CN117748482B - Method and device for predicting generated power, electronic equipment and storage medium - Google Patents
Method and device for predicting generated power, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN117748482B CN117748482B CN202311714781.6A CN202311714781A CN117748482B CN 117748482 B CN117748482 B CN 117748482B CN 202311714781 A CN202311714781 A CN 202311714781A CN 117748482 B CN117748482 B CN 117748482B
- Authority
- CN
- China
- Prior art keywords
- power generation
- equipment
- power
- historical
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000010248 power generation Methods 0.000 claims abstract description 433
- 238000012549 training Methods 0.000 claims description 103
- 230000006870 function Effects 0.000 claims description 21
- 238000012360 testing method Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 11
- 238000012217 deletion Methods 0.000 claims description 4
- 230000037430 deletion Effects 0.000 claims description 4
- 238000012545 processing Methods 0.000 description 8
- 230000005855 radiation Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 230000007613 environmental effect Effects 0.000 description 4
- 238000000354 decomposition reaction Methods 0.000 description 3
- 238000001556 precipitation Methods 0.000 description 3
- 230000007774 longterm Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000000630 rising effect Effects 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 230000022131 cell cycle Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000009469 supplementation Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The application provides a method and a device for predicting generated power, electronic equipment and a storage medium, wherein the method comprises the following steps: acquiring historical equipment power generation data of at least two groups of target power generation equipment under different historical moment information and different historical meteorological data, wherein each group of historical equipment power generation data corresponds to the historical moment information and the historical meteorological data; according to the power generation data of each group of historical equipment, the historical meteorological data corresponding to the power generation data of each group of historical equipment and the corresponding historical moment information, trend characteristic information of the power generation power change of the target power generation equipment is determined; and predicting the future power generation power of the target power generation equipment according to the trend characteristic information. According to the power generation power prediction method, the power generation data of the historical equipment under different time and different meteorological conditions can be obtained, and the power generation power of the power generation equipment can be effectively predicted by analyzing the trend characteristic information of the power generation of the historical equipment under different time and different meteorological conditions.
Description
Technical Field
The application relates to the technical field of power generation equipment, in particular to a power generation power prediction method and device, electronic equipment and a storage medium.
Background
With the global increasing demand for renewable energy sources, for example, the demand for electrical energy increases. Photovoltaic power generation, however, is increasingly important as a clean, renewable energy source in global energy structures. However, the output power of photovoltaic power generation systems is affected by solar irradiance and other environmental factors, and there are significant fluctuations, randomness, and intermittence. These characteristics cause the power output of photovoltaic power generation to dynamically change with time, and this change may break the balance of supply and demand of the power system, and even threaten the stable operation of the whole system. At the same time, this uncertainty also brings economic losses to the operators of the photovoltaic power generation systems. Thus, accurate photovoltaic power predictions are critical to the stable operation of the grid and the efficient operation of the power market. However, the current prediction result of the generated power deviates more than the actual situation.
Disclosure of Invention
The application provides a method for predicting generated power, which can effectively predict the generated power of power generation equipment.
In a first aspect, the present application provides a method for predicting generated power, applied to a power prediction model for completing training, the method comprising:
acquiring historical equipment power generation data of at least two groups of target power generation equipment under different historical moment information and different historical meteorological data, wherein each group of historical equipment power generation data corresponds to the historical moment information and the historical meteorological data;
According to each set of historical equipment power generation data, historical meteorological data corresponding to each set of historical equipment power generation data and corresponding historical time information, trend feature information of the change of the power generation power of the target power generation equipment is determined;
and predicting the future power generation power of the target power generation equipment according to the trend characteristic information.
In some embodiments of the present application, the predicting the future generation power of the target generation device according to the trend feature information includes:
According to each group of the historical equipment power generation data, the historical meteorological data corresponding to each group of the historical equipment power generation data and the corresponding historical time information, determining fluctuation characteristic information of the power generation change of the target power generation equipment;
And predicting future power generation of the target power generation equipment according to the trend characteristic information and the fluctuation characteristic information.
In some embodiments of the present application, the predicting the future generation power of the target generation device according to the trend feature information includes:
And predicting the power generation of the target power generation equipment in a future target time period according to the trend characteristic information, wherein the time difference between the minimum time information and the current time information in the target time period is smaller than a preset time difference threshold value, and the target time period is smaller than a preset time period threshold value.
In some embodiments of the present application, before the obtaining the historical device power generation data under the different historical time information and the different historical meteorological data, the method further includes:
acquiring a plurality of groups of equipment power generation sample data for training the power prediction model, wherein each group of equipment power generation sample data corresponds to time information and meteorological sample data;
dividing each group of the equipment power generation sample data into training set samples and test set samples according to a set proportion;
inputting the power generation sample data of each group of equipment in the training set sample as a current training sample into a power prediction model to be trained to perform training, and obtaining predicted power of a second historical moment after a first historical moment corresponding to the current training sample, wherein the second historical moment corresponds to at least one target training sample except the current training sample in the training set sample;
Calculating a loss function value according to the predicted power of the predicted second historical moment and the power parameter of the equipment power generation sample data set corresponding to the second historical moment in the target training sample;
Iteratively updating the model index of the power prediction model according to the loss function value until the loss function value meets the iteration termination condition, so as to obtain a trained power prediction model;
And testing the trained power prediction model through the equipment power generation sample data set in the test set sample, and completing the training of the power prediction model.
In some embodiments of the present application, the acquiring a plurality of sets of device power generation sample data for the power prediction model training includes:
acquiring a plurality of groups of initial equipment power generation sample data for training the power prediction model;
Respectively taking each group of initial equipment power generation sample data as a target initial power generation sample data group, and determining missing parameters in the target initial power generation sample data group;
if the target initial power generation sample data set has no parameter missing, determining that the target initial power generation sample data set is a set of equipment power generation sample data;
And if the target initial power generation sample data set has parameter deletion, filling the target parameter deleted in the target initial power generation sample data set to obtain a group of equipment power generation sample data.
In some embodiments of the application, the method further comprises:
if the target initial power generation sample data set has the parameter missing, determining the number of the missing parameters in the target initial power generation sample data set;
And deleting the target initial power generation sample data set if the number of the missing parameters in the target initial power generation sample data set is larger than a preset number threshold.
In some embodiments of the present application, the inputting the power generation sample data of each group of devices in the training set sample as the current training sample into the power prediction model to be trained to obtain the predicted power at the second historical time after the first historical time corresponding to the current training sample includes:
Respectively taking each group of equipment power generation sample data in the training set sample as a target equipment power generation sample data set, and determining average power generation sample data of the target equipment power generation sample data set according to the target equipment power generation sample data set and equipment power generation sample data of time information before target time information of the target equipment power generation sample data set;
determining average power generation difference sample data of the target equipment power generation sample data set according to the target equipment power generation sample data set and equipment power generation sample data of time information before target time information of the target equipment power generation sample data set;
According to average power generation sample data and average power generation difference sample data of the target equipment power generation sample data set, trend characteristic information and fluctuation characteristic information between the first historical moment and the second historical moment of the target equipment power generation sample data set are determined;
And determining the predicted power of the second historical moment after the first historical moment corresponding to the current training sample according to the trend characteristic information and the fluctuation characteristic information between the first historical moment and the second historical moment.
In a second aspect, the present application also provides a generated power prediction apparatus applied to a power prediction model for completing training, the apparatus comprising:
the acquisition module is used for acquiring historical equipment power generation data of at least two groups of target power generation equipment under different historical moment information and different historical meteorological data, wherein each group of historical equipment power generation data corresponds to the historical moment information and the historical meteorological data;
The determining module is used for determining trend characteristic information of the change of the power generation power of the target power generation equipment according to each group of the power generation data of the historical equipment, the historical meteorological data corresponding to each group of the power generation data of the historical equipment and the corresponding historical time information;
and the prediction module is used for predicting the future power generation power of the target power generation equipment according to the trend characteristic information.
In a third aspect, the present application also provides an electronic device comprising a processor, a memory and a computer program stored in the memory and executable on the processor, the processor executing the computer program to implement the steps in any of the generated power prediction methods.
In a fourth aspect, the present application also provides a computer-readable storage medium having stored thereon a computer program that is executed by a processor to implement the steps in the generated power prediction method of any one of the above.
According to the power generation power prediction method provided by the application, the historical equipment power generation data of the power generation equipment under different time and different meteorological conditions can be obtained, and the power generation trend characteristic information of the power generation equipment under different time and different meteorological conditions can be analyzed, so that the regularity of the power generation power of the power generation equipment can be judged, and the power generation power of the power generation equipment can be further effectively predicted.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic view of a scenario of a generated power prediction system provided in an embodiment of the present application;
FIG. 2 is a flow chart of an embodiment of a method for predicting generated power in accordance with an embodiment of the present application;
FIG. 3 is a schematic diagram of the architecture of a power prediction model in one embodiment of a method for predicting generated power in an embodiment of the present application;
FIG. 4 is a schematic diagram of the architecture of a power prediction model in one embodiment of a method for predicting generated power in an embodiment of the present application;
FIG. 5 is a schematic diagram of a functional block of a power prediction apparatus according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to fall within the scope of the application.
In the description of the present application, it should be understood that the terms "first," "second," and the like are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. In the description of the present application, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
In the present application, the term "exemplary" is used to mean "serving as an example, instance, or illustration. Any embodiment described as "exemplary" in this disclosure is not necessarily to be construed as preferred or advantageous over other embodiments. Meanwhile, it can be understood that, in the specific embodiment of the present application, related data such as user information and user data are related, when the above embodiment of the present application is applied to specific products or technologies, user permission or consent needs to be obtained, and the collection, use and processing of related data need to comply with related laws and regulations and standards of related countries and regions.
The following description is presented to enable any person skilled in the art to make and use the application. In the following description, details are set forth for purposes of explanation. It will be apparent to one of ordinary skill in the art that the present application may be practiced without these specific details. In other instances, well-known structures and processes have not been described in detail so as not to obscure the description of the application with unnecessary detail. Thus, the present application is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.
The application provides a method, a device, equipment and a storage medium for predicting generated power, which are respectively described in detail below.
Referring to fig. 1, fig. 1 is a schematic diagram of a scenario of a power generation prediction system according to an embodiment of the application, where the power generation prediction system may include an electronic device 100 and a storage device 200, and the storage device 200 may transmit data to the electronic device 100. The electronic device 100 as in fig. 1 may be equipped with a power prediction model that completes training, and the power prediction model that completes training may acquire historical device power generation data and historical heat dissipation parameters of the power generation device stored in the storage device 200, so as to execute the power generation power prediction method in the present application, and implement power prediction for operation of the power generation device at a future time.
In the embodiment of the present application, the electronic device 100 includes, but is not limited to, a desktop computer, a portable computer, a network server, a Personal computer (Personal DIGITAL ASSISTANT, PDA), a tablet computer, a wireless electronic device, an embedded device, an independent server, or a server network or a server cluster formed by servers.
In embodiments of the present application, communication between electronic device 100 and storage device 200 may be accomplished by any means of communication, including, but not limited to, mobile communication based on the third generation partnership project (3rd Generation Partnership Project,3GPP), long term evolution (Long Term Evolution, LTE), worldwide interoperability for microwave access (Worldwide Interoperability for Microwave Access, wiMAX), or computer network communication based on the TCP/IP protocol suite (TCP/IP Protocol Suite, TCP/IP), user datagram protocol (User Datagram Protocol, UDP), and the like.
It should be noted that, the schematic view of the scenario of the power generation prediction system shown in fig. 1 is only an example, and the power generation prediction system and scenario described in the embodiments of the present application are for more clearly describing the technical solution of the embodiments of the present application, and do not constitute a limitation to the technical solution provided by the embodiments of the present application, and those skilled in the art can know that, with the evolution of the power generation prediction system and the appearance of a new service scenario, the technical solution provided by the embodiments of the present application is equally applicable to similar technical problems.
As shown in fig. 2, fig. 2 is a flow chart of a method for predicting generated power according to an embodiment of the present application, where the method for predicting generated power may be applied to a power prediction model for training, and specifically includes the following steps 201 to 203.
201. And acquiring historical equipment power generation data of at least two groups of target power generation equipment under different historical moment information and different historical meteorological data, wherein each group of historical equipment power generation data corresponds to the historical moment information and the historical meteorological data.
In the embodiment of the present application, the target power generation device may be a power generation device to be subjected to power prediction, including a photovoltaic power generation unit or a wind power generation unit, and the embodiment of the present application is not limited. The set of historical device power generation data may include historical power generation of the target power generation device. As an example, the manner of acquiring the historical device power generation data of each group of the target power generation device may include: when the target power generation equipment is in an operating state, the power generation power under certain historical moment information and meteorological data at the moment are recorded. Meanwhile, the operation parameters having the same historical time information may be stored in the same storage location, for example, the historical equipment power generation data and the historical weather data having the same historical time may be stored in the same folder. When the historical equipment power generation data is required to be acquired, the historical equipment power generation data and the meteorological data with the same historical moment are taken as a data set, and a set of historical equipment power generation data is obtained. Wherein, it should be noted that the historical meteorological data may include: solar radiation, temperature, humidity, precipitation, wind speed, etc.
For example: and the power generation power of the target power generation equipment is A when XX is divided into XX in XX year, XX month and XX day. And meanwhile, according to the sensors of all meteorological data, the meteorological data such as solar radiation, temperature, humidity, precipitation, wind speed and the like of XX minutes in XX year, XX month, XX day, XX minute can be obtained and stored in the same position. And if the historical equipment power generation data is required to be acquired later, acquiring the historical equipment power generation data from the corresponding storage position. Thus, historical equipment power generation data comprising meteorological data and time information can be obtained.
202. And determining trend characteristic information of the change of the power generation power of the target power generation equipment according to the power generation data of each group of historical equipment, the historical meteorological data corresponding to the power generation data of each group of historical equipment and the corresponding historical moment information.
After the power generation data of each group of historical equipment are obtained, the power prediction model after training can determine the characteristics of the power generation power of the target power generation equipment under the information of each historical moment and different historical meteorological data according to the power generation data of each group of historical equipment. For example: the power generation of the target power generation facility is in an increasing trend as solar radiation in the historical meteorological data increases during the time period from sunrise to noon each day. Meanwhile, after noon, as solar radiation decreases, the generated power may be in a decreasing trend. Of course, the power prediction model also can determine the specific change trend of the generated power according to other meteorological data, for example, factors such as wind speed, humidity, temperature, rainfall and the like can not influence the generated power according to the generated data of each group of historical equipment, so that the overall generation trend is determined. For example: and determining that the power generation power of the target power generation equipment is in a trend rising state according to the power generation data of each group of historical equipment, and representing the rising trend by using 1. Or the generated power of the target power generation equipment is in a trend descending state, and the descending trend is represented by 0.
In the embodiment of the application, each set of historical equipment power generation data has time information, so that the historical equipment power generation data can be regarded as time series data. The time series data can be arranged from small to large according to historical time information, and the power generation data and the meteorological data of the equipment are corresponding to each piece of historical time information. Preferably, in the embodiment of the present application, specific trend feature information may be determined according to the following formula (1):
Wherein in the formula (1) The trend characteristic information obtained by decomposing the time series data includes time series because of the existence of the historical time information. W t is a weight coefficient preset for the trend sequence. /(I)T is the length of time sequence, the length of history time sequence and C is the number of variables. H t is trend feature information obtained by weight calculation. It should be noted that, in the embodiment of the present application, the initial trend feature information decomposed according to the time series data may be obtained by decomposing through a DLinear model, and of course, other time series decomposition models may also be included, which is not limited by the embodiment of the present application.
203. And predicting the future power generation power of the target power generation equipment according to the trend characteristic information.
After trend characteristic information of the power generated by the target power generation equipment is obtained, the trend characteristic information and the power generation data of each set of historical equipment can be input into an LSTM model to predict the power generation. For example: and inputting each group of historical equipment power generation data and corresponding trend characteristic information into an LSTM model, and predicting the power generation of the target power generation equipment within 1 day in the future. It should be noted that in the embodiment of the present application, the LSTM model may also be a network of other processing time sequences, such as RNN, GRU, or transducer. In summary, in the embodiment of the present application, the power prediction model is a combined model of two models.
According to the power generation power prediction method provided by the application, the historical equipment power generation data of the power generation equipment under different time and different meteorological conditions can be obtained, and the power generation trend characteristic information of the power generation equipment under different time and different meteorological conditions can be analyzed, so that the regularity of the power generation power of the power generation equipment can be judged, and the power generation power of the power generation equipment can be further effectively predicted.
In order to better implement the embodiment of the present application, in an embodiment of the present application, predicting future generation power of a target generation device according to trend feature information includes:
According to the power generation data of each group of historical equipment, the historical meteorological data corresponding to the power generation data of each group of historical equipment and the corresponding historical moment information, fluctuation characteristic information of the power generation power change of the target power generation equipment is determined; and predicting the future power generation power of the target power generation equipment according to the trend characteristic information and the fluctuation characteristic information.
The above embodiment provides a scheme for generating power prediction based on trend feature information. However, in actual cases, when the overall generated power tends to be increased, there is a certain fluctuation and thus a certain power drop. For example: although the generated power tends to rise in a certain period of time, the generated power may fall in a short time in the period of time, that is, there is a case opposite to the trend. For example: when solar radiation is enhanced, wind speed, humidity, temperature and other related factors prevent the efficiency of the power generation device for converting light energy into electric energy, the power generation power of the power generation device may not be increased. Therefore, in the embodiment of the application, the fluctuation feature information also needs to be considered. The fluctuation feature information is a feature information opposite to the trend.
Therefore, in the embodiment of the application, the DLinear model can also decompose the fluctuation characteristic information when the power generation data of each group of historical equipment is acquired. Specifically, after the trend feature information is determined, the situation that the generated power is opposite in the trend feature information is found out, and the situation can be the fluctuation feature information. Specifically, the fluctuation-characteristic information may be determined according to the following formula (2):
wherein in the formula (2) For the fluctuation feature information obtained by decomposing the time series data, the fluctuation feature information also includes time sequence due to the existence of the historical time information. W s is a preset weight coefficient corresponding to the fluctuation sequence. /(I)T is the length of time sequence, the length of history time sequence and C is the number of variables. H s is fluctuation feature information obtained by weight calculation. In the embodiment of the present application, the characteristic information of the decomposition trend according to the time series data may be obtained by decomposing the DLinear model, and may of course also include other time series decomposition models.
After the fluctuation feature information is obtained, the LSTM model can also combine the fluctuation feature information with the trend feature information, and predict the generated power. For example: and inputting each group of historical equipment power generation data, corresponding trend characteristic information and fluctuation characteristic information into an LSTM model, and predicting the power generation of the target power generation equipment within 1 day in the future.
Specifically, reference may be made to fig. 3 and fig. 4, and fig. 3 is a power prediction model of DLinear-LSTM combined model. Fig. 4 shows the structure of an LSTM Cell of fig. 3. When the historical device power generation data X t including the time series information is input to the DLinear-LSTM combination model of fig. 3, it can be decomposed into a sequence of trend characteristic information and a sequence of fluctuation characteristic information through the DLinear model. Then, the sequence of trend feature information and the sequence of fluctuation feature information are input into the LSTM layer, respectively. The LSTM layer is provided with a plurality of LSTM cells, each LSTM Cell is used for extracting features of the sequence of trend feature information and the sequence of fluctuation feature information, then the LSTM cells enter the full-connection layer, and finally, a power prediction result is output, wherein the output power prediction result is a specific power value. Wherein t in the subscript t-m+1 or t-m+2 in fig. 3 is denoted as the current time instant and m is the first m time instants.
Further, the results for each LSTM Cell are shown in fig. 4. Wherein x t may be decomposedOr/>Due toAnd/>The subscripts of (c) are identical and are therefore denoted herein by x t. h t is the output of the LSTM Cell, and h t-1 is the output of the last LSTM Cell. f t determines the deleted target information in the current LSTM Cell, see formula (3):
ft=sigmoid(Wf,xxt+Wf,hht-1+bf)……(3)
In the formula (3), W f,x and W f,h are preset weight coefficients, and b f is a bias coefficient.
I t decides how much information can be saved under the current historical time information, and the calculation formula is shown in formula (4):
it=sigmoid(Wi,xxt+Wi,iht-1+bi)……(4)
in the formula (4), W i,x and W i,h are preset weight coefficients, and b i is a bias coefficient.
In addition, it is also necessary to determine the possible addition of LSTM Cell according to equation (5)
In the formula (5) of the present invention,/>For preset weight coefficient,/>Is the bias factor.
Then, a new CELL STATE is determined according to the formula (6)
The parameters in formula (6) are all mentioned in the above formula, and are not described here again.
Finally, o t decides which information in the cell state is used as output, see in particular equation (7) and equation (8):
ot=sigmoid(Wo,xxt+Wo,hht-1+bo)……(7)
ht=ot×tanh(st)……((8)
In the formula (7), W o,x and W o,h are preset weight coefficients, and b o is a bias coefficient. The parameters in formula (8) are mentioned in the other formulas, and are not described here again. The weight coefficient and the bias coefficient in any embodiment of the present application may be adjusted according to actual situations, and the specific embodiment of the present application is not limited.
In summary, after each LSTM Cell cycle calculation, the LSTM layer results may be input to the full link layer and the final power prediction may be performed.
According to the embodiment of the application, on the basis of trend characteristic information and after combining fluctuation characteristic information, the power generation power prediction model can learn the fluctuation characteristic of power in a certain trend, so that the probability of fluctuation of the power generation power of the power generation equipment under what kind of meteorological data and time of the power generation equipment can be effectively judged, and further the power generation power of the target power generation equipment in a certain future time period can be better predicted.
In order to better implement the embodiment of the present application, in an embodiment of the present application, predicting future generation power of a target generation device according to trend feature information includes:
and according to the trend characteristic information, predicting the power generation of the target power generation equipment in a future target time period, wherein the time difference between the minimum time information and the current time information in the target time period is smaller than a preset time difference threshold value, and the target time period is smaller than the preset time period threshold value.
In actual cases, the weather changes are large, and the weather causes affect the power generated by the power generation device. Therefore, in order to make the accuracy of the prediction higher, in predicting the generated power of the target power generation device, a model may be set to predict only the generated power in a future shorter period of time. For example: the generated power of the target power generation device is predicted within 5 minutes to 15 minutes in the future. At this time, if the time difference threshold is 7 minutes, the time difference between the minimum time information and the current time information, that is, the difference between 5 minutes and the current time information is calculated to be 5 minutes. At the same time, a time period length of between 5 minutes and 15 minutes, which is 10 minutes, is also calculated. If the time period threshold is 15 minutes. At this time, both the time period threshold and the time difference threshold are satisfied, and the model can make predictions. Otherwise, the model does not output the prediction result, and simultaneously reminds the user to reset the prediction time period. The manner of reminding the user to reset the prediction time period may be by performing text prompting on a display device connected to the electronic device on which the power prediction model is mounted, and the specific embodiment of the present application is not limited.
In order to better implement the embodiment of the present application, in one embodiment of the present application, before acquiring the historical equipment power generation data of at least two sets of target power generation equipment under different historical time information and different historical meteorological data, the method further includes:
Acquiring a plurality of groups of equipment power generation sample data for training a power prediction model, wherein each group of equipment power generation sample data corresponds to time information and meteorological sample data; dividing the power generation sample data of each group of equipment into training set samples and test set samples according to a set proportion; inputting the power generation sample data of each group of equipment in the training set sample as a current training sample into a power prediction model to be trained for training, and obtaining the predicted power of a second historical moment after a first historical moment corresponding to the current training sample, wherein the second historical moment corresponds to at least one target training sample except the current training sample in the training set sample; calculating a loss function value according to the predicted power of the predicted second historical moment and the power parameter of the equipment power generation sample data set corresponding to the second historical moment in the target training sample; carrying out iterative updating on model indexes of the power prediction model according to the loss function value until the loss function value meets the iteration termination condition, and obtaining a trained power prediction model; and testing the trained power prediction model through the equipment power generation sample data set in the test set sample, and completing the training of the power prediction model.
In the embodiment of the present application, the acquisition manner of acquiring the plurality of sets of device power generation sample data for power prediction model training may include, for example: and the power generation power of the target power generation equipment is A when XX is divided into XX in XX year, XX month and XX day. And meanwhile, according to the sensors of all meteorological data, the meteorological data such as solar radiation, temperature, humidity, precipitation, wind speed and the like of XX minutes in XX year, XX month, XX day, XX minute can be obtained and stored in the same position. And if the power prediction model needs to be trained, acquiring the power prediction model from the corresponding storage position, so that multiple groups of equipment power generation sample data for training the power prediction model, including meteorological data and time information, can be acquired.
After the power generation sample data of each group of equipment are obtained, the training sample set and the test sample set can be divided according to the power generation sample data of each group of equipment. The scheme of dividing the training sample set and the test sample set may be divided in a 9:1 manner, or may be divided in other manners, which is not limited in the embodiment of the present application.
Since the equipment power generation sample data includes time sequence information, and the power generation data and the meteorological data of the power generation equipment related in the equipment power generation sample data are collected in the process of historical operation of the power generation equipment. Therefore, the power prediction model can predict the equipment power generation sample data in the training sample set according to the learned characteristics to obtain the power prediction result of the training sample set. It is to be noted that, assuming that the total range of the time sequences in the training sample set is the power generation data and the weather data of the power generation equipment in one year, the power prediction model may predict the first half year data of the data in the one year to obtain a prediction result. I.e. the first historical moment may be half a year before data in one year of data, and the second historical moment after the first historical moment may be the second half of year of data. Since the power generation data of the power generation equipment and the corresponding meteorological data are also real parameters in nature. Therefore, the sample data of the latter half year can be regarded as real data for comparing the prediction result of the power prediction model aiming at the data of the first half year, so as to obtain the training loss of the model, and then the loss value is smaller than the preset requirement by adjusting the parameters of the model. The first history time and the second history time listed in the present application are not necessarily half a year, but may be other time points, and specific embodiments of the present application are not limited thereto.
For example, the current training sample set includes sample 1 and sample 2, and sample 2 is a training sample arranged in time series after sample 1. Sample 1 corresponds to a first historical time T1, sample 2 corresponds to a second historical time T2, and the first historical time T1 is earlier than the second historical time T2. During training, the power generation data and the meteorological data of the power generation equipment in the sample 1 are adopted to predict the power generation power Pre-PowerT2 of the power generation equipment at the second historical moment T2, then the power generation power Sam-PowerT of the power generation equipment corresponding to the second historical moment T2 recorded in the sample 2 is compared and analyzed, and a loss function value is calculated and used for iteratively updating the weight parameters of the power prediction model according to the set gradient direction. The training mode can save the time of sample processing without manually calibrating or machine calibrating each training sample. The loss function according to the present application may refer to equation (9).
In formula (9), y i and y i ′ are the real power and the predicted power, respectively; m is the ordinal number of the sample. Meanwhile, equation (10) needs to be considered.
In formula (10), y i and y i ′ are the real power and the predicted power, respectively; m is the ordinal number of the sample; n is the number of samples; cap is the rated capacity of the photovoltaic power station; acc is the prediction accuracy. When the prediction accuracy of each sample is higher than a certain accuracy threshold, the model can be determined to complete training.
In addition, the training samples in the test set samples are the same as the training samples in the training set samples, and detailed description thereof will not be repeated here.
In order to better implement the embodiment of the present application, in an embodiment of the present application, obtaining a plurality of sets of device power generation sample data for training a power prediction model includes:
Acquiring a plurality of groups of initial equipment power generation sample data for training a power prediction model; respectively taking each group of initial equipment power generation sample data as a target initial power generation sample data group, and determining missing parameters in the target initial power generation sample data group; if the target initial power generation sample data set has no parameter deletion, determining the target initial power generation sample data set as a group of equipment power generation sample data; if the target initial power generation sample data set has the parameter missing, filling the missing target parameter in the target initial power generation sample data set to obtain a group of equipment power generation sample data.
In the above embodiment, an acquisition manner of acquiring power generation sample data of a device for model training is provided. However, in actual processes, there may be some data missing from a set of plant power generation sample data, such as some meteorological data missing, or power generation data. Thus, when there is a data loss, the set of plant power generation sample data cannot be used for model training in theory. However, in order to increase the number of samples, the embodiment of the application can perform data supplementation on the data of the power generation sample data of the equipment with the missing data, so that the data of the power generation sample data of the equipment with the missing data is reserved.
In the embodiment of the application, if the target initial power generation sample data set with missing data exists, a specific data missing type, such as missing the power generation of power generation equipment or missing the environmental temperature in meteorological data, can be determined. Assuming that the ambient temperature is missing from the target initial power generation sample data set, it is necessary to perform data population for the ambient temperature in the target initial power generation sample data set. For example: the method comprises the steps of determining target time information corresponding to a target initial power generation sample data set, searching the environmental temperature corresponding to the target time information in a weather website of the Internet according to the corresponding target time information, and further filling the environmental temperature information of the target initial power generation sample data set. Or the part of the data type network can not be searched, for example, the generated power can be determined, the initial generated power sample data set corresponding to the adjacent time information of the target time information corresponding to the target initial generated power sample data set can be determined, and the average of the generated power in the corresponding initial generated power sample data set is taken for filling. Specifically, there are a plurality of data filling modes, and the embodiment of the present application is not limited.
In order to better implement the embodiments of the present application, in an embodiment of the present application, the method further includes:
If the target initial power generation sample data set has the parameter missing, determining the number of the missing parameters in the target initial power generation sample data set; and deleting the target initial power generation sample data set if the number of the missing parameters in the target initial power generation sample data set is larger than a preset number threshold.
The above embodiments provide a scheme of data population of a target initial power generation sample data set of missing data. However, if there are too many missing parameters in some of the target initial power generation sample data sets, for example, if there are missing parameters such as power generation, temperature, solar radiation, humidity, etc. at the same time. The target initial power generation sample data set is not suitable for training of the power prediction model. At this time, the target initial power generation sample data set may be deleted. In the embodiment of the present application, the specific number threshold may be set according to the actual situation, and the specific embodiment of the present application is not limited.
In order to better implement the embodiment of the present application, in one embodiment of the present application, each set of device power generation sample data in a training set sample is input as a current training sample into a power prediction model to be trained to perform training, so as to obtain predicted power at a second historical time after a first historical time corresponding to the current training sample, including:
Respectively taking each group of equipment power generation sample data in the training set sample as a target equipment power generation sample data set, and determining average power generation sample data of the target equipment power generation sample data set according to the target equipment power generation sample data set and the equipment power generation sample data of the time information before the target time information of the target equipment power generation sample data set; determining average power generation difference sample data of the target equipment power generation sample data set according to the target equipment power generation sample data set and equipment power generation sample data of time information before target time information of the target equipment power generation sample data set; according to average power generation sample data and average power generation difference sample data of the target equipment power generation sample data set, trend characteristic information and fluctuation characteristic information of the target equipment power generation sample data set from the first historical moment to the second historical moment are determined; and determining the predicted power of the second historical moment after the first historical moment corresponding to the current training sample according to the trend characteristic information and the fluctuation characteristic information between the first historical moment and the second historical moment.
In the embodiment of the application, in order that the power prediction model can better learn the law of the change of the power generated by the power generation equipment in the past time, the embodiment of the application provides a learning mode of the power prediction model.
Specifically, in the learning process of the power prediction model, when the power prediction model obtains the power generation sample data of each device for training, average power generation sample data of each target device power generation sample data set may be determined according to the corresponding time information. Assuming that the power generation sample data is power generation sample power, the manner of determining average power generation sample data of the power generation sample power may include: an average value of the generated power of each target device generated sample data set and the generated power corresponding to the first 14 pieces of adjacent time information thereof is determined. For example: at this time, the power generated in the device power generation sample data set corresponding to 14 pieces of time information adjacent to the time information a is determined. Then, the average value of the 15 generated powers is calculated. According to the embodiment of the application, the power prediction model can calculate the corresponding average power for each equipment power generation sample data set capable of calculating the average power, so that the power prediction model can better learn trend characteristic information of power change according to the average power. Of course, in the embodiment of the present application, the number of samples required for calculating the average power is not necessarily 15, and may be specifically set according to practical situations.
Similarly, the average power generation difference sample data is determined in the same manner as the average power generation sample data in the above description. For example: and determining the power generation power in the equipment power generation sample data set adjacent to the time information of the current target equipment power generation sample data set, and calculating the difference value between the adjacent power in the adjacent data set and the power generation power in the current target equipment power generation sample data set. In the embodiment of the application, the average power generation difference sample data can be calculated by taking the power generation power in the first 4 equipment power generation sample data sets adjacent to the current time information. For example: and respectively calculating the difference values of the power generation powers in the adjacent first 4 equipment power generation sample data sets and the current target equipment power generation sample data set, so as to obtain 4 difference values, and calculating the average value of the 4 difference values to obtain average power generation difference sample data. Of course, in the embodiment of the present application, the number of data sets for calculating the average power generation difference sample data may be other than 4, and the data sets may be specifically determined according to the actual situation. By means of calculating average power generation difference sample data, the power prediction model can learn fluctuation characteristic information better.
In order to better implement the power generation prediction method in the embodiment of the present application, on top of the power generation prediction method, the embodiment of the present application further provides a power generation prediction device, which is applied to a power prediction model for completing training, as shown in fig. 5, the power generation prediction device 300 includes:
The acquiring module 301 is configured to acquire historical equipment power generation data of at least two groups of target power generation equipment under different historical time information and different historical meteorological data, where each group of historical equipment power generation data corresponds to the historical time information and the historical meteorological data;
The determining module 302 is configured to determine trend feature information of a change in power generated by the target power generating device according to each set of power generating data of the historical device, historical meteorological data corresponding to each set of power generating data of the historical device, and corresponding historical time information;
And the prediction module 303 is used for predicting the future power generation power of the target power generation equipment according to the trend characteristic information.
According to the power generation power prediction device provided by the application, the historical equipment power generation data of the power generation equipment under different time and different meteorological conditions can be obtained through the obtaining module 301, and then the trend characteristic information of the power generation of the historical equipment power generation data under different time and different meteorological conditions is analyzed through the determining module 302, so that the regularity of the power generation power of the power generation equipment can be judged, and the power generation power of the power generation equipment can be effectively predicted through the prediction module 303.
In some embodiments of the present application, the prediction module 303 is specifically configured to:
According to the power generation data of each group of historical equipment, the historical meteorological data corresponding to the power generation data of each group of historical equipment and the corresponding historical moment information, fluctuation characteristic information of the power generation power change of the target power generation equipment is determined;
and predicting the future power generation power of the target power generation equipment according to the trend characteristic information and the fluctuation characteristic information.
In some embodiments of the present application, the prediction module 303 is specifically further configured to:
and according to the trend characteristic information, predicting the power generation of the target power generation equipment in a future target time period, wherein the time difference between the minimum time information and the current time information in the target time period is smaller than a preset time difference threshold value, and the target time period is smaller than the preset time period threshold value.
In some embodiments of the present application, the apparatus further comprises a training module, wherein the training module is specifically configured to:
Acquiring a plurality of groups of equipment power generation sample data for training a power prediction model, wherein each group of equipment power generation sample data corresponds to time information and meteorological sample data;
dividing the power generation sample data of each group of equipment into training set samples and test set samples according to a set proportion;
Inputting the power generation sample data of each group of equipment in the training set sample as a current training sample into a power prediction model to be trained for training, and obtaining the predicted power of a second historical moment after a first historical moment corresponding to the current training sample, wherein the second historical moment corresponds to at least one target training sample except the current training sample in the training set sample;
Calculating a loss function value according to the predicted power of the predicted second historical moment and the power parameter of the equipment power generation sample data set corresponding to the second historical moment in the target training sample;
Carrying out iterative updating on model indexes of the power prediction model according to the loss function value until the loss function value meets the iteration termination condition, and obtaining a trained power prediction model;
And testing the trained power prediction model through the equipment power generation sample data set in the test set sample, and completing the training of the power prediction model.
In some embodiments of the present application, the training module is specifically further configured to:
Acquiring a plurality of groups of initial equipment power generation sample data for training a power prediction model;
Respectively taking each group of initial equipment power generation sample data as a target initial power generation sample data group, and determining missing parameters in the target initial power generation sample data group;
If the target initial power generation sample data set has no parameter deletion, determining the target initial power generation sample data set as a group of equipment power generation sample data;
If the target initial power generation sample data set has the parameter missing, filling the missing target parameter in the target initial power generation sample data set to obtain a group of equipment power generation sample data.
In some embodiments of the present application, the training module is specifically further configured to:
If the target initial power generation sample data set has the parameter missing, determining the number of the missing parameters in the target initial power generation sample data set;
And deleting the target initial power generation sample data set if the number of the missing parameters in the target initial power generation sample data set is larger than a preset number threshold.
In some embodiments of the present application, the training module is specifically further configured to:
Respectively taking each group of equipment power generation sample data in the training set sample as a target equipment power generation sample data set, and determining average power generation sample data of the target equipment power generation sample data set according to the target equipment power generation sample data set and the equipment power generation sample data of the time information before the target time information of the target equipment power generation sample data set;
Determining average power generation difference sample data of the target equipment power generation sample data set according to the target equipment power generation sample data set and equipment power generation sample data of time information before target time information of the target equipment power generation sample data set;
According to average power generation sample data and average power generation difference sample data of the target equipment power generation sample data set, trend characteristic information and fluctuation characteristic information of the target equipment power generation sample data set from the first historical moment to the second historical moment are determined;
And determining the predicted power of the second historical moment after the first historical moment corresponding to the current training sample according to the trend characteristic information and the fluctuation characteristic information between the first historical moment and the second historical moment.
The embodiment of the application also provides electronic equipment, which comprises a processor, a memory and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to realize the steps in the generated power prediction method of any one of the embodiments of the application. The electronic device integrates any of the generated power prediction methods provided by the embodiments of the present application, as shown in fig. 6, which shows a schematic structural diagram of the electronic device according to the embodiments of the present application, specifically:
The electronic device may include one or more processing cores 'processors 401, one or more computer-readable storage media's memory 402, power supply 403, and input unit 404, among other components. It will be appreciated by those skilled in the art that the electronic device structure shown in fig. 6 is not limiting of the electronic device and may include more or fewer components than shown, or may combine certain components, or a different arrangement of components. Wherein:
the processor 401 is a control center of the electronic device, connects various parts of the entire electronic device using various interfaces and lines, and performs various functions of the electronic device and processes data by running or executing software programs and/or modules stored in the memory 402, and calling data stored in the memory 402, thereby performing overall monitoring of the electronic device. Optionally, processor 401 may include one or more processing cores; the Processor 401 may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (DIGITAL SIGNAL Processor, DSP), application SPECIFIC INTEGRATED Circuit (ASIC), off-the-shelf Programmable gate array (Field-Programmable GATE ARRAY, FPGA) or other Programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like, preferably, the processor 401 may integrate an application processor, which primarily handles operating systems, user interfaces, applications, etc., with a modem processor, which primarily handles wireless communications. It will be appreciated that the modem processor described above may not be integrated into the processor 401.
The memory 402 may be used to store software programs and modules, and the processor 401 executes various functional applications and data processing by executing the software programs and modules stored in the memory 402. The memory 402 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program (such as a sound playing function, an image playing function, etc.) required for at least one function, and the like; the storage data area may store data created according to the use of the electronic device, etc. In addition, memory 402 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 with access to the memory 402.
The electronic device further comprises a power supply 403 for supplying power to the various components, preferably the power supply 403 may be logically connected to the processor 401 by a power management system, so that functions of managing charging, discharging, and power consumption are performed by the power management system. The power supply 403 may also include one or more of any of a direct current or alternating current power supply, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, and the like.
The electronic device may further comprise an input unit 404, which input unit 404 may be used for receiving input digital or character information and generating keyboard, mouse, joystick, optical or trackball signal inputs in connection with user settings and function control.
Although not shown, the electronic device may further include a display unit or the like, which is not described herein. In particular, in this embodiment, the processor 401 in the electronic device loads executable files corresponding to the processes of one or more application programs into the memory 402 according to the following instructions, and the processor 401 executes the application programs stored in the memory 402, so as to implement various functions, for example:
acquiring historical equipment power generation data of at least two groups of target power generation equipment under different historical moment information and different historical meteorological data, wherein each group of historical equipment power generation data corresponds to the historical moment information and the historical meteorological data;
according to the power generation data of each group of historical equipment, the historical meteorological data corresponding to the power generation data of each group of historical equipment and the corresponding historical moment information, trend characteristic information of the power generation power change of the target power generation equipment is determined;
and predicting the future power generation power of the target power generation equipment according to the trend characteristic information.
Those of ordinary skill in the art will appreciate that all or a portion of the steps of the various methods of the above embodiments may be performed by instructions, or by instructions controlling associated hardware, which may be stored in a computer-readable storage medium and loaded and executed by a processor.
To this end, an embodiment of the present application provides a computer-readable storage medium, which may include: read Only Memory (ROM), random access Memory (RAM, random Access Memory), magnetic or optical disk, and the like. On which a computer program is stored, the computer program being loaded by a processor to perform the steps of any of the generated power prediction methods provided by the embodiments of the present application. For example, the loading of the computer program by the processor may perform the steps of:
acquiring historical equipment power generation data of at least two groups of target power generation equipment under different historical moment information and different historical meteorological data, wherein each group of historical equipment power generation data corresponds to the historical moment information and the historical meteorological data;
according to the power generation data of each group of historical equipment, the historical meteorological data corresponding to the power generation data of each group of historical equipment and the corresponding historical moment information, trend characteristic information of the power generation power change of the target power generation equipment is determined;
and predicting the future power generation power of the target power generation equipment according to the trend characteristic information.
In the foregoing embodiments, the descriptions of the embodiments are focused on, and the portions of one embodiment that are not described in detail in the foregoing embodiments may be referred to in the foregoing detailed description of other embodiments, which are not described herein again.
In the implementation, each unit or structure may be implemented as an independent entity, or may be implemented as the same entity or several entities in any combination, and the implementation of each unit or structure may be referred to the foregoing method embodiments and will not be repeated herein.
The specific implementation of each operation above may be referred to the previous embodiments, and will not be described herein.
The above description is made in detail of a method and apparatus for predicting generated power provided by the embodiment of the present application, and specific examples are applied to illustrate the principles and embodiments of the present application, and the above description of the embodiment is only used to help understand the method and core idea of the present application; meanwhile, as those skilled in the art will vary in the specific embodiments and application scope according to the ideas of the present application, the present description should not be construed as limiting the present application in summary.
Claims (9)
1. A method of generating power prediction applied to a trained power prediction model, the method comprising:
acquiring historical equipment power generation data of at least two groups of target power generation equipment under different historical moment information and different historical meteorological data, wherein each group of historical equipment power generation data corresponds to the historical moment information and the historical meteorological data;
Inputting each set of historical equipment power generation data, historical meteorological data corresponding to each set of historical equipment power generation data and corresponding historical time information into DLinear layers in the power prediction model to obtain trend characteristic information and fluctuation characteristic information of the change of the power generated by the target power generation equipment;
inputting the trend characteristic information and the fluctuation characteristic information into an LSTM layer in the power prediction model to obtain an output result of the LSTM layer;
And inputting an output result of the LSTM layer into a full connection layer in the power prediction model, and predicting the future power generation power of the target power generation equipment, wherein the future power generation power of the target power generation equipment comprises a power value.
2. The generated power prediction method according to claim 1, wherein predicting the target power generation device future generated power based on the trend feature information includes:
And predicting the power generation of the target power generation equipment in a future target time period according to the trend characteristic information, wherein the time difference between the minimum time information and the current time information in the target time period is smaller than a preset time difference threshold value, and the target time period is smaller than a preset time period threshold value.
3. The method of claim 1, wherein the acquiring at least two sets of target power generation devices includes, prior to the acquiring historical device power generation data at different historical time information and different historical meteorological data,:
acquiring a plurality of groups of equipment power generation sample data for training the power prediction model, wherein each group of equipment power generation sample data corresponds to time information and meteorological sample data;
dividing each group of the equipment power generation sample data into training set samples and test set samples according to a set proportion;
inputting the power generation sample data of each group of equipment in the training set sample as a current training sample into a power prediction model to be trained to perform training, and obtaining predicted power of a second historical moment after a first historical moment corresponding to the current training sample, wherein the second historical moment corresponds to at least one target training sample except the current training sample in the training set sample;
Calculating a loss function value according to the predicted power of the predicted second historical moment and the power parameter of the equipment power generation sample data set corresponding to the second historical moment in the target training sample;
Iteratively updating the model index of the power prediction model according to the loss function value until the loss function value meets the iteration termination condition, so as to obtain a trained power prediction model;
And testing the trained power prediction model through the equipment power generation sample data set in the test set sample, and completing the training of the power prediction model.
4. A method of generating power prediction according to claim 3, wherein the obtaining sets of device generation sample data for training of the power prediction model comprises:
acquiring a plurality of groups of initial equipment power generation sample data for training the power prediction model;
Respectively taking each group of initial equipment power generation sample data as a target initial power generation sample data group, and determining missing parameters in the target initial power generation sample data group;
if the target initial power generation sample data set has no parameter missing, determining that the target initial power generation sample data set is a set of equipment power generation sample data;
And if the target initial power generation sample data set has parameter deletion, filling the target parameter deleted in the target initial power generation sample data set to obtain a group of equipment power generation sample data.
5. The generated power prediction method according to claim 4, characterized in that the method further comprises:
if the target initial power generation sample data set has the parameter missing, determining the number of the missing parameters in the target initial power generation sample data set;
And deleting the target initial power generation sample data set if the number of the missing parameters in the target initial power generation sample data set is larger than a preset number threshold.
6. The method for predicting generated power according to claim 3, wherein the step of inputting the generated power sample data of each set of devices in the training set sample as a current training sample into a power prediction model to be trained to perform training, and obtaining the predicted power at a second historical time after the first historical time corresponding to the current training sample comprises:
Respectively taking each group of equipment power generation sample data in the training set sample as a target equipment power generation sample data set, and determining average power generation sample data of the target equipment power generation sample data set according to the target equipment power generation sample data set and equipment power generation sample data of time information before target time information of the target equipment power generation sample data set;
determining average power generation difference sample data of the target equipment power generation sample data set according to the target equipment power generation sample data set and equipment power generation sample data of time information before target time information of the target equipment power generation sample data set;
According to average power generation sample data and average power generation difference sample data of the target equipment power generation sample data set, trend characteristic information and fluctuation characteristic information between the first historical moment and the second historical moment of the target equipment power generation sample data set are determined;
And determining the predicted power of the second historical moment after the first historical moment corresponding to the current training sample according to the trend characteristic information and the fluctuation characteristic information between the first historical moment and the second historical moment.
7. A generated power prediction apparatus for use in a trained power prediction model, the apparatus comprising:
the acquisition module is used for acquiring historical equipment power generation data of at least two groups of target power generation equipment under different historical moment information and different historical meteorological data, wherein each group of historical equipment power generation data corresponds to the historical moment information and the historical meteorological data;
The determining module is used for inputting each group of historical equipment power generation data, historical meteorological data corresponding to each group of historical equipment power generation data and corresponding historical time information into DLinear layers in the power prediction model to obtain trend characteristic information and fluctuation characteristic information of the change of the power generated by the target power generation equipment;
The prediction module is used for inputting the trend characteristic information and the fluctuation characteristic information into an LSTM layer in the power prediction model to obtain an output result of the LSTM layer;
And inputting an output result of the LSTM layer into a full connection layer in the power prediction model, and predicting the future power generation power of the target power generation equipment, wherein the future power generation power of the target power generation equipment comprises a power value.
8. An electronic device comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, the processor executing the computer program to perform the steps in the generated power prediction method of any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program that is executed by a processor to implement the steps in the generated power prediction method of any one of claims 1 to 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311714781.6A CN117748482B (en) | 2023-12-13 | 2023-12-13 | Method and device for predicting generated power, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311714781.6A CN117748482B (en) | 2023-12-13 | 2023-12-13 | Method and device for predicting generated power, electronic equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117748482A CN117748482A (en) | 2024-03-22 |
CN117748482B true CN117748482B (en) | 2024-06-14 |
Family
ID=90251989
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311714781.6A Active CN117748482B (en) | 2023-12-13 | 2023-12-13 | Method and device for predicting generated power, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117748482B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110781947A (en) * | 2019-10-22 | 2020-02-11 | 北京交通大学 | Power load prediction model training and power load prediction method and device |
CN112906757A (en) * | 2021-01-29 | 2021-06-04 | 广东晨兴智能科技有限公司 | Model training method, optical power prediction method, device, equipment and storage medium |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113919232A (en) * | 2021-10-25 | 2022-01-11 | 北京航天创智科技有限公司 | Photovoltaic power station power prediction method and system based on recurrent neural network |
CN114330097A (en) * | 2021-11-26 | 2022-04-12 | 中国华能集团清洁能源技术研究院有限公司 | Heterogeneous data and deep learning wind power cluster power prediction method and device |
CN116502692A (en) * | 2023-03-24 | 2023-07-28 | 华为技术有限公司 | Model training method, photovoltaic power generation power prediction method and device |
-
2023
- 2023-12-13 CN CN202311714781.6A patent/CN117748482B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110781947A (en) * | 2019-10-22 | 2020-02-11 | 北京交通大学 | Power load prediction model training and power load prediction method and device |
CN112906757A (en) * | 2021-01-29 | 2021-06-04 | 广东晨兴智能科技有限公司 | Model training method, optical power prediction method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN117748482A (en) | 2024-03-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110556820A (en) | Method and apparatus for determining energy system operating scenarios | |
CN115425680B (en) | Power prediction model construction and prediction method of multi-energy combined power generation system | |
Li et al. | States prediction for solar power and wind speed using BBA‐SVM | |
Bouguera et al. | A novel solar energy predictor for communicating sensors | |
Askarzadeh | Voltage prediction of a photovoltaic module using artificial neural networks | |
CN116739172B (en) | Method and device for ultra-short-term prediction of offshore wind power based on climbing identification | |
CN117613883A (en) | Method and device for predicting generated power, computer equipment and storage medium | |
Binelo et al. | Mathematical modeling and parameter estimation of battery lifetime using a combined electrical model and a genetic algorithm | |
CN114676923A (en) | Method and device for predicting generated power, computer equipment and storage medium | |
Wu et al. | SOH prediction for lithium-ion battery based on improved support vector regression | |
CN112906757A (en) | Model training method, optical power prediction method, device, equipment and storage medium | |
Rafeeq Ahmed et al. | Maximum power point tracking of PV grids using deep learning | |
Sheikh et al. | A data-driven comparative analysis of lithium-ion battery state of health and capacity estimation | |
Bhatt et al. | Machine learning‐based approach for useful capacity prediction of second‐life batteries employing appropriate input selection | |
Lee et al. | Confidence‐aware deep learning forecasting system for daily solar irradiance | |
Zhang et al. | Prediction method of line loss rate in low‐voltage distribution network based on multi‐dimensional information matrix and dimensional attention mechanism‐long‐and short‐term time‐series network | |
Ouyang et al. | Optimisation of time window size for wind power ramps prediction | |
CN117748482B (en) | Method and device for predicting generated power, electronic equipment and storage medium | |
KR20220088946A (en) | Radial prediction processing method, stack generalization model training method and apparatus | |
CN117289167A (en) | Battery remaining life prediction method, device and medium based on multiple neural network | |
CN116937560A (en) | New energy output prediction method, device, equipment and readable storage medium | |
CN115275975B (en) | Method and device for determining matching degree of electric power data of optical storage charging station | |
Wang et al. | [Retracted] The Management of Energy Transformation through Laser Charging in WPT for 5G Application: Prediction Model of the In0. 3Ga0. 7As Solar Cell | |
CN114493051A (en) | Photovoltaic power prediction method and device for improving precision based on combined prediction | |
Bin et al. | Simple and effective fault diagnosis method of power lithium-ion battery based on GWA-DBN |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |