying out normalization processing on the matrix obtained in the step S1 to obtain a normalized matrix;
s3, inputting the normalized matrix elements into the trained AE neural network model, encoding and decoding the data, and obtaining the dimension reduction representation of the data by using a layer with the minimum number of neurons in a hidden layer of the AE neural network model, wherein the method specifically comprises the following steps: 2-dimensional data output by a third hidden layer of the AE neural network model is taken;
s4, classifying the 2-dimensional data by using the trained k-means to obtain the class of the 2-dimensional data as a fault state predicted by the model.
2. The photovoltaic power plant fault diagnosis algorithm based on the automatic encoder and the K-means clustering as claimed in claim 1, wherein the photovoltaic power plant fault diagnosis algorithm is characterized in that: the normalized calculation formula in the step S2 is:
wherein: i, j are the indices of the rows and columns, respectively, in the original data matrix, x ij Matrix element, v, representing normalized i, j positions ij Matrix elements representing the i, j positions before normalization,to normalize the smallest matrix element in the first j columns,to normalize the largest matrix element in the first j columns.
3. The photovoltaic power plant fault diagnosis algorithm based on the automatic encoder and the K-means clustering as claimed in claim 1, wherein the photovoltaic power plant fault diagnosis algorithm is characterized in that: the AE neural network model in the step S3 is composed of an input layer, a hidden layer and an output layer, wherein the input layer is used for inputting data, the hidden layer is used for encoding and decoding the input data, namely, compressing and recovering information, and the output layer is used for outputting a result of the model.
4. The photovoltaic power plant fault diagnosis algorithm based on the automatic encoder and the K-means clustering as claimed in claim 3, wherein: parameters to be set in the AE neural network model are as follows:
the number of neurons of the input layer is 7, and the number of columns after normalization matrix is corresponding;
number of neurons of the hidden layers first through fifth layers: 5,3,2,3,5 respectively;
the number of neurons of the output layer is 7, which is equal to the number of neurons of the input layer.
5. The photovoltaic power plant fault diagnosis algorithm based on the automatic encoder and the K-means clustering as claimed in claim 1, wherein the photovoltaic power plant fault diagnosis algorithm is characterized in that: the training mode of the AE neural network model in the step S3 is as follows:
and (3) carrying out screening in the step (S1) and normalization in the step (S2) on the photovoltaic module production data in the previous year, wherein each row is used as one piece of data, each 1024 pieces of data is used as 1 batch, namely, the batch-size is 1024, the number of training wheels epoch is set to 200, and the AE is trained according to the setting.
6. The photovoltaic power plant fault diagnosis algorithm based on the automatic encoder and the K-means clustering as claimed in claim 5, wherein the photovoltaic power plant fault diagnosis algorithm is characterized in that: and inputting the normalized matrix one by one, namely inputting the trained AE, taking the 2-dimensional data output by the third hidden layer, obtaining M pieces of 2-dimensional data, and drawing a scatter diagram for the M pieces of 2-dimensional data for observing the distribution of the data.
7. The photovoltaic power plant fault diagnosis algorithm based on the automatic encoder and the K-means clustering as claimed in claim 6, wherein the photovoltaic power plant fault diagnosis algorithm is characterized in that: and clustering the obtained M2-dimensional data by using k-means, searching fault records of the previous year, and taking the category of the data corresponding to different fault moments as a corresponding fault state.
8. The photovoltaic power plant fault diagnosis algorithm based on the automatic encoder and the K-means clustering as claimed in claim 7, wherein: in addition, a month of production data test is selected, trained k-means is used for classifying, the classification, namely the model prediction state, is obtained, the model prediction state is compared with the real state corresponding to the data, the accuracy rate of model prediction for normal states and various fault states is calculated, and if the prediction accuracy rate is lower than 90%, k-means adjustment parameters are retrained.
9. The photovoltaic power plant fault diagnosis algorithm based on the automatic encoder and the K-means clustering as claimed in claim 8, wherein: the photovoltaic module production data of the current year 1 month is subjected to screening in the step S1 and normalization in the step S2, N faults occur in the current year 1 month, 10 pieces of normalization matrix data in a corresponding time period are taken for each fault, 10 pieces of data at normal time are randomly taken, 10 pieces of common test data (N+1) are input into the trained AE, 2-dimensional data output by a third hidden layer are taken, 10 pieces of 2-dimensional data (N+1) are obtained, and the 2-dimensional data are classified by using the trained k-means.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010952632.3A CN112085108B (en) | 2020-09-11 | 2020-09-11 | Photovoltaic power station fault diagnosis algorithm based on automatic encoder and K-means clustering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010952632.3A CN112085108B (en) | 2020-09-11 | 2020-09-11 | Photovoltaic power station fault diagnosis algorithm based on automatic encoder and K-means clustering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112085108A CN112085108A (en) | 2020-12-15 |
| CN112085108B true CN112085108B (en) | 2023-08-18 |
Family
ID=73736631
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010952632.3A Active CN112085108B (en) | 2020-09-11 | 2020-09-11 | Photovoltaic power station fault diagnosis algorithm based on automatic encoder and K-means clustering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112085108B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113688910A (en) * | 2021-08-26 | 2021-11-23 | 国网江苏省电力有限公司常州供电分公司 | Power grid operation multi-source information fusion method based on k-BP algorithm |
| CN115169544A (en) * | 2022-09-05 | 2022-10-11 | 广东工业大学 | Short-term photovoltaic power generation power prediction method and system |
| CN115601197B (en) * | 2022-11-28 | 2023-03-10 | 深圳市峰和数智科技有限公司 | Abnormal state detection method and device for photovoltaic power station |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106021806A (en) * | 2016-06-06 | 2016-10-12 | 福州大学 | Photovoltaic string fault diagnosis method based on kernel function limit learning machine |
| CN109117865A (en) * | 2018-07-13 | 2019-01-01 | 福州大学 | One kind realizing photovoltaic array Working state analysis and method for diagnosing faults based on principal component analysis |
| CN109670553A (en) * | 2018-12-25 | 2019-04-23 | 福州大学 | Diagnosing failure of photovoltaic array method based on Adaptive Neuro-fuzzy Inference |
| CN109766952A (en) * | 2019-01-21 | 2019-05-17 | 福州大学 | Photovoltaic array fault detection method based on Partial Least Squares and extreme learning machine |
| CN110068776A (en) * | 2019-05-16 | 2019-07-30 | 合肥工业大学 | Three-level inverter open-circuit fault diagnostic method based on Support Vector Machines Optimized |
| KR102148761B1 (en) * | 2019-11-20 | 2020-08-27 | 주식회사 케이디티 | Apparatus diagnosing fault of photovoltaic system |
-
2020
- 2020-09-11 CN CN202010952632.3A patent/CN112085108B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106021806A (en) * | 2016-06-06 | 2016-10-12 | 福州大学 | Photovoltaic string fault diagnosis method based on kernel function limit learning machine |
| CN109117865A (en) * | 2018-07-13 | 2019-01-01 | 福州大学 | One kind realizing photovoltaic array Working state analysis and method for diagnosing faults based on principal component analysis |
| CN109670553A (en) * | 2018-12-25 | 2019-04-23 | 福州大学 | Diagnosing failure of photovoltaic array method based on Adaptive Neuro-fuzzy Inference |
| CN109766952A (en) * | 2019-01-21 | 2019-05-17 | 福州大学 | Photovoltaic array fault detection method based on Partial Least Squares and extreme learning machine |
| CN110068776A (en) * | 2019-05-16 | 2019-07-30 | 合肥工业大学 | Three-level inverter open-circuit fault diagnostic method based on Support Vector Machines Optimized |
| KR102148761B1 (en) * | 2019-11-20 | 2020-08-27 | 주식회사 케이디티 | Apparatus diagnosing fault of photovoltaic system |
Non-Patent Citations (1)
| Title |
|---|
| 基于BP神经网络的光伏阵列故障诊断研究;王元章等;电力系统保护与控制(16);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112085108A (en) | 2020-12-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112085108B (en) | Photovoltaic power station fault diagnosis algorithm based on automatic encoder and K-means clustering | |
| WO2019214268A1 (en) | Photovoltaic array fault diagnosis method based on composite information | |
| CN111539553B (en) | Wind turbine generator fault early warning method based on SVR algorithm and off-peak degree | |
| CN111507376B (en) | Single-index anomaly detection method based on fusion of multiple non-supervision methods | |
| CN110458230A (en) | A kind of distribution transforming based on the fusion of more criterions is with adopting data exception discriminating method | |
| CN107657250B (en) | Bearing fault detection and positioning method and detection and positioning model implementation system and method | |
| CN109660206B (en) | Wasserstein GAN-based photovoltaic array fault diagnosis method | |
| CN106712713A (en) | Monitoring system and monitoring-anomaly location method for photovoltaic power stations | |
| CN113887616A (en) | Real-time abnormity detection system and method for EPG (electronic program guide) connection number | |
| CN112085621B (en) | Distributed photovoltaic power station fault early warning algorithm based on K-Means-HMM model | |
| CN110879377A (en) | Metering device fault tracing method based on deep belief network | |
| CN111753891A (en) | Rolling bearing fault diagnosis method based on unsupervised feature learning | |
| CN115809405A (en) | Fan main shaft gear box temperature anomaly detection method based on multi-feature fusion | |
| CN115545473A (en) | LSTM-based intelligent prediction method for domestic garbage throwing trend | |
| CN118051827A (en) | Power grid fault prediction method based on deep learning | |
| CN116317937A (en) | Distributed photovoltaic power station operation fault diagnosis method | |
| CN115034485A (en) | Wind power interval prediction method and device based on data space | |
| CN111091141B (en) | Photovoltaic backboard fault diagnosis method based on layered Softmax | |
| CN116542170A (en) | Drainage pipeline siltation disease dynamic diagnosis method based on SSAE and MLSTM | |
| CN117060353A (en) | Fault diagnosis method and system for high-voltage direct-current transmission system based on feedforward neural network | |
| CN115165366A (en) | Variable working condition fault diagnosis method and system for rotary machine | |
| CN114581699A (en) | Transformer state evaluation method based on deep learning model in consideration of multi-source information | |
| CN114764538B (en) | Equipment sound signal mode identification method | |
| CN117893058A (en) | Method and system for comprehensively evaluating performance of photovoltaic field station | |
| CN116933194A (en) | Abnormality screening, repairing and active complement method for power report data of wind power plant |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |