CN115758162A - Data prediction model training and photovoltaic inverter fault prediction method and device - Google Patents
Data prediction model training and photovoltaic inverter fault prediction method and device Download PDFInfo
- Publication number
- CN115758162A CN115758162A CN202211575714.6A CN202211575714A CN115758162A CN 115758162 A CN115758162 A CN 115758162A CN 202211575714 A CN202211575714 A CN 202211575714A CN 115758162 A CN115758162 A CN 115758162A
- Authority
- CN
- China
- Prior art keywords
- data
- prediction model
- phase
- data prediction
- training
- 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
Images
Landscapes
- Inverter Devices (AREA)
Abstract
The invention discloses a data prediction model training method and a photovoltaic inverter fault prediction method and device, wherein the data prediction model training method comprises the following steps: acquiring a data set of historical state quantities of the photovoltaic inverter, wherein each historical state quantity comprises phase voltage, phase current, line voltage, line current and IGBT temperature of each phase; taking phase voltage, phase current, line voltage and line current in each historical state quantity as characteristics, and taking each phase IGBT temperature in each historical state quantity as a label value; and training a preset data prediction model by using the data set to obtain the target weight of the data prediction model and obtain the trained data prediction model. The data prediction model training and photovoltaic inverter fault prediction method and device can effectively improve the generalization capability of the model, further effectively improve the accuracy of the predicted data and enable the fault prediction of the photovoltaic inverter to be more accurate.
Description
Technical Field
The invention relates to the technical field of photovoltaic power generation, in particular to a method and a device for training a data prediction model and predicting a fault of a photovoltaic inverter.
Background
Photovoltaic energy is a clean energy source, and as more and more photovoltaic power generation systems are incorporated into the power grid globally, effective maintenance of the equipment is currently important. The photovoltaic inverter is a core device in the photovoltaic power generation system, and the state of the photovoltaic inverter is effectively analyzed, so that the whole photovoltaic power generation system can safely and stably operate.
At present, most photovoltaic inverter fault prediction methods perform data prediction based on real-time monitored state quantities, and perform fault prediction on a photovoltaic inverter by using prediction data, but the existing training methods for data prediction models of photovoltaic inverters are simple, so that the generalization capability of the models is poor, the prediction data is not accurate enough, and the fault prediction of the photovoltaic inverter is also inaccurate.
Disclosure of Invention
The purpose of the invention is: the method, the device, the computer equipment and the storage medium for training the data prediction model and predicting the faults of the photovoltaic inverter are provided, the generalization capability of the model can be effectively improved, the accuracy of predicted data can be effectively improved, and the faults of the photovoltaic inverter can be more accurately predicted.
In order to achieve the above object, in a first aspect, the present invention provides a data prediction model training method, including:
acquiring a data set of historical state quantities of the photovoltaic inverter, wherein each historical state quantity comprises phase voltage, phase current, line voltage, line current and IGBT temperature of each phase;
taking phase voltage, phase current, line voltage and line current in each historical state quantity as characteristics, and taking each phase IGBT temperature in each historical state quantity as a label value;
and training a preset data prediction model by using the data set to obtain the target weight of the data prediction model and obtain the trained data prediction model.
In a preferred embodiment of the present invention, the training a preset data prediction model by using the data set to obtain a target weight of the data prediction model includes:
determining two target subdata sets conforming to preset dispersion from the data set;
and training a preset data prediction model by using the two target sub-data sets to obtain the target weight of the data prediction model.
In a preferred embodiment of the present invention, the determining two target sub-data sets meeting a preset dispersion from the data set includes:
repeatedly sampling the data sets by using a preset sampling method and a preset proportion to obtain a plurality of groups of sampling data sets, wherein each group of sampling data sets comprises two subdata sets;
and respectively calculating the dispersion of each group of the sampling data sets, determining a target sampling data set conforming to the preset dispersion, and determining two sub data sets of the target sampling data set as two target sub data sets.
In a preferred embodiment of the present invention, the training a preset data prediction model by using the two target sub-data sets to obtain a target weight of the data prediction model includes:
respectively taking the two target subdata sets as source domain data and target domain data;
and inputting the source domain data and the target domain data to a preset data prediction model for training to obtain the target weight of the data prediction model.
In a preferred embodiment of the present invention, the data prediction model uses the following expression of the loss function:
wherein LOSS is a LOSS value, D s For source domain characteristic data, y s For source domain tag data, D t For the target domain syndrome data, λ ∈ [0, + ∞) ] is the hyper-parameter, n is the source domain syndrome data D s The number of the cells.
In a second aspect, the present invention provides a method for predicting a fault of a photovoltaic inverter, including:
acquiring real-time state quantities of the photovoltaic inverter, wherein the real-time state quantities comprise phase voltage, phase current, line voltage, line current and IGBT (insulated gate bipolar translator) temperature of each phase;
taking the phase voltage, the phase current, the line voltage and the line current in the real-time state quantity as characteristics, and taking the temperature of each phase of IGBT in the real-time state quantity as a label value;
obtaining the IGBT temperature at the next moment by using the real-time state quantity and a data prediction model obtained by the data prediction model training method;
and predicting the fault of the photovoltaic inverter according to the IGBT temperature at the next moment to obtain a fault prediction result.
In a third aspect, the present invention provides a data prediction model training apparatus, including:
the photovoltaic inverter control system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring a data set of historical state quantities of a photovoltaic inverter, and each historical state quantity comprises a phase voltage, a phase current, a line voltage, a line current and an IGBT temperature of each phase;
the data processing module is used for taking phase voltage, phase current, line voltage and line current in each historical state quantity as characteristics and taking the IGBT temperature of each phase in each historical state quantity as a label value;
and the training module is used for training a preset data prediction model by using the data set to obtain the target weight of the data prediction model and obtain the trained data prediction model.
In a fourth aspect, the present invention provides a photovoltaic inverter failure prediction apparatus, including:
the photovoltaic inverter comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring real-time state quantities of the photovoltaic inverter, and the real-time state quantities comprise phase voltage, phase current, line voltage, line current and the temperature of each phase IGBT;
the data processing module is used for taking the phase voltage, the phase current, the line voltage and the line current in the real-time state quantity as characteristics and taking the temperature of each phase of IGBT in the real-time state quantity as a label value;
the data prediction module is used for obtaining the IGBT temperature at the next moment by utilizing the real-time state quantity and a data prediction model obtained by the data prediction model training device;
and the fault prediction module is used for predicting the fault of the photovoltaic inverter according to the IGBT temperature at the next moment to obtain a fault prediction result.
In a fifth aspect, the present invention provides a computer device, including a memory for storing a computer program and a processor for executing the computer program to make the computer device execute the above data prediction model training method or the above photovoltaic inverter fault prediction method.
In a sixth aspect, the present invention provides a computer-readable storage medium, which stores a computer program, and the computer program, when executed by a processor, implements the above data prediction model training method, or the above photovoltaic inverter fault prediction method.
Compared with the prior art, the data prediction model training method, the data prediction model training device, the photovoltaic inverter fault prediction method, the photovoltaic inverter fault prediction device, the computer equipment and the storage medium have the beneficial effects that:
according to the method, the phase voltage, the phase current, the line voltage and the line current in each historical state quantity are taken as characteristics through the acquired data set of the historical state quantity of the photovoltaic inverter, the IGBT temperature of each phase in each historical state quantity is taken as a label value, the historical state quantity of the photovoltaic inverter is divided more clearly, the historical state quantity of the photovoltaic inverter can be more effectively utilized, the preset data prediction model is trained by the data set, the target weight of the data prediction model is obtained, the trained data prediction model is obtained, the generalization capability of the model can be effectively improved, the accuracy of the predicted data can be effectively improved, and the fault prediction of the photovoltaic inverter is more accurate.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments of the present invention will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and it will be apparent to those skilled in the art that other relevant drawings can be obtained based on the drawings without inventive effort.
FIG. 1 is a schematic flow chart of a data prediction model training method according to an embodiment of the present invention;
FIG. 2 is a block diagram of a data prediction model training apparatus according to a second embodiment of the present invention;
fig. 3 is a schematic flowchart of a method for predicting a fault of a photovoltaic inverter according to a third embodiment of the present invention;
fig. 4 is a block diagram of a photovoltaic inverter failure prediction apparatus according to a fourth embodiment of the present invention;
fig. 5 is a schematic diagram of an internal structure of a computer device according to a fifth embodiment of the present invention.
Detailed Description
The following detailed description of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
At present, most photovoltaic inverter fault prediction methods perform data prediction based on real-time monitored state quantities, and perform fault prediction on a photovoltaic inverter by using prediction data, but the existing training methods for data prediction models of photovoltaic inverters are simple, so that the generalization capability of the models is poor, the prediction data is not accurate enough, and the fault prediction of the photovoltaic inverter is also inaccurate.
In view of the problems in the prior art, embodiments of the present invention provide a method, an apparatus, a computer device, and a storage medium for training a data prediction model and predicting a fault of a photovoltaic inverter, which can effectively improve the generalization ability of the model, and further can effectively improve the accuracy of data prediction, so that the fault prediction of the photovoltaic inverter is more accurate.
Example one
Referring to fig. 1, fig. 1 is a schematic flow chart of a data prediction model training method according to an embodiment of the present invention.
The data prediction model training method in the embodiment of the invention can be applied to computer equipment such as a server.
In one embodiment, the present invention provides a data prediction model training method, comprising the steps of:
step S110, acquiring a data set of historical state quantities of the photovoltaic inverter, wherein each historical state quantity comprises phase voltage, phase current, line voltage, line current and IGBT temperature of each phase.
In one embodiment, the data prediction model trained by the data prediction model training method of the invention is used for data prediction of the photovoltaic inverter.
In one embodiment, the computer device can obtain a data set of historical state quantities of the photovoltaic inverter through historical data acquired by the sensor; the temperature of each phase of IGBT (Insulated gate bipolar Transistor) is the temperature of each phase of IGBT; optionally, the data set of the historical state quantities of the photovoltaic inverter is a time series data set.
And step S120, taking the phase voltage, the phase current, the line voltage and the line current in each historical state quantity as characteristics, and taking the IGBT temperature of each phase in each historical state quantity as a label value.
In one embodiment, the computer device may employ a sliding window approach to characterize the phase voltages, phase currents, line voltages, line currents in each historical state quantity and label each phase IGBT temperature in each historical state quantity.
Step S130, training a preset data prediction model by using a data set to obtain the target weight of the data prediction model, and obtaining the trained data prediction model.
In one embodiment, the predetermined data prediction model is a neural network model.
In one embodiment, when the computer device trains a preset data prediction model by using a data set to obtain a target weight of the data prediction model, the computer device may:
determining two target subdata sets conforming to preset dispersion from the data set;
and training a preset data prediction model by using the two target sub-data sets to obtain the target weight of the data prediction model.
In this embodiment, the preset dispersion may be the maximum KL dispersion.
In one embodiment, the data used for training the data prediction model can be greatly reduced through the method, the training time of the data prediction model is greatly reduced, the two determined target sub-data sets conforming to the preset dispersion degree have better training significance and training effect, the data of the data prediction model training can be reduced, meanwhile, the data prediction model can be better trained, and the model training efficiency is improved.
In this embodiment, when the computer device determines two target sub-data sets conforming to the preset dispersion from the data set, the computer device may:
repeatedly sampling the data sets by using a preset sampling method and a preset proportion to obtain a plurality of groups of sampling data sets, wherein each group of sampling data sets comprises two subdata sets;
and respectively calculating the dispersion of each group of sampling data sets, determining a target sampling data set conforming to the preset dispersion, and determining two sub-data sets of the target sampling data set as two target sub-data sets.
In this embodiment, the computer device may repeatedly sample the data set by using a monte carlo sampling method according to a preset ratio to obtain a plurality of sets of sampled data sets, where each set of sampled data sets includes two sub data sets.
In this embodiment, it can be understood that the dispersion of each set of sampled data is calculated, that is, the dispersion of two subdata sets in each set of sampled data is calculated.
In this embodiment, the data set may be partitioned more simply, quickly, and reasonably by the above manner, and the sub data sets and the target sub data set may be obtained more reasonably.
In this embodiment, when the computer device trains a preset data prediction model by using two target sub-data sets to obtain a target weight of the data prediction model, the computer device may:
respectively taking the two target subdata sets as source domain data and target domain data;
and inputting the source domain data and the target domain data to a preset data prediction model for training to obtain the target weight of the data prediction model.
In this embodiment, the two target sub-data sets are respectively input to a preset data prediction model as source domain data and target domain data for training, so that the training effect of the data prediction model is better.
In this embodiment, the data prediction model may include a feedforward network and a loss function, wherein, optionally, the feedforward network may select a GRU-Attention network; optionally, the data prediction model employs the following expression of the loss function:
wherein LOSS is a LOSS value, D s For source domain characteristic data, y s For source domain tagged data, D t For the target domain characteristic data, λ ∈ [0, + ∞ ]) is hyper-parameter, n is the source domain characteristic data D s The number of the cells.
According to the data prediction model training method, the phase voltage, the phase current, the line voltage and the line current in each historical state quantity are taken as characteristics through the acquired data set of the historical state quantity of the photovoltaic inverter, the IGBT temperature of each phase in each historical state quantity is taken as a label value, the historical state quantity of the photovoltaic inverter is more clearly divided, the historical state quantity of the photovoltaic inverter can be more effectively utilized, the preset data prediction model is trained through the data set, the target weight of the data prediction model is obtained, the trained data prediction model is obtained, the generalization capability of the model can be effectively improved, the accuracy of predicted data can be effectively improved, and the fault prediction of the photovoltaic inverter is more accurate.
Example two
In order to implement the method corresponding to the above embodiment to achieve the corresponding functions and technical effects, a data prediction model training apparatus is provided below.
Referring to fig. 2, fig. 2 is a block diagram of a data prediction model training apparatus according to an embodiment of the present invention.
In one embodiment, the data prediction model training apparatus of the present invention includes:
an obtaining module 210, configured to obtain a data set of historical state quantities of the photovoltaic inverter, where each historical state quantity includes a phase voltage, a phase current, a line voltage, a line current, and an IGBT temperature of each phase;
the data processing module 220 is used for taking the phase voltage, the phase current, the line voltage and the line current in each historical state quantity as characteristics and taking the IGBT temperature of each phase in each historical state quantity as a tag value;
the training module 230 is configured to train a preset data prediction model by using a data set, obtain a target weight of the data prediction model, and obtain the trained data prediction model.
According to the data prediction model training device, phase voltage, phase current, line voltage and line current in each historical state quantity are taken as characteristics through the acquired data set of the historical state quantity of the photovoltaic inverter, the IGBT temperature of each phase in each historical state quantity is taken as a label value, the historical state quantity of the photovoltaic inverter is subjected to data division more clearly, the historical state quantity of the photovoltaic inverter can be used more effectively, the preset data prediction model is trained by the data set, the target weight of the data prediction model is obtained, the trained data prediction model is obtained, the generalization capability of the model can be improved effectively, the accuracy of the predicted data can be improved effectively, and the fault prediction of the photovoltaic inverter is more accurate.
In one embodiment, the training module 230 may be specifically configured to:
determining two target subdata sets conforming to preset dispersion from the data set;
and training a preset data prediction model by using the two target sub-data sets to obtain the target weight of the data prediction model.
In this embodiment, when determining two target sub-data sets conforming to the preset dispersion from the data set, the training module 230 may:
repeatedly sampling the data sets by using a preset sampling method and a preset proportion to obtain a plurality of groups of sampling data sets, wherein each group of sampling data sets comprises two subdata sets;
and respectively calculating the dispersion of each group of sampling data sets, determining a target sampling data set conforming to the preset dispersion, and determining two sub-data sets of the target sampling data set as two target sub-data sets.
In this embodiment, when the training module 230 trains a preset data prediction model by using two target sub-data sets to obtain a target weight of the data prediction model, the training module may:
respectively taking the two target subdata sets as source domain data and target domain data;
and inputting the source domain data and the target domain data to a preset data prediction model for training to obtain the target weight of the data prediction model.
The data prediction model training apparatus may implement the data prediction model training method described above. For specific limitations and the rest of the above data prediction model training apparatus, reference may be made to the above data prediction model training method, and details are not repeated in the embodiments.
EXAMPLE III
Referring to fig. 3, fig. 3 is a schematic flowchart of a method for predicting a fault of a photovoltaic inverter according to an embodiment of the present invention.
The photovoltaic inverter fault prediction method disclosed by the embodiment of the invention can be applied to computer equipment such as a server.
In one embodiment, the present invention provides a method for predicting a fault of a photovoltaic inverter, comprising the steps of:
step S310, acquiring real-time state quantities of the photovoltaic inverter, wherein the real-time state quantities comprise phase voltage, phase current, line voltage, line current and IGBT temperature of each phase.
In one embodiment, the computer device can acquire the real-time state quantity of the photovoltaic inverter through a sensor; the temperature of each phase of IGBT (Insulated Gate Bipolar Transistor) is the temperature of each phase of IGBT.
And step S320, taking the phase voltage, the phase current, the line voltage and the line current in the real-time state quantity as characteristics, and taking the IGBT temperature of each phase in the real-time state quantity as a label value.
In one embodiment, the computer device may employ a sliding window approach, characterizing the phase voltages, phase currents, line voltages, line currents in the real-time state quantities, and tagging the IGBT temperature for each phase in the real-time state quantities.
And step S330, obtaining the IGBT temperature at the next moment by using the real-time state quantity and the data prediction model.
It is to be understood that the data prediction model is the data prediction model obtained by training in the first embodiment.
It can be understood that the next time is a time next to the time corresponding to the real-time state quantity.
For the data prediction model, the contents in the above first embodiment can be referred to, and details are not repeated in the embodiments.
And step S340, performing fault prediction on the photovoltaic inverter according to the IGBT temperature at the next moment to obtain a fault prediction result.
In one embodiment, the computer device may perform fault prediction on the photovoltaic inverter according to whether the temperature of the IGBT at the next moment is greater than a preset temperature, so as to obtain a fault prediction result; understandably, when the temperature of the IGBT at the next moment is greater than the preset temperature, obtaining a fault prediction result of the photovoltaic inverter fault; and when the IGBT temperature at the next moment is not higher than the preset temperature, obtaining the normal fault prediction result of the photovoltaic inverter.
In one embodiment, the method for predicting the failure of the photovoltaic inverter of the present invention may further include the steps of:
and updating the target weight of the data prediction model on line according to the IGBT temperature at the next moment and the IGBT temperature in the real-time state quantity at the next moment.
According to the photovoltaic inverter fault prediction method, the obtained real-time state quantity of the photovoltaic inverter and the data prediction model with higher generalization capability and higher data prediction accuracy of the first embodiment are utilized, the IGBT temperature at the next moment can be predicted more accurately, and therefore the fault prediction of the photovoltaic inverter can be more accurate.
Example four
In order to implement the method corresponding to the above embodiment to achieve the corresponding functions and technical effects, a photovoltaic inverter failure prediction apparatus is provided below.
Referring to fig. 4, fig. 4 is a block diagram of a failure prediction apparatus for a photovoltaic inverter according to an embodiment of the present invention.
In one embodiment, the photovoltaic inverter failure prediction apparatus of the present invention includes:
the obtaining module 410 is configured to obtain real-time state quantities of the photovoltaic inverter, where the real-time state quantities include phase voltage, phase current, line voltage, line current, and each phase IGBT temperature;
the data processing module 420 is configured to use the phase voltage, the phase current, the line voltage, and the line current in the real-time state quantity as characteristics, and use the IGBT temperature of each phase in the real-time state quantity as a tag value;
the data prediction module 430 is configured to obtain the IGBT temperature at the next moment by using the real-time state quantity and the data prediction model;
and the fault prediction module 440 is configured to perform fault prediction on the photovoltaic inverter according to the IGBT temperature at the next time, so as to obtain a fault prediction result.
According to the photovoltaic inverter fault prediction device, the obtained real-time state quantity of the photovoltaic inverter and the data prediction model with higher generalization capability and higher data prediction accuracy of the first embodiment can be used for more accurately predicting the IGBT temperature at the next moment, so that the photovoltaic inverter fault prediction is more accurate.
In one embodiment, the failure prediction apparatus for a photovoltaic inverter according to the present invention may further include:
and the online updating module is used for updating the target weight of the data prediction model online according to the IGBT temperature at the next moment and the IGBT temperature in the real-time state quantity at the next moment.
The photovoltaic inverter failure prediction device can implement the photovoltaic inverter failure prediction method. Specific limitations and other contents of the above embodiments of the photovoltaic inverter fault prediction apparatus may refer to the contents of the above photovoltaic inverter fault prediction method, and details are not repeated in the embodiments.
EXAMPLE five
In one embodiment, the present invention provides a computer device, including a memory for storing a computer program and a processor for executing the computer program to make the computer device execute the above data prediction model training method or the above photovoltaic inverter failure prediction method.
Alternatively, the computer device may be a server.
In one embodiment, the internal structure of the computer device of the present invention may be as shown in FIG. 5.
In one embodiment, the present invention provides a computer-readable storage medium storing a computer program, which when executed by a processor, implements the above-mentioned data prediction model training method, or the above-mentioned photovoltaic inverter failure prediction method.
In the embodiments provided by the present invention, it should be understood that the disclosed apparatus and method may be implemented in other manners. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, the functional modules in the embodiments of the present invention may be integrated together to form an independent part, or each module may exist alone, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The above description is only an example of the present invention, and is not intended to limit the scope of the present invention, and it will be apparent to those skilled in the art that various modifications and variations can be made in the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of additional like elements in a process, method, article, or apparatus that comprises the element.
Claims (10)
1. A data prediction model training method is characterized by comprising the following steps:
acquiring a data set of historical state quantities of a photovoltaic inverter, wherein each historical state quantity comprises a phase voltage, a phase current, a line voltage, a line current and an IGBT temperature of each phase;
taking phase voltage, phase current, line voltage and line current in each historical state quantity as characteristics, and taking each phase IGBT temperature in each historical state quantity as a label value;
and training a preset data prediction model by using the data set to obtain the target weight of the data prediction model and obtain the trained data prediction model.
2. The method for training the data prediction model according to claim 1, wherein the training a preset data prediction model by using the data set to obtain a target weight of the data prediction model comprises:
determining two target subdata sets conforming to preset dispersion from the data sets;
and training a preset data prediction model by using the two target sub-data sets to obtain the target weight of the data prediction model.
3. The method for training a data prediction model according to claim 2, wherein the determining two target sub-data sets conforming to a preset dispersion from the data set comprises:
repeatedly sampling the data sets by using a preset sampling method and a preset proportion to obtain a plurality of groups of sampling data sets, wherein each group of sampling data sets comprises two subdata sets;
and respectively calculating the dispersion of each group of sampling data sets, determining a target sampling data set conforming to the preset dispersion, and determining two sub data sets of the target sampling data set as two target sub data sets.
4. The method for training the data prediction model according to claim 2, wherein the training a preset data prediction model by using the two target sub data sets to obtain a target weight of the data prediction model comprises:
respectively taking the two target subdata sets as source domain data and target domain data;
and inputting the source domain data and the target domain data to a preset data prediction model for training to obtain the target weight of the data prediction model.
5. The method for training the data prediction model according to claim 4, wherein the loss function adopted by the data prediction model is expressed as follows:
wherein LOSS is a LOSS value, D s For source domain characteristic data, y s For source domain tagged data, D t For the target domain syndrome data, λ ∈ [0, + ∞) ] is the hyper-parameter, n is the source domain syndrome data D s The number of the cells.
6. A method for predicting a fault of a photovoltaic inverter, comprising:
acquiring real-time state quantities of the photovoltaic inverter, wherein the real-time state quantities comprise phase voltage, phase current, line voltage, line current and IGBT (insulated gate bipolar translator) temperature of each phase;
taking phase voltage, phase current, line voltage and line current in the real-time state quantity as characteristics, and taking the temperature of each phase of IGBT in the real-time state quantity as a label value;
obtaining the IGBT temperature at the next moment by using the real-time state quantity and a data prediction model obtained by the data prediction model training method according to any one of claims 1 to 5;
and predicting the fault of the photovoltaic inverter according to the IGBT temperature at the next moment to obtain a fault prediction result.
7. A data prediction model training apparatus, comprising:
the photovoltaic inverter control system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring a data set of historical state quantities of the photovoltaic inverter, and each historical state quantity comprises a phase voltage, a phase current, a line voltage, a line current and an IGBT temperature of each phase;
the data processing module is used for taking phase voltage, phase current, line voltage and line current in each historical state quantity as characteristics and taking each phase IGBT temperature in each historical state quantity as a label value;
and the training module is used for training a preset data prediction model by using the data set to obtain the target weight of the data prediction model and obtain the trained data prediction model.
8. A photovoltaic inverter failure prediction apparatus, comprising:
the acquisition module is used for acquiring real-time state quantities of the photovoltaic inverter, wherein the real-time state quantities comprise phase voltage, phase current, line voltage, line current and IGBT temperature of each phase;
the data processing module is used for taking the phase voltage, the phase current, the line voltage and the line current in the real-time state quantity as characteristics and taking the temperature of each phase of IGBT in the real-time state quantity as a label value;
a data prediction module for obtaining the IGBT temperature at the next moment using the real-time state quantity and the data prediction model obtained by the data prediction model training apparatus according to claim 7;
and the fault prediction module is used for predicting the fault of the photovoltaic inverter according to the IGBT temperature at the next moment to obtain a fault prediction result.
9. A computer device comprising a memory for storing a computer program and a processor executing the computer program to cause the computer device to perform the data prediction model training method of any one of claims 1 to 5, or the pv inverter fault prediction method of claim 6.
10. A computer-readable storage medium, characterized in that it stores a computer program which, when executed by a processor, implements the data prediction model training method of any one of claims 1 to 5, or the photovoltaic inverter fault prediction method of claim 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211575714.6A CN115758162A (en) | 2022-12-07 | 2022-12-07 | Data prediction model training and photovoltaic inverter fault prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211575714.6A CN115758162A (en) | 2022-12-07 | 2022-12-07 | Data prediction model training and photovoltaic inverter fault prediction method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115758162A true CN115758162A (en) | 2023-03-07 |
Family
ID=85344846
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211575714.6A Pending CN115758162A (en) | 2022-12-07 | 2022-12-07 | Data prediction model training and photovoltaic inverter fault prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115758162A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117713580A (en) * | 2024-02-06 | 2024-03-15 | 杭州利沃得电源有限公司 | Switching method and device for modulation mode of photovoltaic inverter |
| CN117713580B (en) * | 2024-02-06 | 2024-05-24 | 杭州利沃得电源有限公司 | Switching method and device for modulation mode of photovoltaic inverter |
-
2022
- 2022-12-07 CN CN202211575714.6A patent/CN115758162A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117713580A (en) * | 2024-02-06 | 2024-03-15 | 杭州利沃得电源有限公司 | Switching method and device for modulation mode of photovoltaic inverter |
| CN117713580B (en) * | 2024-02-06 | 2024-05-24 | 杭州利沃得电源有限公司 | Switching method and device for modulation mode of photovoltaic inverter |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Pi et al. | A grey prediction approach to forecasting energy demand in China | |
| Ayvaz et al. | Electricity consumption forecasting for Turkey with nonhomogeneous discrete grey model | |
| Goh et al. | Statistical control of a six sigma process | |
| Kumar et al. | Availability analysis of repairable mechanical systems using analytical semi-Markov approach | |
| CN109245099A (en) | A kind of electric load discrimination method, device, equipment and readable storage medium storing program for executing | |
| Cheng et al. | Enhanced state estimation and bad data identification in active power distribution networks using photovoltaic power forecasting | |
| Ahmadi et al. | A new false data injection attack detection model for cyberattack resilient energy forecasting | |
| Hüsler et al. | Estimation for the generalized Pareto distribution using maximum likelihood and goodness of fit | |
| Al-Hussaini et al. | Accelerated life tests under finite mixture models | |
| Kang et al. | Robust output feedback H∞ control of uncertain Markovian jump systems with mode-dependent time-delays | |
| Yang et al. | Monitoring data factorization of high renewable energy penetrated grids for probabilistic static voltage stability assessment | |
| Malik et al. | Selection of most relevant input parameters using waikato environment for knowledge analysis for gene expression programming based power transformer fault diagnosis | |
| Shaheen et al. | Representations of solar photovoltaic triple-diode models using artificial hummingbird optimizer | |
| Besheer et al. | Single-diode model based photovoltaic module: analysis and comparison approach | |
| Çetintaş | Energy consumption and economic growth: The case of transition economies | |
| Tian | Chaotic characteristic analysis of short-term wind speed time series with different time scales | |
| Kuyumani et al. | A hybrid model based on CNN-LSTM to detect and forecast harmonics: a case study of an eskom substation in South Africa | |
| CN117556369B (en) | Power theft detection method and system for dynamically generated residual error graph convolution neural network | |
| Luo et al. | Detection and isolation of false data injection attack via adaptive Kalman filter bank | |
| Liu et al. | Wind power forecasting method based on bidirectional long short-term memory neural network and error correction | |
| Wadi et al. | Reliability and sensitivity analysis for closed-ring distribution power systems | |
| Datta et al. | S-transform based kurtosis analysis for detection of LG and LL faults in 14 Bus microgrid system | |
| CN109768549B (en) | Method for constructing thermal stability security domain of electric power system | |
| Mak et al. | Load Embeddings for Scalable {AC-OPF} Learning | |
| Sangeeth et al. | High impedance fault detection using multi-domain feature with artificial neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |