CN109039280B - Photovoltaic array fault diagnosis method based on non-principal component data characteristics - Google Patents
Photovoltaic array fault diagnosis method based on non-principal component data characteristics Download PDFInfo
- Publication number
- CN109039280B CN109039280B CN201810706750.9A CN201810706750A CN109039280B CN 109039280 B CN109039280 B CN 109039280B CN 201810706750 A CN201810706750 A CN 201810706750A CN 109039280 B CN109039280 B CN 109039280B
- Authority
- CN
- China
- Prior art keywords
- parameter
- matrix
- data
- principal component
- sample
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000003745 diagnosis Methods 0.000 title claims abstract description 15
- 239000011159 matrix material Substances 0.000 claims abstract description 93
- 238000001514 detection method Methods 0.000 claims abstract description 20
- 238000010248 power generation Methods 0.000 claims abstract description 19
- 230000009466 transformation Effects 0.000 claims abstract description 17
- 238000007621 cluster analysis Methods 0.000 claims abstract description 6
- 238000004088 simulation Methods 0.000 claims description 16
- 238000013507 mapping Methods 0.000 claims description 15
- 238000005286 illumination Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 7
- 239000013598 vector Substances 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000006835 compression Effects 0.000 claims description 3
- 238000007906 compression Methods 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 238000003064 k means clustering Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 13
- 238000000513 principal component analysis Methods 0.000 description 10
- 238000005259 measurement Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000000875 corresponding effect Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 1
- 238000011550 data transformation method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013082 photovoltaic technology Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02S—GENERATION OF ELECTRIC POWER BY CONVERSION OF INFRARED RADIATION, VISIBLE LIGHT OR ULTRAVIOLET LIGHT, e.g. USING PHOTOVOLTAIC [PV] MODULES
- H02S50/00—Monitoring or testing of PV systems, e.g. load balancing or fault identification
- H02S50/10—Testing of PV devices, e.g. of PV modules or single PV cells
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/50—Photovoltaic [PV] energy
Abstract
The invention relates to a photovoltaic array fault diagnosis method based on non-principal component data characteristics, which is characterized by comprising the following steps of: s1, collecting related parameter samples of the photovoltaic power generation array to obtain a parameter sample combination; step S2, according to the parameter sample combination, each parameter sample is normalized to obtain a parameter matrix to be measured; step S3, PCA is carried out on the standard data matrix to obtain a transformation matrix; step S4, multiplying the parameter matrix to be measured by the transformation matrix to obtain a transformed parameter matrix, and selecting two dimensions of the non-principal component of the transformed parameter matrix according to the transformed parameter matrix to obtain a two-parameter non-principal component matrix; and step S5, mixing the two-parameter non-principal component matrix with known label data, performing cluster analysis, judging the class of the two-parameter non-principal component matrix data according to the label data in the class, and completing fault detection and classification. The photovoltaic array fault detection method can effectively identify the faults of the photovoltaic array and classify the working state of the photovoltaic array.
Description
Technical Field
The invention relates to the field of photovoltaic power generation fault detection and classification, in particular to a photovoltaic array fault diagnosis method based on non-principal component data characteristics.
Background
In order to alleviate fossil energy requirements and solve ecological crisis, photovoltaic technology has become a middle-strength force in the field of new energy. The solar photovoltaic array is easy to obtain and abundant, the global deployment amount of the photovoltaic array is steadily increased, and the installed capacity of the photovoltaic array reaches hundreds to thousands of megawatts.
However, due to the special working environment of the photovoltaic system, the photovoltaic system is often threatened by various faults from the outside or the inside, such as short circuit, open circuit, shadow shielding and the like of components. Failure of a photovoltaic system can result in reduced system efficiency and potential safety risks. Therefore, efficient fault diagnosis technology becomes an indispensable component for maintenance of photovoltaic systems. When the photovoltaic power generation system breaks down, a lot of time is needed for manual detection, and the photovoltaic power generation system must be suspended to protect the safety of personnel. Photovoltaic modules are still expensive today, so longer detection times lead to more serious losses. Furthermore, photovoltaic power generation systems are often installed outdoors, often in locations that are difficult for personnel to reach, making manual inspection difficult and heavy, and also increasing the risk of maintenance work. In view of the above problems, it is imperative to establish an automatic fault diagnosis system for a photovoltaic power generation system.
Currently, many scholars propose various photovoltaic array fault detection means. Such as Artificial Neural Network (ANN) -based photovoltaic fault detection methods, simulation model building and analysis-based detection methods, statistical analysis methods, and the like. However, the usual method can only detect a particular photovoltaic array, and once the array is replaced, the parameters in the model need to be changed.
Principal Component Analysis (PCA) is a classical statistical process that yields a set of values of linearly uncorrelated variables, i.e., principal components, by orthogonal transformation. And the values of the set of linearly uncorrelated variables are transformed from the observed values of a set of possible correlated variables. The number of principal components is less than or equal to the number of observations, so PCA is typically used for dimensionality reduction. PCA always provides the best orthogonal transformation to keep the original data set with the largest variance subspace.
The output data set of the photovoltaic array has a plurality of characteristic parameters, each characteristic parameter can be regarded as one dimension of the data set, if the data set is displayed from the dimensions of common voltage, current, illumination, temperature and the like, the similarity is extremely high, and the data set is not beneficial to displaying and classifying in the same coordinate space. Therefore, the PCA method is introduced to analyze the image, and a method for effectively visually displaying and classifying the image is found.
At present, no research for applying PCA (principal component analysis) -based data transformation method to fault diagnosis and classification of photovoltaic power generation array is found in publicly published documents and patents
Disclosure of Invention
In view of this, the present invention provides a method for diagnosing a failure of a photovoltaic array based on non-principal component data characteristics, which can effectively identify the failure of the photovoltaic array and classify the operating state of the photovoltaic array.
In order to achieve the purpose, the invention adopts the following technical scheme:
a photovoltaic array fault diagnosis method based on non-principal component data features is characterized by comprising the following steps:
s1, collecting related parameter samples of the photovoltaic power generation array to obtain a parameter sample combination;
step S2, according to the parameter sample combination, each parameter sample is normalized to obtain a parameter matrix to be measured;
step S3, PCA is carried out on the standard data matrix to obtain a transformation matrix;
step S4, multiplying the parameter matrix to be measured by the transformation matrix to obtain a transformed parameter matrix, and selecting two dimensions of the non-principal component of the transformed parameter matrix according to the transformed parameter matrix to obtain a two-parameter non-principal component matrix;
and step S5, mixing the two-parameter non-principal component matrix with known label data, performing cluster analysis, judging the class of the two-parameter non-principal component matrix data according to the label data in the class, and completing fault detection and classification.
Further, the relevant parameter samples include a photovoltaic array voltage parameter sample of a maximum power point, each set of string current parameter samples of the maximum power point, a temperature parameter sample, and an illuminance parameter sample.
Further, the parameter sample combination is marked as { Uk，Ik，Tk，Sk}
Wherein k is a sample collection sequence number, where k is an integer from 1 to N, UkFor the voltage parameter samples of the kth combination of electrical parameter samples, IkRepresenting the current parameter samples, T, in the kth combination of electrical parameter sampleskRepresenting temperature parameter samples, S, in a kth combination of electrical parameter sampleskRepresenting the illumination parameter samples in the kth combination of electrical parameter samples.
Further, the step S2 is specifically: mapping the same parameter sample into an interval [0,1] by adopting a proportional compression method;
voltage parameter sample U ═ U1,U2...Uk...UN) The specific mapping formula of (2) is:
wherein U represents the data obtained after normalization of the voltage sample, UmaxRepresents the maximum value in the data set U, UminRepresents the minimum value, y, in the data set UmaxIs set to 1, yminIs set to-1;
current parameter sample I ═ I (I)1,I2...Ik...IN) The specific mapping formula of (2) is:
wherein I represents the data obtained by normalizing the voltage sample, ImaxRepresents the maximum value in the data set I, IminRepresents the minimum value in the data set I;
temperature parameter sample T ═ T1,T2...Tk...TN) The specific mapping formula of (2) is:
wherein T represents the data obtained by normalizing the voltage sample, TmaxRepresenting the maximum value in the data set T, TminRepresents the minimum value in the data set T;
illumination parameter sample S ═ S (S)1,S2...Sk…SN) The specific mapping formula of (2) is:
wherein S represents the data obtained by normalizing the voltage sample, SmaxRepresents the maximum value in the data set S, SminRepresenting the minimum value in the data set S.
Further, the step S3 specifically includes the following steps:
step S31: calculating covariance matrix C of standard data matrixXThe standard data are typical parameter values of the photovoltaic array obtained by the simulation system in various running states;
step S32: according to formula CX＝WDW-1For covariance matrix CXPerforming eigenvalue decomposition, wherein the matrix W is formed by CXA matrix composed of the feature vectors of (a), i.e. a transformation matrix;
further, the step S4 specifically includes the following steps,
step S41: collecting p samples x of relevant parameters1,x2,…xpAnd each related parameter sample acquires n samples, so that the parameter to be measuredThe number matrix X is represented as follows:
step S42: multiplying the data matrix X to be measured by the transformation matrix W to obtain a new data matrix F, and then F ═ F1,F2,…,Fp) The specific calculation method is as follows:
step S43: sequentially calculate F1,F2,…,FpAnd arranging the variances from large to small, and selecting two vectors with the minimum variance to form a two-parameter non-principal component matrix Y.
Further, the step S5 specifically includes the following steps,
step S51: adding the tag data into the obtained two-parameter non-principal component matrix Y to obtain a new matrix Z, wherein the tag data is also a two-parameter matrix;
step S52: performing K-means clustering analysis on the matrix Z;
step S53: and observing the class of the data in the two-parameter non-principal component matrix Y, and judging which state type the label data in the class belongs to, namely judging the state type of the test data.
Further, the operating state of the photovoltaic power generation array comprises: NORMAL Operation (NORMAL), single set of string OPEN (OPEN), single set of 1 component short (LL) on string, for a total of three states.
Compared with the prior art, the invention has the following beneficial effects:
the method can detect and classify faults under the condition of not influencing the work of the photovoltaic power generation system, does not need to collect a large amount of sample data to train a diagnosis model, only needs to perform principal component analysis on the standard data obtained by simulation, can directly transform the sample test data by the obtained transformation matrix, and detects the faults by comparing the clustered standard data with the label data. Through simulation and experimental test of different daily running states, the running condition of the photovoltaic power generation system can be accurately identified by the scheme.
Drawings
FIG. 1 is a block flow diagram of the present invention;
FIG. 2 is a system circuit diagram according to an embodiment of the present invention
Fig. 3 is a diagram of a system experimental photovoltaic platform in an embodiment of the invention.
FIG. 4 is a simulated photovoltaic status data tag used in an embodiment of the invention
FIG. 5 is a diagram illustrating the detection result of the simulation data of the open circuit fault in the embodiment of the present invention
FIG. 6 is a diagram illustrating the detection result of the normal state simulation data according to an embodiment of the present invention
FIG. 7 is a diagram showing the detection result of the simulation data of the short circuit fault in the embodiment of the present invention
FIG. 8 is a diagram showing the detection results of simulation data of open circuit fault and normal state in the embodiment of the present invention
FIG. 9 shows the detection results of simulation data of short-circuit fault and normal state in the embodiment of the present invention
FIG. 10 shows the result of testing simulation data of open circuit fault and short circuit fault in the embodiment of the present invention
FIG. 11 is a measured photovoltaic status data tag used in an embodiment of the present invention
FIG. 12 is a diagram illustrating the detection result of the actual measurement data of the open circuit fault in the embodiment of the present invention
FIG. 13 is a diagram illustrating the result of detecting the measured data in the normal state according to an embodiment of the present invention
FIG. 14 is a diagram illustrating the detection result of the measured short-circuit fault data according to an embodiment of the present invention
FIG. 15 shows the detection results of the actual measurement data in the normal state and the open-circuit fault in the embodiment of the present invention
FIG. 16 shows the detection results of the measured data in the short-circuit fault and normal state in the embodiment of the present invention
FIG. 17 shows the result of detecting the actual measurement data of the open circuit fault and the short circuit fault in the embodiment of the present invention
Detailed Description
The invention is further explained below with reference to the drawings and the embodiments.
The invention provides a photovoltaic array fault diagnosis method based on non-principal component data characteristics, and a flow diagram is shown in figure 1. Fig. 2 is a topological diagram of the photovoltaic power generation system of the embodiment, and the system is formed by a photovoltaic array formed by m × n photovoltaic modules and connected with a power grid through a grid-connected inverter. Under different atmospheric temperatures and irradiance, different working conditions appearing in daily operation of the photovoltaic power generation array are simulated, and data acquisition of the photovoltaic power generation system is carried out. The specific operations of the embodiment include the following steps:
s1, collecting related parameter samples of the photovoltaic power generation array to obtain a parameter sample combination;
step S2, according to the parameter sample combination, each parameter sample is normalized to obtain a parameter matrix to be measured;
step S3, PCA is carried out on the standard data matrix to obtain a transformation matrix;
step S4, multiplying the parameter matrix to be measured by the transformation matrix to obtain a transformed parameter matrix, and selecting two dimensions of the non-principal component of the transformed parameter matrix according to the transformed parameter matrix to obtain a two-parameter non-principal component matrix;
and step S5, mixing the two-parameter non-principal component matrix with known label data, performing cluster analysis, judging the class of the two-parameter non-principal component matrix data according to the label data in the class, and completing fault detection and classification.
Preferably, the photovoltaic system used for collecting data in this embodiment is composed of 18 solar panels, and 6 series-parallel and 3-parallel solar panels are formed, and grid-connected power generation is performed through an inverter, a system experiment photovoltaic platform diagram in the embodiment is shown in fig. 3, and system detailed parameters are shown in table 1.
TABLE 1 detailed parameters of the System
In this embodiment, the parameter combination described in step S1 includes a photovoltaic array voltage parameter sample of the maximum power point, and a maximum power point current parameterSamples and temperature and illumination parameter samples. The parameter sample combination is noted as (U)k，Ik，Tk，Sk) Wherein k is a sample collection sequence number, where k is an integer from 1 to N, UkFor the voltage parameter samples of the kth combination of electrical parameter samples, IkRepresenting the current parameter samples, T, in the kth combination of electrical parameter sampleskRepresenting temperature parameter samples, S, in a kth combination of electrical parameter sampleskRepresenting the illumination parameter samples in the kth combination of electrical parameter samples.
In this embodiment, the working states include: normal operation, single group series open circuit, 1 component short circuit on single group series. In particular, the embodiment collects data in 3 working states of the simulation photovoltaic power generation system: NORMAL Operation (NORMAL), single set of string OPEN circuits (OPEN), single set of string 1 component short circuits (LL). The specific information of each type of simulation data collected is shown in table 2.
TABLE 2 simulation data detailed information
In particular, the present embodiment collects measured data in 3 working states of the actual photovoltaic power generation system: NORMAL Operation (NORMAL), single set of string OPEN circuits (OPEN), single set of string 1 component short circuits (LL). The specific information of each type of simulation data collected is shown in table 3.
TABLE 3 actual measurement data details
In this embodiment, the step S2 specifically includes: mapping the same parameter sample into an interval [0,1] by adopting a proportional compression method;
voltage parameter sample U ═ U1,U2...Uk...UN) The specific mapping formula of (2) is:
wherein U represents the data obtained after normalization of the voltage sample, UmaxRepresents the maximum value in the data set U, UminRepresents the minimum value, y, in the data set UmaxIs set to 1, yminIs set to-1;
current parameter sample I ═ I (I)1,I2...Ik...IN) The specific mapping formula of (2) is:
wherein I represents the data obtained by normalizing the voltage sample, ImaxRepresents the maximum value in the data set I, IminRepresents the minimum value in the data set I;
temperature parameter sample T ═ T1,T2...Tk...TN) The specific mapping formula of (2) is:
wherein T represents the data obtained by normalizing the voltage sample, TmaxRepresenting the maximum value in the data set T, TminRepresents the minimum value in the data set T;
illumination parameter sample S ═ S (S)1,S2...Sk...SN) The specific mapping formula of (2) is:
wherein S represents the data obtained by normalizing the voltage sample, SmaxRepresents the maximum value in the data set S, SminRepresenting the minimum value in the data set S.
In this embodiment, the step S3 specifically includes the following steps:
step S31: calculating a standard data matrix X covariance matrix CXThe standard data are obtained from the simulation systemTypical parameter values under the seed operating conditions;
step S32: according to formula CX＝WDW-1For covariance matrix CXPerforming eigenvalue decomposition, wherein the matrix W is formed by CXA matrix composed of the feature vectors of (a), i.e. a transformation matrix;
further, the step S4 specifically includes the following steps, where the number p of the characteristic parameters is 4:
step S41: the collected test data has p characteristic parameters x1,x2,…xpAnd each characteristic parameter acquires n samples, and the data matrix X to be detected is represented as follows:
step S42: multiplying the data matrix X to be measured by the transformation matrix W to obtain a new data matrix F, and then F ═ F1,F2,…,Fp) The specific calculation method is as follows:
step S43: sequentially calculate F1,F2,…,FpAnd arranging the variances from large to small, and selecting two vectors with the minimum variance to form a two-parameter non-principal component matrix Y.
Further, the step S5 specifically includes the following steps,
step S51: adding the label data into the two-parameter non-principal component matrix Y obtained in the step S43 to obtain a new matrix Z, wherein the label data is also a two-parameter matrix;
step S52: performing K-means clustering analysis on the matrix Z;
step S53: and observing the class of the data in the two-parameter non-principal component matrix Y, and judging which state type the label data in the class belongs to, namely judging the state type of the test data.
Preferably, when the emulation data is used, the tag data corresponding to the normal, open and short states is as shown in fig. 4 according to step S5. The classification results obtained after the cluster analysis are shown in fig. 5, 6, and 7, and fig. 8, 9, and 10. Fig. 5 shows the identification results when 100 collected data points are all open-circuit faults, and all data of the experimental results are accurately classified as open-circuit faults; FIG. 6 shows the recognition results when all the 100 collected data points are in the normal state, and all the data are accurately classified as the normal state according to the experimental results; FIG. 7 shows the identification results of 100 collected data points with short-circuit faults, and all the data are accurately classified as short-circuit faults according to the experimental results; FIG. 8 shows the identification results of 100 collected data points, one part of which is open-circuit fault and the other part of which is normal state, all the data of the experimental results are accurately classified; FIG. 9 shows the identification results of 100 collected data points, one part of which is short-circuit fault and the other part of which is normal state, all the data of the experimental results are accurately classified; fig. 10 shows the identification results of 100 collected data points, one part is open-circuit fault and the other part is short-circuit fault, and all data are accurately classified according to the experimental results.
Preferably, when the measured data is used, the label data corresponding to the normal, open and short states is as shown in fig. 11 according to step S5. The classification results obtained after the cluster analysis are shown in fig. 12, 13, and 14, and fig. 15, 16, and 17. Fig. 12 shows the identification results when 100 collected data points are all open-circuit faults, and all data of the experimental results are accurately classified as open-circuit faults; FIG. 13 shows the recognition results when all the 100 collected data points are in the normal state, and all the data are accurately classified as the normal state according to the experimental results; FIG. 14 shows the identification results of 100 collected data points with short-circuit faults, and all the data are accurately classified as short-circuit faults according to the experimental results; FIG. 15 shows the identification results of 100 collected data points, one part of which is open-circuit fault and the other part of which is normal state, all the data of the experimental results are accurately classified; FIG. 16 shows the identification results of 100 collected data points, one part of which is short-circuit fault and the other part of which is normal state, all the data of the experimental results are accurately classified; fig. 17 shows the identification results of 100 collected data points, one part of which is open-circuit fault and the other part of which is short-circuit fault, and all the data of the experimental results are accurately classified.
The above description is only a preferred embodiment of the present invention, and all equivalent changes and modifications made in accordance with the claims of the present invention should be covered by the present invention.
Claims (6)
1. A photovoltaic array fault diagnosis method based on non-principal component data features is characterized by comprising the following steps:
s1, collecting related parameter samples of the photovoltaic power generation array to obtain a parameter sample combination;
step S2, according to the parameter sample combination, each parameter sample is normalized to obtain a parameter matrix to be measured;
step S3, PCA is carried out on the standard data matrix to obtain a transformation matrix; the method specifically comprises the following steps:
step S31: calculating covariance matrix C of standard data matrixXThe standard data are typical parameter values of the photovoltaic array obtained by the simulation system in various running states;
step S32: according to formula CX＝WDW-1For covariance matrix CXPerforming eigenvalue decomposition, wherein the matrix W is formed by CXA matrix composed of the feature vectors of (a), i.e. a transformation matrix;
step S4, multiplying the parameter matrix to be measured by the transformation matrix to obtain a transformed parameter matrix, and selecting two dimensions of the non-principal component of the transformed parameter matrix according to the transformed parameter matrix to obtain a two-parameter non-principal component matrix;
and step S5, mixing the two-parameter non-principal component matrix with known label data, performing cluster analysis, judging the class of the two-parameter non-principal component matrix data according to the label data in the class, and completing fault detection and classification.
2. The photovoltaic array fault diagnosis method based on non-principal component data features according to claim 1, characterized in that: the related parameter samples comprise photovoltaic array voltage parameter samples of a maximum power point, various groups of series current parameter samples of the maximum power point, temperature parameter samples and illumination parameter samples.
3. The photovoltaic array fault diagnosis method based on non-principal component data features according to claim 2, characterized in that:
the parameter sample combination is denoted as { Uk，Ik，Tk，Sk}
Wherein k is a sample collection sequence number, where k is an integer from 1 to N, UkFor the voltage parameter samples of the kth combination of electrical parameter samples, IkRepresenting the current parameter samples, T, in the kth combination of electrical parameter sampleskRepresenting temperature parameter samples, S, in a kth combination of electrical parameter sampleskRepresenting the illumination parameter samples in the kth combination of electrical parameter samples.
4. The photovoltaic array fault diagnosis method based on the non-principal component data characteristics according to claim 3, characterized in that: the step S2 specifically includes: mapping the same parameter sample into an interval [0,1] by adopting a proportional compression method;
voltage parameter sample U ═ U1,U2...Uk...UN) The specific mapping formula of (2) is:
wherein U represents the data obtained after normalization of the voltage sample, UmaxRepresents the maximum value in the data set U, UminRepresents the minimum value, y, in the data set UmaxIs set to 1, yminIs set to-1;
current parameter sample I ═ I (I)1,I2...Ik...IN) The specific mapping formula of (2) is:
wherein I represents the data obtained by normalizing the voltage sample, ImaxRepresents the maximum value in the data set I, IminRepresents the minimum value in the data set I;
temperature parameter sample T ═ T1,T2...Tk...TN) The specific mapping formula of (2) is:
wherein T represents the data obtained by normalizing the voltage sample, TmaxRepresenting the maximum value in the data set T, TminRepresents the minimum value in the data set T;
illumination parameter sample S ═ S (S)1,S2...Sk...SN) The specific mapping formula of (2) is:
wherein S represents the data obtained by normalizing the voltage sample, SmaxRepresents the maximum value in the data set S, SminRepresenting the minimum value in the data set S.
5. The method according to claim 1, wherein the method comprises the following steps: the step S4 specifically includes the following steps,
step S41: collecting p samples x of relevant parameters1,x2,…xpAnd each relevant parameter sample acquires n samples, and the parameter matrix X to be measured is represented as follows:
step S42: multiplying the data matrix X to be measured by the transformation matrix W to obtain a new data matrix F, and then F ═ F1,F2,…,Fp) The specific calculation method is as follows:
step S43: sequentially calculate F1,F2,…,FpAnd arranging the variances from large to small, and selecting two vectors with the minimum variance to form a two-parameter non-principal component matrix Y.
6. The photovoltaic array fault diagnosis method based on non-principal component data features according to claim 5, characterized in that: the step S5 specifically includes the following steps,
step S51: adding the tag data into the obtained two-parameter non-principal component matrix Y to obtain a new matrix Z, wherein the tag data is also a two-parameter matrix;
step S52: performing K-means clustering analysis on the matrix Z;
step S53: and observing the class of the data in the two-parameter non-principal component matrix Y, and judging which state type the label data in the class belongs to, namely judging the state type of the test data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810706750.9A CN109039280B (en) | 2018-07-02 | 2018-07-02 | Photovoltaic array fault diagnosis method based on non-principal component data characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810706750.9A CN109039280B (en) | 2018-07-02 | 2018-07-02 | Photovoltaic array fault diagnosis method based on non-principal component data characteristics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109039280A CN109039280A (en) | 2018-12-18 |
| CN109039280B true CN109039280B (en) | 2020-02-18 |
Family
ID=65522057
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810706750.9A Expired - Fee Related CN109039280B (en) | 2018-07-02 | 2018-07-02 | Photovoltaic array fault diagnosis method based on non-principal component data characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109039280B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109583515A (en) * | 2018-12-20 | 2019-04-05 | 福州大学 | A kind of photovoltaic power generation fault detection and classification method based on BP_Adaboost |
| CN110286330A (en) * | 2019-08-15 | 2019-09-27 | 莆田市烛火信息技术有限公司 | One kind being used for lithium battery fault detection system |
| CN112182967B (en) * | 2020-09-28 | 2022-10-18 | 四川渔光物联技术有限公司 | Automatic photovoltaic module modeling method based on thermal imaging instrument |
| CN116111948B (en) * | 2022-12-07 | 2024-04-23 | 中广核贵州安顺关岭新能源有限公司 | Photovoltaic branch equipment fault early warning system |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103294896B (en) * | 2013-05-09 | 2016-08-17 | 国家电网公司 | A kind of photovoltaic plant benchmark photovoltaic component system of selection based on principal component analysis |
| CN106324405A (en) * | 2016-09-07 | 2017-01-11 | 南京工程学院 | Transformer fault diagnosis method based on improved principal component analysis |
| CN106777984B (en) * | 2016-12-19 | 2019-02-22 | 福州大学 | A method of photovoltaic array Working state analysis and fault diagnosis are realized based on density clustering algorithm |
| CN107273917A (en) * | 2017-05-26 | 2017-10-20 | 电子科技大学 | A kind of Method of Data with Adding Windows based on parallelization Principal Component Analysis Algorithm |
-
2018
- 2018-07-02 CN CN201810706750.9A patent/CN109039280B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN109039280A (en) | 2018-12-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109039280B (en) | Photovoltaic array fault diagnosis method based on non-principal component data characteristics | |
| Harrou et al. | Improved $ k $ NN-based monitoring schemes for detecting faults in PV systems | |
| CN107340456B (en) | Power distribution network operating condition intelligent identification Method based on multiple features analysis | |
| CN108062571B (en) | Photovoltaic array fault diagnosis method based on differential evolution random forest classifier | |
| CN106777984B (en) | A method of photovoltaic array Working state analysis and fault diagnosis are realized based on density clustering algorithm | |
| CN104753461B (en) | Method for diagnosing and classifying faults of photovoltaic power generation arrays on basis of particle swarm optimization support vector machines | |
| CN109660206B (en) | Wasserstein GAN-based photovoltaic array fault diagnosis method | |
| CN109766952B (en) | Photovoltaic array fault detection method based on partial least square method and extreme learning machine | |
| CN105141255A (en) | Fault diagnosis method of photovoltaic array | |
| Zaki et al. | Deep‐learning–based method for faults classification of PV system | |
| CN103116090A (en) | Three-phrase pulse-width modulation (PWM) rectifier fault diagnosis method based on wavelet packet analysis and support vector machine | |
| Liu et al. | Time series clustering and physical implication for photovoltaic array systems with unknown working conditions | |
| CN109034220A (en) | A kind of intelligent photovoltaic array method for diagnosing faults based on optimal rotation forest | |
| Fatama et al. | A multi feature based islanding classification technique for distributed generation systems | |
| CN113702767B (en) | Island direct-current microgrid fault diagnosis method based on wavelet sliding window energy | |
| CN113326667B (en) | Solar cell panel fault diagnosis method and system | |
| CN117110798B (en) | Fault detection method and system for intelligent power distribution network | |
| Zhu et al. | Photovoltaic failure diagnosis using sequential probabilistic neural network model | |
| Joga et al. | Fault Diagnosis in PV system using DWT and Ensembled k-NN Machine Learning Classifier | |
| Dong et al. | Fault diagnosis and classification in photovoltaic systems using scada data | |
| CN108875796A (en) | Diagnosing failure of photovoltaic array method based on linear discriminant analysis and support vector machines | |
| CN109117865A (en) | One kind realizing photovoltaic array Working state analysis and method for diagnosing faults based on principal component analysis | |
| CN112132226A (en) | Method for diagnosing defects of +/-1100 kV composite insulator | |
| CN115906350A (en) | Micro-grid fault positioning and diagnosing method | |
| Attouri et al. | Fault detection in photovoltaic systems using machine learning technique |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200218 Termination date: 20210702 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |