CN114358371A - Photovoltaic short-term power prediction method and device based on deep learning - Google Patents
Photovoltaic short-term power prediction method and device based on deep learning Download PDFInfo
- Publication number
- CN114358371A CN114358371A CN202111424158.8A CN202111424158A CN114358371A CN 114358371 A CN114358371 A CN 114358371A CN 202111424158 A CN202111424158 A CN 202111424158A CN 114358371 A CN114358371 A CN 114358371A
- Authority
- CN
- China
- Prior art keywords
- photovoltaic
- data
- prediction
- real
- short
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000013135 deep learning Methods 0.000 title claims abstract description 28
- 230000004927 fusion Effects 0.000 claims abstract description 27
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 21
- 230000002457 bidirectional effect Effects 0.000 claims abstract description 9
- 230000007246 mechanism Effects 0.000 claims abstract description 8
- 238000012544 monitoring process Methods 0.000 claims description 58
- 238000000605 extraction Methods 0.000 claims description 41
- 238000012549 training Methods 0.000 claims description 31
- 238000007781 pre-processing Methods 0.000 claims description 17
- 230000006870 function Effects 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 11
- 238000013528 artificial neural network Methods 0.000 claims description 10
- 230000002452 interceptive effect Effects 0.000 claims description 6
- 238000011176 pooling Methods 0.000 claims description 6
- 230000005540 biological transmission Effects 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000012512 characterization method Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000010248 power generation Methods 0.000 abstract description 6
- 239000000463 material Substances 0.000 abstract description 5
- 230000009467 reduction Effects 0.000 abstract description 5
- 230000006872 improvement Effects 0.000 abstract description 3
- 238000005457 optimization Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000000053 physical method Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005684 electric field Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000007526 fusion splicing Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a photovoltaic short-term power prediction method and device based on deep learning, which are characterized in that photovoltaic real-time data and historical power data are respectively input into a method for constructing a deep learning prediction model through a convolutional neural network, a BilStm network model and a bidirectional Attention mechanism of convolution kernels with different sizes, and finally, two obtained data text characteristics are combined to obtain more effective fusion characteristic representation capability, so that the optimal text characteristics can be used for accurately predicting the photovoltaic short-term power. By the method, the accuracy of photovoltaic short-term power prediction is improved, optimization is facilitated, and the operation cost of the photovoltaic power generation system is greatly reduced, including cost reduction and efficiency improvement of manpower, material resources, time efficiency and the like.
Description
Technical Field
The invention relates to the fields of artificial intelligence technology, deep learning, neural networks, natural language processing, new energy and photovoltaic electric fields, in particular to a photovoltaic short-term power prediction method and device based on deep learning, computer equipment and a storage medium.
Background
With the rapid development of deep learning, not only application scenes such as finance and insurance are rapidly fused and applied to the ground, but also a fusion scheme for a photovoltaic power generation system is accompanied. At present, the traditional machine learning method provides greater challenges for the accuracy of photovoltaic short-term power prediction, such as: the confusion of how to improve the accuracy of photovoltaic short-term power from two dimensions of photovoltaic real-time data and historical power data by applying a traditional machine learning method and a physical method provides decision basis for the reliability and robustness of photovoltaic power generation. The past photovoltaic power prediction method comprises the following steps: statistical methods, physical methods, machine learning methods and integration methods, but these methods have a number of defects that lead to inaccurate photovoltaic short-term power prediction, and these defects are expressed as: insufficient acquired data may cause a fitting phenomenon, or the network structure is simple, the network performance is poor, the feature characterization capability is insufficient or even limited, and the like. Therefore, if the problem of photovoltaic short-term power prediction is not solved under the condition that the current photovoltaic power generation ratio is good, many disadvantages are brought to a power supply system, including not only increasing various operation costs, indirectly increasing expenses such as manpower, and even forming safe operation of a power dispatching system.
Disclosure of Invention
The invention provides a photovoltaic short-term power prediction method and device based on deep learning, computer equipment and a storage medium, and aims to improve the accuracy of photovoltaic short-term power prediction and reduce the operation cost of a photovoltaic power generation system, including cost reduction and efficiency improvement of manpower, material resources, time efficiency and the like.
To this end, a first object of the present invention is to provide a deep learning photovoltaic short-term power prediction method, including:
acquiring photovoltaic real-time monitoring data and historical photovoltaic power data in a specified time interval, performing data preprocessing, and taking the preprocessed photovoltaic real-time monitoring data and the historical photovoltaic power data as training sets;
building a photovoltaic short-term power prediction network model, and training the built photovoltaic short-term power prediction through a training set; the photovoltaic short-term power prediction network model comprises a feature extraction module, a context information extraction module, a prediction information enhancement module, a feature fusion module and a result prediction module which are connected in sequence;
and preprocessing the photovoltaic real-time monitoring data and the historical photovoltaic power data which are generated in real time, inputting the preprocessed photovoltaic real-time monitoring data and the preprocessed historical photovoltaic power data into a trained photovoltaic short-term power prediction network model, and outputting a result as a prediction result of the photovoltaic power in a future specified time interval.
Wherein, the data preprocessing step comprises:
and (3) data format analysis: analyzing the data formats of the photovoltaic real-time monitoring data and/or the historical photovoltaic power data in different data formats, and converting the data formats into a uniform format;
analyzing data correlation to exclude photovoltaic real-time monitoring data and historical photovoltaic power data with weights lower than a set threshold;
normalization processing, namely normalizing the photovoltaic real-time monitoring data and the historical photovoltaic power data according to a formula (1);
where w' is a normalized value, w represents the true value of the sample, wminAnd wmaxRepresenting the minimum and maximum values selected.
The photovoltaic short-term power prediction network model comprises a feature extraction module, a context information extraction module, a prediction information enhancement module, a feature fusion module and a result prediction module which are connected in sequence; wherein,
the characteristic extraction module is a characteristic extraction neural network and is used for extracting characteristics of different scales of the photovoltaic real-time monitoring data and the historical photovoltaic power data;
the context information extraction module is used for acquiring the sequence relation of the photovoltaic real-time monitoring data and the historical photovoltaic power data on a time sequence to obtain the context information of the photovoltaic real-time monitoring data and the historical photovoltaic power data;
the prediction information enhancement module is used for acquiring the correlation weight characteristics of the photovoltaic real-time monitoring data and the historical photovoltaic power data;
the characteristic fusion module is used for fusion splicing of correlation weight characteristics of photovoltaic real-time monitoring data and historical photovoltaic power data to obtain characteristic fusion information;
and the result prediction module is used for calculating a prediction result according to the feature fusion information and completing photovoltaic short-term power prediction.
The characteristic extraction neural network is a convolutional neural network model CNN network; the convolutional neural network model CNN network comprises two convolutional layers and two pooling layers, and a maximum pooling method is adopted to obtain high-frequency space-time characteristics.
The context information extraction module adopts a BilSTM network model, inputs high-frequency spatio-temporal characteristics into the BilSTM network model, outputs context relations of data at different moments, and improves text characteristic representation and fitting capacity.
The prediction information enhancement module obtains different weights in the characteristics and different associated weight information between the photovoltaic real-time monitoring data and the historical photovoltaic power data by utilizing a self-attention and interactive attention mechanism in the bidirectional attention to the photovoltaic real-time monitoring data and the historical photovoltaic power data.
The method comprises the following steps of constructing a photovoltaic short-term power prediction network model by using a training set, wherein the steps of training the constructed photovoltaic short-term power prediction network model by using the training set comprise:
inputting the preprocessed training set data into a feature extraction neural network of a feature extraction module, and performing feature extraction through a convolutional neural network model (CNN) network;
obtaining a sequence relation of a past moment and a future moment by using a BilSTM network model so as to obtain context information;
different weights in text features and different associated weight information between the historical data and the real-time data are obtained by utilizing a self-attention and interactive attention mechanism in bidirectional attention on the photovoltaic historical data and the real-time data respectively, so that the correlation in the features of the two data and the associated characteristics between the two data are mined, and the two data respectively have different scales, context information and correlation and associated characteristics;
combining the features with different scales, context information and correlation and relevance to obtain a fusion feature, wherein the fusion feature contains different sizes, context information, correlation and relevance and fully embodies the contribution of real-time data and historical data to photovoltaic short-term power prediction;
and calculating a prediction result by using the full connection layer, comparing the prediction result with a marked detection result, and finishing network training by continuously adjusting network functions and parameters until the prediction result is consistent with an actual power result.
Wherein, after the step of outputting the predicted result of the photovoltaic power in the future designated time interval, the method further comprises the step of displaying the result; wherein, the result display mode at least comprises: the method comprises the following steps of character display, voice broadcast, terminal outbound, mail and short message transmission and intelligent sound box voice awakening.
The second objective of the present invention is to provide a photovoltaic short-term power prediction apparatus based on deep learning, which includes:
the data acquisition module is used for acquiring photovoltaic real-time monitoring data and historical photovoltaic power data in a specified time interval, carrying out data preprocessing, and taking the preprocessed photovoltaic real-time monitoring data and the historical photovoltaic power data as a training set;
the network construction module is used for constructing a photovoltaic short-term power prediction network model and training the constructed photovoltaic short-term power prediction through a training set; the photovoltaic short-term power prediction network model comprises a feature extraction module, a context information extraction module, a prediction information enhancement module, a feature fusion module and a result prediction module which are connected in sequence;
and the power prediction module is used for preprocessing the photovoltaic real-time monitoring data and the historical photovoltaic power data which are generated in real time, inputting the preprocessed photovoltaic real-time monitoring data and the historical photovoltaic power data into the trained photovoltaic short-term power prediction network model, and outputting a result as a prediction result of the photovoltaic power in a future specified time interval.
A third object of the present invention is to provide a computer device, which includes a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the method according to the foregoing technical solution.
A fourth object of the invention is to propose a non-transitory computer-readable storage medium on which a computer program is stored, which computer program, when executed by a processor, implements the method of the aforementioned technical solution.
Different from the prior art, the photovoltaic short-term power prediction method based on deep learning provided by the invention is characterized in that the photovoltaic real-time data and the historical power data are respectively input into a deep learning prediction model method constructed by a convolutional neural network, a BilStm network model and a bidirectional Attention mechanism of convolution kernels with different sizes, and finally, two obtained data text characteristics are subjected to merging operation to obtain more effective fusion characteristic characterization capability, so that the optimal text characteristics can be used for accurately predicting the photovoltaic short-term power. By the method, the accuracy of photovoltaic short-term power prediction is improved, optimization is facilitated, and the operation cost of the photovoltaic power generation system is greatly reduced, including cost reduction and efficiency improvement of manpower, material resources, time efficiency and the like.
Drawings
The foregoing and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a schematic flow chart of a photovoltaic short-term power prediction method based on deep learning provided by the invention.
Fig. 2 is a schematic structural diagram of a photovoltaic short-term power prediction network model of the photovoltaic short-term power prediction method based on deep learning provided by the invention.
Fig. 3 is a schematic structural diagram of a photovoltaic short-term power prediction device based on deep learning according to the present invention.
Fig. 4 is a schematic structural diagram of a non-transitory computer-readable storage medium according to the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
Fig. 1 is a schematic flowchart of a photovoltaic short-term power prediction method based on deep learning according to an embodiment of the present invention. The method comprises the following steps:
The invention constructs a deep learning prediction model with multi-scale convolution kernel CNN + BiLSTM + bidirectional Attention, aiming at the problems of inaccurate photovoltaic short-term power prediction caused by insufficient characteristics of extracted photovoltaic real-time monitoring data and historical photovoltaic power data.
Firstly, acquiring data from a database, and extracting photovoltaic real-time monitoring data and historical photovoltaic power data within a specified time interval; in the invention, the time interval of the photovoltaic real-time monitoring data can be 24 hours, and the historical photovoltaic power data is data within one year from the time limit of the photovoltaic real-time monitoring data.
After the data acquisition is completed, the method further comprises a step of data preprocessing, wherein the data preprocessing is shown as 101 in fig. 2:
and (3) data format analysis: analyzing the data formats of the photovoltaic real-time monitoring data and/or the historical photovoltaic power data in different data formats, and converting the data formats into a uniform format;
analyzing data correlation to exclude photovoltaic real-time monitoring data and historical photovoltaic power data with weights lower than a set threshold;
normalization processing, namely normalizing the photovoltaic real-time monitoring data and the historical photovoltaic power data according to a formula (1);
where w' is a normalized value, w represents the true value of the sample, wminAnd wmaxRepresenting the minimum and maximum values selected.
After the data preprocessing is completed, the process proceeds to step 102.
Step 102: building a photovoltaic short-term power prediction network model, and training the built photovoltaic short-term power prediction through a training set; as shown in fig. 2, the photovoltaic short-term power prediction network model includes a feature extraction module 102, a context information extraction module 103, a prediction information enhancement module 104, a feature fusion module 105, and a result prediction module 106, which are connected in sequence.
The feature extraction module 102 is a feature extraction neural network, and is configured to perform feature extraction on the photovoltaic real-time monitoring data and the historical photovoltaic power data at different scales.
The characteristic extraction neural network is a convolutional neural network model CNN network; the convolutional neural network model CNN network comprises two convolutional layers and two pooling layers, and a maximum pooling method is adopted to obtain high-frequency space-time characteristics, so that unified reduction and compression of data are facilitated, and generalization capability is improved.
The context information extraction module 103 is configured to obtain a sequence relationship between the photovoltaic real-time monitoring data and the historical photovoltaic power data in a time sequence, and obtain context information of the photovoltaic real-time monitoring data and the historical photovoltaic power data.
The context information extraction module 103 adopts a BilSTM network model, inputs high-frequency spatio-temporal characteristics into the BilSTM network model, outputs context relationships of data at different moments, and improves text characteristic representation and fitting capability.
The prediction information enhancement module 104 is used for acquiring the correlation weight characteristics of the photovoltaic real-time monitoring data and the historical photovoltaic power data; the prediction information enhancement module obtains different weights in the characteristics and different associated weight information between the photovoltaic real-time monitoring data and the historical photovoltaic power data by utilizing a self-attention and interactive attention mechanism in the bidirectional attention to the photovoltaic real-time monitoring data and the historical photovoltaic power data. By mining the correlation inside the respective characteristics of the two kinds of data and the correlation characteristics between the two kinds of data, the two kinds of data are respectively formed with different sizes, context information and correlation characteristics.
The feature fusion module 105 is configured to fuse and splice associated weight features of the photovoltaic real-time monitoring data and the historical photovoltaic power data to obtain feature fusion information. The fusion characteristics contain different sizes, context information, correlation and relevance, and fully reflect the contribution degree of real-time data and historical data to photovoltaic short-term power prediction.
And the result prediction module 106 is used for calculating a prediction result according to the feature fusion information and completing photovoltaic short-term power prediction.
Calculating a photovoltaic short-term power predicted value through a full connection layer, adopting an activation function ReLU function as a Dense activation function, and calculating to obtain a predicted result and calculating by using a normalized reduction function to obtain the original size of the predicted result; then, the prediction result is displayed, and the average absolute error MAE and the root mean square error RMSE are selected.
wo＝wpre(wmax-wmin)+wmin (2)
Wherein wprePredicting an output value, W, for a network modeloRepresenting the restored power prediction.
The step of training the constructed photovoltaic short-term power prediction network model through the training set comprises the following steps:
inputting the preprocessed training set data into a feature extraction neural network of a feature extraction module, and performing feature extraction through a convolutional neural network model (CNN) network;
obtaining a sequence relation of a past moment and a future moment by using a BilSTM network model so as to obtain context information;
different weights in text features and different associated weight information between the historical data and the real-time data are obtained by utilizing a self-attention and interactive attention mechanism in bidirectional attention on the photovoltaic historical data and the real-time data respectively, so that the correlation in the features of the two data and the associated characteristics between the two data are mined, and the two data respectively have different scales, context information and correlation and associated characteristics;
combining the features with different scales, context information and correlation and relevance to obtain a fusion feature, wherein the fusion feature contains different sizes, context information, correlation and relevance and fully embodies the contribution of real-time data and historical data to photovoltaic short-term power prediction;
and calculating a prediction result by using the full connection layer, comparing the prediction result with a marked detection result, and finishing network training by continuously adjusting network functions and parameters until the prediction result is consistent with an actual power result.
S103: and preprocessing the photovoltaic real-time monitoring data and the historical photovoltaic power data which are generated in real time, inputting the preprocessed photovoltaic real-time monitoring data and the preprocessed historical photovoltaic power data into a trained photovoltaic short-term power prediction network model, and outputting a result as a prediction result of the photovoltaic power in a future specified time interval.
The mean absolute error MAE and the root mean square error RMSE of the system are reduced by historical power data within 1 year and monitoring data acquired in real time within 24 hours, and the method has important significance for improving a photovoltaic short-term power prediction model and scheduling operation and photovoltaic proportion of a power system.
After the step of outputting the predicted result of the photovoltaic power within the future specified time interval, the method further comprises the step of displaying the result; the result is shown in fig. 2 as 107, and at least includes: the method comprises the following steps of character display, voice broadcast, terminal outbound, mail and short message transmission and intelligent sound box voice awakening.
In order to implement the foregoing embodiment, the present invention further provides a photovoltaic short-term power prediction apparatus based on deep learning, including:
the data acquisition module 310 is configured to acquire photovoltaic real-time monitoring data and historical photovoltaic power data within a specified time interval, perform data preprocessing, and use the preprocessed photovoltaic real-time monitoring data and the historical photovoltaic power data as a training set;
the network construction module 320 is used for constructing a photovoltaic short-term power prediction network model and training the constructed photovoltaic short-term power prediction through a training set; the photovoltaic short-term power prediction network model comprises a feature extraction module, a context information extraction module, a prediction information enhancement module, a feature fusion module and a result prediction module which are connected in sequence;
and the power prediction module 330 is configured to input the photovoltaic real-time monitoring data and the historical photovoltaic power data generated in real time into the trained photovoltaic short-term power prediction network model after preprocessing, and output a result as a prediction result of the photovoltaic power within a future specified time interval.
After the prediction is completed, the method further includes result presentation, as shown in fig. 3, where the manner of result presentation at least includes: the method comprises the following steps of character display, voice broadcast, terminal outbound, mail and short message transmission and intelligent sound box voice awakening.
In order to implement the above embodiment, the present invention further provides another computer device, including: the system comprises a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to realize photovoltaic short-term power prediction according to the embodiment of the invention.
As shown in fig. 4, the non-transitory computer readable storage medium includes a memory 810 of instructions executable by a processor 820 of a deep learning based photovoltaic short term power prediction device to perform the above-described method, an interface 830. Alternatively, the storage medium may be a non-transitory computer readable storage medium, for example, the non-transitory computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.
In order to achieve the above embodiments, the present invention also proposes a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, enables a photovoltaic short term power prediction as an embodiment of the present invention.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (11)
1. A photovoltaic short-term power prediction method based on deep learning is characterized by comprising the following steps:
acquiring photovoltaic real-time monitoring data and historical photovoltaic power data in a specified time interval, performing data preprocessing, and taking the preprocessed photovoltaic real-time monitoring data and the historical photovoltaic power data as training sets;
building a photovoltaic short-term power prediction network model, and training the built photovoltaic short-term power prediction through the training set; the photovoltaic short-term power prediction network model comprises a feature extraction module, a context information extraction module, a prediction information enhancement module, a feature fusion module and a result prediction module which are connected in sequence;
and preprocessing the photovoltaic real-time monitoring data and the historical photovoltaic power data which are generated in real time, inputting the preprocessed photovoltaic real-time monitoring data and the historical photovoltaic power data into the trained photovoltaic short-term power prediction network model, and outputting a result as a prediction result of the photovoltaic power in a future specified time interval.
2. The deep learning-based photovoltaic short-term power prediction method according to claim 1, wherein the step of data preprocessing comprises:
and (3) data format analysis: analyzing the data formats of the photovoltaic real-time monitoring data and/or the historical photovoltaic power data in different data formats, and converting the data formats into a uniform format;
analyzing data correlation to exclude photovoltaic real-time monitoring data and historical photovoltaic power data with weights lower than a set threshold;
normalization processing, namely normalizing the photovoltaic real-time monitoring data and the historical photovoltaic power data according to a formula (1);
where w' is a normalized value, w represents the true value of the sample, wminAnd wmaxRepresenting the minimum and maximum values selected.
3. The deep learning-based photovoltaic short-term power prediction method according to claim 2, wherein the photovoltaic short-term power prediction network model comprises a feature extraction module, a context information extraction module, a prediction information enhancement module, a feature fusion module and a result prediction module which are connected in sequence; wherein,
the characteristic extraction module is a characteristic extraction neural network and is used for extracting characteristics of different scales of the photovoltaic real-time monitoring data and the historical photovoltaic power data;
the context information extraction module is used for acquiring a sequence relation of the photovoltaic real-time monitoring data and the historical photovoltaic power data on a time sequence to obtain context information of the photovoltaic real-time monitoring data and the historical photovoltaic power data;
the prediction information enhancement module is used for acquiring the correlation weight characteristics of the photovoltaic real-time monitoring data and the historical photovoltaic power data;
the characteristic fusion module is used for fusing and splicing the correlation weight characteristics of the photovoltaic real-time monitoring data and the historical photovoltaic power data to obtain characteristic fusion information;
and the result prediction module is used for calculating a prediction result according to the feature fusion information and completing photovoltaic short-term power prediction.
4. The deep learning-based photovoltaic short-term power prediction method according to claim 3, wherein the feature extraction neural network is a convolutional neural network model (CNN) network; the convolutional neural network model CNN network comprises two convolutional layers and two pooling layers, and a maximum pooling method is adopted to obtain high-frequency space-time characteristics.
5. The photovoltaic short-term power prediction method based on deep learning of claim 4, wherein the context information extraction module adopts a BilSTM network model, inputs the high-frequency spatiotemporal features into the BilSTM network model, outputs context relationships of data at different times, and improves text feature characterization and fitting capabilities.
6. The deep learning-based photovoltaic short-term power prediction method as claimed in claim 3, wherein the prediction information enhancement module obtains different weights in the features and different weight information of the association between the photovoltaic real-time monitoring data and the historical photovoltaic power data by using a self-attention and interactive attention mechanism in bidirectional attention on the photovoltaic real-time monitoring data and the historical photovoltaic power data.
7. The deep learning-based photovoltaic short-term power prediction method according to claim 4, wherein the step of training the constructed photovoltaic short-term power prediction network model through the training set comprises:
inputting the preprocessed training set data into a feature extraction neural network of a feature extraction module, and performing feature extraction through a convolutional neural network model (CNN) network;
obtaining a sequence relation of a past moment and a future moment by using a BilSTM network model so as to obtain context information;
different weights in text features and different associated weight information between the historical data and the real-time data are obtained by utilizing a self-attention and interactive attention mechanism in bidirectional attention on the photovoltaic historical data and the real-time data respectively, so that the correlation in the features of the two data and the associated characteristics between the two data are mined, and the two data respectively have different scales, context information and correlation and associated characteristics;
merging the features with different scales, context information and correlation and relevance to obtain a fusion feature, wherein the fusion feature contains different sizes, context information, correlation and relevance and fully embodies the contribution of real-time data and historical data to photovoltaic short-term power prediction;
and calculating a prediction result by using the full connection layer, comparing the prediction result with a marked detection result, and finishing network training by continuously adjusting network functions and parameters until the prediction result is consistent with an actual power result.
8. The deep learning based photovoltaic short-term power prediction method according to claim 4, characterized in that after the step of outputting the prediction result of photovoltaic power within a specified time interval in the future, a step of result presentation is further included; wherein, the result display mode at least comprises: the method comprises the following steps of character display, voice broadcast, terminal outbound, mail and short message transmission and intelligent sound box voice awakening.
9. A photovoltaic short-term power prediction device based on deep learning is characterized by comprising:
the data acquisition module is used for acquiring photovoltaic real-time monitoring data and historical photovoltaic power data in a specified time interval, carrying out data preprocessing, and taking the preprocessed photovoltaic real-time monitoring data and the historical photovoltaic power data as a training set;
the network construction module is used for constructing a photovoltaic short-term power prediction network model and training the constructed photovoltaic short-term power prediction through the training set; the photovoltaic short-term power prediction network model comprises a feature extraction module, a context information extraction module, a prediction information enhancement module, a feature fusion module and a result prediction module which are connected in sequence;
and the power prediction module is used for preprocessing the photovoltaic real-time monitoring data and the historical photovoltaic power data which are generated in real time, inputting the preprocessed photovoltaic real-time monitoring data and the historical photovoltaic power data into the trained photovoltaic short-term power prediction network model, and outputting a result as a prediction result of the photovoltaic power in a future specified time interval.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of any one of claims 1-8 when executing the computer program.
11. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program, when executed by a processor, implements the method of any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111424158.8A CN114358371A (en) | 2021-11-26 | 2021-11-26 | Photovoltaic short-term power prediction method and device based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111424158.8A CN114358371A (en) | 2021-11-26 | 2021-11-26 | Photovoltaic short-term power prediction method and device based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114358371A true CN114358371A (en) | 2022-04-15 |
Family
ID=81096592
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111424158.8A Pending CN114358371A (en) | 2021-11-26 | 2021-11-26 | Photovoltaic short-term power prediction method and device based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114358371A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115630726A (en) * | 2022-09-01 | 2023-01-20 | 华能江苏综合能源服务有限公司 | Roof photovoltaic power prediction method based on VMD-BILSTM neural network fusion attention mechanism |
| WO2023093010A1 (en) * | 2021-11-26 | 2023-06-01 | 中国华能集团清洁能源技术研究院有限公司 | Wind power prediction method and device based on deep learning fusion model |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110909919A (en) * | 2019-11-07 | 2020-03-24 | 哈尔滨工程大学 | Photovoltaic power prediction method of depth neural network model with attention mechanism fused |
| CN111341386A (en) * | 2020-02-17 | 2020-06-26 | 大连理工大学 | Attention-introducing multi-scale CNN-BilSTM non-coding RNA interaction relation prediction method |
| CN112784960A (en) * | 2021-01-21 | 2021-05-11 | 中国电建集团西北勘测设计研究院有限公司 | Photovoltaic power generation power prediction method based on sliding window and local time sequence characteristics |
| CN113128793A (en) * | 2021-05-19 | 2021-07-16 | 中国南方电网有限责任公司 | Photovoltaic power combination prediction method and system based on multi-source data fusion |
| CN113205226A (en) * | 2021-05-28 | 2021-08-03 | 河北工业大学 | Photovoltaic power prediction method combining attention mechanism and error correction |
| CN113627070A (en) * | 2021-05-24 | 2021-11-09 | 国网新疆电力有限公司经济技术研究院 | Short-term photovoltaic power prediction method |
| CN113627674A (en) * | 2021-08-12 | 2021-11-09 | 中国华能集团清洁能源技术研究院有限公司 | Distributed photovoltaic power station output prediction method and device and storage medium |
-
2021
- 2021-11-26 CN CN202111424158.8A patent/CN114358371A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110909919A (en) * | 2019-11-07 | 2020-03-24 | 哈尔滨工程大学 | Photovoltaic power prediction method of depth neural network model with attention mechanism fused |
| CN111341386A (en) * | 2020-02-17 | 2020-06-26 | 大连理工大学 | Attention-introducing multi-scale CNN-BilSTM non-coding RNA interaction relation prediction method |
| CN112784960A (en) * | 2021-01-21 | 2021-05-11 | 中国电建集团西北勘测设计研究院有限公司 | Photovoltaic power generation power prediction method based on sliding window and local time sequence characteristics |
| CN113128793A (en) * | 2021-05-19 | 2021-07-16 | 中国南方电网有限责任公司 | Photovoltaic power combination prediction method and system based on multi-source data fusion |
| CN113627070A (en) * | 2021-05-24 | 2021-11-09 | 国网新疆电力有限公司经济技术研究院 | Short-term photovoltaic power prediction method |
| CN113205226A (en) * | 2021-05-28 | 2021-08-03 | 河北工业大学 | Photovoltaic power prediction method combining attention mechanism and error correction |
| CN113627674A (en) * | 2021-08-12 | 2021-11-09 | 中国华能集团清洁能源技术研究院有限公司 | Distributed photovoltaic power station output prediction method and device and storage medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023093010A1 (en) * | 2021-11-26 | 2023-06-01 | 中国华能集团清洁能源技术研究院有限公司 | Wind power prediction method and device based on deep learning fusion model |
| CN115630726A (en) * | 2022-09-01 | 2023-01-20 | 华能江苏综合能源服务有限公司 | Roof photovoltaic power prediction method based on VMD-BILSTM neural network fusion attention mechanism |
| CN115630726B (en) * | 2022-09-01 | 2024-01-30 | 华能江苏综合能源服务有限公司 | Roof photovoltaic power prediction method based on VMD-BILSTM neural network fused attention mechanism |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114330495A (en) | Wind power prediction method and device based on deep learning fusion model | |
| WO2023087525A1 (en) | Multi-modal deep learning power generation device anomaly integrated identification method and device | |
| WO2023093774A1 (en) | Deep learning-based wind power cluster power prediction method | |
| CN109740657B (en) | Training method and device of neural network model for image data classification | |
| EP4053751A1 (en) | Method and apparatus for training cross-modal retrieval model, device and storage medium | |
| CN114358371A (en) | Photovoltaic short-term power prediction method and device based on deep learning | |
| WO2022188644A1 (en) | Word weight generation method and apparatus, and device and medium | |
| CN114663383B (en) | Blood cell segmentation and identification method and device, electronic equipment and storage medium | |
| CN116596916B (en) | Training of defect detection model and defect detection method and device | |
| CN113252701B (en) | Cloud edge cooperation-based power transmission line insulator self-explosion defect detection system and method | |
| CN113658122A (en) | Image quality evaluation method, device, storage medium and electronic equipment | |
| CN110046279A (en) | Prediction technique, medium, device and the calculating equipment of video file feature | |
| CN115861462A (en) | Training method and device for image generation model, electronic equipment and storage medium | |
| CN112200772A (en) | Pox check out test set | |
| CN114372532B (en) | Method, device, equipment, medium and product for determining label labeling quality | |
| CN115775349A (en) | False news detection method and device based on multi-mode fusion | |
| CN118297911A (en) | Insulator defect fault detection method and device, storage medium and computer equipment | |
| CN118053027A (en) | Defect identification method and device, electronic equipment and storage medium | |
| CN113743355A (en) | Method, device and system for checking state of switch device and computer equipment | |
| CN117038099A (en) | Medical term standardization method and device | |
| CN115457329B (en) | Training method of image classification model, image classification method and device | |
| CN114863450B (en) | Image processing method, device, electronic equipment and storage medium | |
| CN114462526B (en) | Classification model training method and device, computer equipment and storage medium | |
| CN115908969A (en) | Method and apparatus for image processing and model training | |
| CN114330239A (en) | Text processing method and device, storage medium and electronic equipment |
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 |