CN116722545B - Photovoltaic power generation prediction method based on multi-source data and related equipment - Google Patents
Photovoltaic power generation prediction method based on multi-source data and related equipment Download PDFInfo
- Publication number
- CN116722545B CN116722545B CN202310984296.4A CN202310984296A CN116722545B CN 116722545 B CN116722545 B CN 116722545B CN 202310984296 A CN202310984296 A CN 202310984296A CN 116722545 B CN116722545 B CN 116722545B
- Authority
- CN
- China
- Prior art keywords
- data
- photovoltaic
- power generation
- historical
- prediction model
- 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
- 238000010248 power generation Methods 0.000 title claims abstract description 224
- 238000000034 method Methods 0.000 title claims abstract description 62
- 238000012549 training Methods 0.000 claims abstract description 31
- 230000007613 environmental effect Effects 0.000 claims description 41
- 238000000605 extraction Methods 0.000 claims description 22
- 230000004927 fusion Effects 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 5
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 2
- 239000013598 vector Substances 0.000 description 14
- 230000008859 change Effects 0.000 description 8
- 238000004891 communication Methods 0.000 description 5
- 230000000007 visual effect Effects 0.000 description 5
- 238000013527 convolutional neural network Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000013178 mathematical model Methods 0.000 description 4
- 230000005855 radiation Effects 0.000 description 4
- 230000004044 response Effects 0.000 description 4
- 238000013459 approach Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 230000032683 aging Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000003749 cleanliness Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000011109 contamination Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
- H02J2300/20—The dispersed energy generation being of renewable origin
- H02J2300/22—The renewable source being solar energy
-
- 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
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Economics (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Power Engineering (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a photovoltaic power generation prediction method based on multi-source data and related equipment, and relates to the technical field of photovoltaic power generation, wherein the method comprises the following steps: acquiring historical data and photovoltaic equipment data, wherein the historical data comprises historical power generation data and historical environment data, and the photovoltaic equipment data comprises equipment three-dimensional data and equipment working data; constructing a photovoltaic equipment model according to the photovoltaic equipment data, wherein the photovoltaic equipment model comprises an initial prediction model; training the initial prediction model according to the historical power generation data, the historical environment data and the photovoltaic equipment data until the initial prediction model converges to obtain a photovoltaic power generation prediction model; when a prediction instruction is detected, the prediction instruction is predicted based on the photovoltaic power generation prediction model, and predicted photovoltaic data are obtained. The method and the device can improve the accuracy and the adaptability of photovoltaic power generation prediction.
Description
Technical Field
The invention relates to the technical field of photovoltaic power generation, in particular to a photovoltaic power generation prediction method based on multi-source data and related equipment.
Background
Photovoltaic power generation is a technology for generating power by utilizing solar energy, and has the advantages of cleanliness, reproducibility, distribution and the like. However, the output power of photovoltaic power generation is affected by various factors, such as solar radiation intensity, temperature, wind speed, cloud cover, and the like, and thus has strong randomness and uncertainty. In order to ensure stable operation of the photovoltaic power generation system and improve the dispatching efficiency of the power grid, accurate prediction of photovoltaic power generation is required.
The current photovoltaic power generation prediction mode mainly relies on historical power generation to predict, and environmental conditions are ignored. Although the method is simple and easy to implement, the actual condition of photovoltaic power generation cannot be reflected, and large errors are easy to occur. For this reason, it has been proposed to use convolutional neural networks for prediction, and to use meteorological data as input to learn complex nonlinear relationships between photovoltaic power generation and environmental factors. This approach may improve the accuracy and robustness of the prediction, but has some limitations. This approach ignores conditions of the photovoltaic device itself, such as component aging, contamination, failure, etc., which also affect the output power of the photovoltaic power generation. Therefore, this approach creates a model that is too static to accommodate the conditions of different photovoltaic devices.
Disclosure of Invention
The invention provides a photovoltaic power generation prediction method based on multi-source data, which is used for solving the defects of a photovoltaic power generation prediction method in the prior art, realizing the photovoltaic power generation prediction based on the multi-source data and improving the adaptability and the accuracy of the photovoltaic prediction.
The invention provides a photovoltaic power generation prediction method based on multi-source data, which comprises the following steps:
acquiring historical data and photovoltaic equipment data, wherein the historical data comprises historical power generation data and historical environment data, and the photovoltaic equipment data comprises equipment three-dimensional data and equipment working data;
constructing a photovoltaic equipment model according to the photovoltaic equipment data, wherein the photovoltaic equipment model comprises an initial prediction model;
training the initial prediction model according to the historical power generation data, the historical environment data and the photovoltaic equipment data until the initial prediction model converges to obtain a photovoltaic power generation prediction model;
when a prediction instruction is detected, the prediction instruction is predicted based on the photovoltaic power generation prediction model, and predicted photovoltaic data are obtained.
According to the photovoltaic power generation prediction method based on multi-source data provided by the invention, the training of the initial prediction model is performed according to the historical power generation data, the historical environment data and the photovoltaic equipment data until the initial prediction model converges, and the obtaining of the photovoltaic power generation prediction model comprises the following steps:
The characteristic extraction module is used for respectively extracting the historical environmental characteristics corresponding to the historical environmental data, the historical power generation characteristics corresponding to the historical power generation data and the photovoltaic equipment characteristics corresponding to the photovoltaic equipment data based on the initial prediction model;
predicting based on the historical environmental characteristics to obtain a first predicted value; the method comprises the steps of,
predicting based on the characteristics of the photovoltaic equipment to obtain a second predicted value;
performing feature fusion on the first predicted value and the second predicted value to obtain predicted power generation features;
and adjusting parameters of the initial prediction model according to the predicted power generation characteristics and the historical power generation characteristics until the initial prediction model converges to obtain a photovoltaic power generation prediction model.
According to the photovoltaic power generation prediction method based on multi-source data provided by the invention, the feature extraction module based on the initial prediction model is used for respectively extracting the historical environmental features corresponding to the historical environmental data, the historical power generation features corresponding to the historical power generation data and the photovoltaic equipment features corresponding to the photovoltaic equipment data, and the method comprises the following steps:
convolving the historical environmental data to obtain historical environmental characteristics; the method comprises the steps of,
Convolving the photovoltaic equipment data to obtain photovoltaic equipment characteristics;
according to the historical power generation data, calculating the attention weight corresponding to each piece of historical power generation data;
and extracting the historical power generation characteristics corresponding to the historical power generation data according to the attention weight.
According to the photovoltaic power generation prediction method based on multi-source data provided by the invention, the steps of adjusting parameters of the initial prediction model according to the prediction power generation characteristics and the historical power generation characteristics until the initial prediction model converges, and obtaining the photovoltaic power generation prediction model comprise the following steps:
calculating a loss value corresponding to the predicted power generation characteristic according to the historical power generation characteristic;
and carrying out internal circulation and external circulation on the photovoltaic equipment model according to a preset circulation rule and the loss value until the photovoltaic equipment model converges to obtain a photovoltaic power generation prediction model, wherein the internal circulation comprises:
and according to the loss value, carrying out parameter adjustment on a feature extraction module and a feature fusion module in the photovoltaic equipment model.
According to the photovoltaic power generation prediction method based on multi-source data provided by the invention, the construction of the photovoltaic equipment model according to the photovoltaic equipment data comprises the following steps:
Constructing an equipment three-dimensional model according to the equipment three-dimensional data in the photovoltaic equipment data;
and carrying out data fusion on the three-dimensional model of the equipment according to the equipment working data to obtain a photovoltaic equipment model.
According to the photovoltaic power generation prediction method based on multi-source data provided by the invention, when a prediction instruction is detected, the prediction instruction is predicted based on the photovoltaic power generation prediction model, and after the predicted photovoltaic data is obtained, the method further comprises the steps of:
according to the prediction instruction, determining prediction equipment to be predicted;
determining an equipment display model corresponding to the prediction equipment according to the photovoltaic equipment model;
and fusing and displaying the predicted photovoltaic data with the equipment display model.
The photovoltaic power generation prediction method based on the multi-source data provided by the invention further comprises the following steps:
for each photovoltaic device, when motion information corresponding to the photovoltaic device is detected, calculating updated device information corresponding to the photovoltaic device according to the motion information;
generating an update animation corresponding to the photovoltaic equipment according to the update equipment information; the method comprises the steps of,
and updating the data of the photovoltaic equipment model.
The invention also provides a photovoltaic power generation prediction device based on the multi-source data, which comprises:
the photovoltaic power generation system comprises an acquisition module, a storage module and a storage module, wherein the acquisition module is used for acquiring historical data and photovoltaic equipment data, and the historical data comprise historical power generation data and historical environment data;
the construction module is used for constructing a photovoltaic equipment model according to the photovoltaic equipment data, wherein the photovoltaic equipment model comprises an initial prediction model;
the training module is used for training the initial prediction model according to the historical power generation data, the historical environment data and the photovoltaic equipment data until the initial prediction model converges to obtain a photovoltaic power generation prediction model;
and the prediction module is used for predicting the prediction instruction based on the photovoltaic power generation prediction model when the prediction instruction is detected, so as to obtain predicted photovoltaic data.
The invention also provides electronic equipment, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the photovoltaic power generation prediction method based on the multi-source data when executing the computer program.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a multi-source data based photovoltaic power generation prediction method as described in any of the above.
According to the photovoltaic power generation prediction method and the related equipment based on the multi-source data, the historical data and the photovoltaic equipment data are acquired to collect the related information of photovoltaic power generation, wherein the related information comprises the historical environment data, the historical power generation data and the photovoltaic equipment data. These data can help understand the characteristics and regularity of photovoltaic power generation, as well as the performance and status of photovoltaic devices. And then, constructing a photovoltaic equipment model according to the photovoltaic equipment data to simulate the working principle and the output characteristic of the photovoltaic equipment. This model can help analyze the response and impact of the photovoltaic device on environmental changes. And then, based on the initial prediction model, extracting the historical environmental characteristics corresponding to the historical environmental data, the historical power generation characteristics corresponding to the historical power generation data and the photovoltaic equipment characteristics corresponding to the photovoltaic equipment data, and training the initial prediction model based on the historical environmental characteristics and the historical power generation characteristics to adjust and optimize parameters of the prediction model so that the parameters can be better fit with the rule of photovoltaic power generation. On one hand, photovoltaic equipment data are used in the originally constructed model, so that the similarity of the model and real photovoltaic equipment is improved, and on the other hand, historical environment data and photovoltaic equipment data are combined in the training process, so that the accuracy and reliability of a prediction model are improved, errors and deviation of the prediction model are reduced, and the applicability of the model is improved.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a photovoltaic power generation prediction method based on multi-source data;
fig. 2 is a schematic diagram of a photovoltaic data fusion display predicted in the photovoltaic power generation prediction method based on multi-source data;
fig. 3 is a schematic structural diagram of a photovoltaic power generation prediction device based on multi-source data provided by the invention;
fig. 4 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The photovoltaic power generation prediction method based on multi-source data of the present invention is described below with reference to fig. 1 to 2. The method can be realized in the forms of application programs, plug-ins and the like, and specifically comprises the following steps:
s100, acquiring historical data and photovoltaic equipment data, wherein the historical data comprises historical power generation data and historical environment data, and the photovoltaic equipment data comprises equipment three-dimensional data and equipment working data;
s200, constructing a photovoltaic equipment model according to the photovoltaic equipment data, wherein the photovoltaic equipment model comprises an initial prediction model;
s300, training the initial prediction model according to the historical power generation data, the historical environment data and the photovoltaic equipment data until the initial prediction model converges to obtain a photovoltaic power generation prediction model;
and S400, when a prediction instruction is detected, predicting the prediction instruction based on the photovoltaic power generation prediction model to obtain predicted photovoltaic data.
Specifically, the historical data refers to data that can be used to predict photovoltaic power generation, and in this embodiment includes historical power generation data and historical environmental data that affects power generation. Distributed multi-source data are collected from photovoltaic stations, historical power generation data comprise power generation data (KW) of once every 30 minutes, and time intervals corresponding to specific power generation data can be freely adjusted. The historical environment data comprise weather data, radiation intensity data and the like at corresponding moments in the historical power generation data, and the common weather data comprise temperature, humidity and illumination intensity. Photovoltaic device data refers to data related to a photovoltaic device that can be used to implement a simulated state of the photovoltaic device. Including the dimensions, size, inclination angle, degree of aging, etc. of the photovoltaic device.
In one implementation, the photovoltaic device model includes only one initial predictive model, which is a mathematical model of an initial state based on physical principles, which can be subsequently stabilized into a photovoltaic predictive model by training. The initial parameters of the initial predictive model may be determined based on the photovoltaic device data adjustments, e.g., applied to the initial predictive model according to model parameter referenceability trained specifically for the same type of photovoltaic device. The initial predictive model may be in a variety of forms, including convolutional neural networks, time series models based on deep learning, and the like.
After the initial prediction model is obtained, based on the initial prediction model, feature extraction of historical environment data, historical power generation data and photovoltaic equipment data can be achieved, and historical environment features, historical power generation features and photovoltaic equipment features are obtained respectively. A feature is a numeric or classification value used to capture certain aspects or attributes of data. For example, the historical environmental characteristics may include an average, maximum, minimum, or standard deviation of solar radiation, temperature, humidity, wind speed, etc. over each time period, with the historical environmental characteristics and photovoltaic device characteristics taking the form of multidimensional vectors in order to preserve the richness of the characteristics. The historical power generation characteristics may include actual or normalized power output for each time period.
Training is a process of finding the best value of the parameter, minimizing the error between the predicted and actual power generation. The trained predictive model is more accurate and reliable than the initial predictive model. A suitable algorithm is pre-selected to define the structure and loss function of the initial predictive model. And taking the historical environmental characteristics and the photovoltaic equipment characteristics as input data, and taking the historical power generation characteristics as supervision data to construct a supervision learning problem. And then, updating parameters of the initial prediction model by using an optimization method such as gradient descent, random gradient descent, newton method and the like, so that the loss function reaches the minimum value or the training times reach a preset threshold value, and realizing convergence of the initial prediction model to obtain the photovoltaic power generation prediction model. In addition, a portion of the historical data may be retained as a validation set to evaluate the performance of the trained predictive model, such as accuracy, mean square error, correlation coefficient, and the like.
And finally, generating predicted photovoltaic data according to the requirements of the user by using the trained prediction model. First, a prediction instruction is analyzed, and information such as a time zone, a place, and equipment that a user wants to predict is acquired. The corresponding environmental features and photovoltaic device features are then queried or inferred from the information provided by the user and used as input data. And then, the input data is sent into a trained photovoltaic power generation prediction model to obtain output data, namely predicted photovoltaic data. Finally, the predicted photovoltaic data is returned to the user in a suitable format, such as tables, charts, text, and the like.
The scheme is used for understanding and simulating the rules and characteristics of photovoltaic power generation and the performances and states of photovoltaic equipment by collecting and analyzing historical data and photovoltaic equipment data. And then, the accuracy and reliability of the prediction model are improved through training and optimizing parameters of the prediction model, and the capability of adapting to different environmental changes is improved, so that the management and optimization of photovoltaic power generation are facilitated, energy sources are saved, and the environment is protected.
In another implementation manner, the training the initial prediction model according to the historical power generation data, the historical environment data and the photovoltaic device data until the initial prediction model converges, and obtaining the photovoltaic power generation prediction model includes:
the characteristic extraction module is used for respectively extracting the historical environmental characteristics corresponding to the historical environmental data, the historical power generation characteristics corresponding to the historical power generation data and the photovoltaic equipment characteristics corresponding to the photovoltaic equipment data based on the initial prediction model;
predicting based on the historical environmental characteristics to obtain a first predicted value; and predicting based on the photovoltaic equipment characteristics to obtain a second predicted value;
performing feature fusion on the first predicted value and the second predicted value to obtain predicted power generation features;
And adjusting parameters of the initial prediction model according to the predicted power generation characteristics and the historical power generation characteristics until the initial prediction model converges to obtain a photovoltaic power generation prediction model.
In particular, the feature extraction module is part of an initial predictive model that can convert input historical environmental data and photovoltaic device data into a higher level representation of features to facilitate model prediction. In one implementation, the historical environmental characteristics, the photovoltaic power generation data characteristics, and the photovoltaic device data characteristics may all be obtained by convolution calculations. After the characteristics are obtained, the characteristics of the historical environment and the characteristics of the photovoltaic equipment are used for predicting the power generation, so that the characteristics are required to be fused. In this embodiment, the prediction is performed based on the historical environmental characteristics to obtain a first predicted value, and the prediction is performed based on the photovoltaic device characteristics to obtain a second predicted value. The two predictions correspond to two different sets of calculation parameters. And finally, fusing the first predicted value and the second predicted value, and searching the balance between the first predicted value and the second predicted value to obtain more accurate predicted power generation characteristics. The fusion process may perform a weighted average or other operation on the first predicted value and the second predicted value to obtain a comprehensive predicted power generation characteristic. The characteristic can comprehensively consider the influence of environmental factors and equipment factors on the generated energy, so that the prediction accuracy is improved.
And finally, further optimizing the initial prediction model by utilizing the prediction power generation characteristics and the historical power generation characteristics. And using the historical power generation characteristics as target values, using the predicted power generation characteristics and the historical power generation characteristics as input characteristics, and updating the parameters of the initial prediction model again so as to minimize the error between the predicted value and the target value of the model. When the error of the model reaches a smaller threshold value or is not reduced significantly, the initial prediction model is converged, and a more accurate photovoltaic power generation prediction model is obtained. It is noted that training of the model is performed in this way, and when the model is applied later, the predicted result is a feature, and the feature needs to be converted into a numerical value. The conversion mode and the feature extraction can be performed in the opposite steps, and other modes of recovering values according to the features can also be adopted.
Compared with the prior scheme, the scheme adopts the historical power generation characteristics as the object for calculating the loss value instead of the historical power generation data, so that the model overfitting after training can be reduced, and the prediction result is more adaptive. In addition, the power generation characteristic is predicted by adopting a mode of fusing the first predicted value and the second predicted value, which is favorable for the situation of subsequent equipment change or environment change, for example, the equipment is greatly changed and cannot adapt to the model, only the parameters aiming at the characteristics of the photovoltaic equipment are required to be trained, and the subsequent adjustment cost is increased.
In another implementation manner, the feature extraction module based on the initial prediction model extracts the historical environmental feature corresponding to the historical environmental data, the historical power generation feature corresponding to the historical power generation data, and the photovoltaic device feature corresponding to the photovoltaic device data respectively, where the feature extraction module includes:
convolving the historical environmental data to obtain historical environmental characteristics; the method comprises the steps of,
convolving the photovoltaic equipment data to obtain photovoltaic equipment characteristics;
according to the historical power generation data, calculating the attention weight corresponding to each piece of historical power generation data;
and extracting the historical power generation characteristics corresponding to the historical power generation data according to the attention weight.
Specifically, first, according to an initial prediction model, historical environment data, historical power generation data, and photovoltaic device data are input. These data are time series data such as temperature, humidity, power, wind direction, radiation intensity, etc.
And then, carrying out convolution on the historical environment data to obtain the historical environment characteristics. Convolution is a mathematical operation by which local features and spatial relationships in data are extracted by a convolution kernel. Convolution may be implemented using a convolutional neural network. And then, convolving the photovoltaic equipment data to obtain the characteristics of the photovoltaic equipment. This step is similar to the previous step except that the input data is converted to photovoltaic device data.
Then, according to the historical power generation data, the attention weight corresponding to each historical power generation data is calculated. Attention weighting is a mechanism that can be used to measure the importance of different parts of the input data, thereby increasing the relevance between the data. For each input data, a query vector, a key vector and a value vector corresponding to the input data are calculated. And then calculating the attention score corresponding to the input data according to the query vector, the key vector and the value vector. And finally, normalizing each attention score according to the value vector to obtain a corresponding attention weight. One way of calculating the attention score is as follows:
performing dot product on the query vector and the key vector to obtain a vector dot product;
calculating the square root of the dimension of the key vector to obtain a scaling factor;
dividing the vector dot product by the scaling factor to obtain an attention score.
And finally, extracting the historical power generation characteristics corresponding to the historical power generation data according to the attention weight. This step is connected to the previous step, except that the output vector is the historical power generation signature. And for the attention weight, when the historical power generation characteristics are extracted, carrying out characteristic extraction on the historical power generation data with different dimensions according to different attention weights, so as to obtain the corresponding historical power generation characteristics.
The scheme can automatically learn local features and spatial relations in data by utilizing the convolutional neural network, and improves the efficiency and accuracy of feature extraction. And meanwhile, the correlation and the importance in the data are automatically learned by using an attention mechanism, so that the flexibility and the robustness of feature extraction are improved. The accuracy and reliability of the predicted photovoltaic power generation performance and response are improved by such features.
In another implementation manner, the adjusting the parameters of the initial prediction model according to the predicted power generation characteristics and the historical power generation characteristics until the initial prediction model converges, and obtaining the photovoltaic power generation prediction model includes:
calculating a loss value corresponding to the predicted power generation characteristic according to the historical power generation characteristic;
and carrying out internal circulation and external circulation on the photovoltaic equipment model according to a preset circulation rule and the loss value until the photovoltaic equipment model converges to obtain a photovoltaic power generation prediction model, wherein the internal circulation comprises:
and according to the loss value, carrying out parameter adjustment on a feature extraction module and a feature fusion module in the photovoltaic equipment model.
In particular, the loss value may be a numerical indicator, such as a mean square error or a mean absolute error, that measures the difference between the predicted power generation characteristic and the actual power generation characteristic. The smaller the loss value, the closer the predicted result is to the true value, and the more accurate the model. Comparing the predicted power generation characteristics with the historical power generation characteristics one by one, calculating errors between the predicted power generation characteristics and the historical power generation characteristics, and summing or averaging to obtain a loss value, so that the loss value can be obtained.
The preset circulation rule may be a rule for setting parameters such as frequency, interval, etc. of the inner circulation and the outer circulation. The inner loop and the outer loop can be two different levels of training loop processes, the inner loop is used for adjusting parameters of a feature extraction module and a feature fusion module in the model, and the outer loop trains all parameters of the whole model.
According to a preset circulation rule, a batch of samples (namely a group of historical environmental characteristics, photovoltaic equipment characteristics and real power generation characteristics) are selected from historical data according to a certain sequence, and are input into a photovoltaic equipment model to obtain predicted power generation characteristics and loss values. And then, according to the loss value, utilizing a back propagation algorithm to carry out parameter adjustment on a feature extraction module and a feature fusion module in the photovoltaic equipment model so as to reduce the loss value. This is an internal loop. After internal circulation is carried out for 5 times, the same or different samples are adopted, parameters of the whole model are adjusted, the process is repeated until the loss value reaches a preset termination condition or the maximum circulation times, and finally the initial prediction model is trained into the photovoltaic power generation prediction model.
According to the scheme, the internal circulation training process and the external circulation training process are adopted, parameters in the model can be dynamically adjusted, the structure and the performance of the model are optimized, and the stability and the robustness of prediction are improved.
In another implementation, the photovoltaic device model includes not only an initial prediction model that is trainable for prediction, but also a visualization model corresponding to the photovoltaic device. The photovoltaic equipment data comprises equipment three-dimensional data and equipment working data, and the building of the photovoltaic equipment model according to the photovoltaic equipment data comprises the following steps:
constructing an equipment three-dimensional model according to the equipment three-dimensional data in the photovoltaic equipment data;
and carrying out data fusion on the three-dimensional model of the equipment according to the equipment working data to obtain a photovoltaic equipment model.
Specifically, the spatial morphology and layout of the photovoltaic device are described according to the physical structure and parameters of the photovoltaic device, such as size, shape, position, inclination angle, azimuth angle and the like, and a three-dimensional model of the device is constructed. The three-dimensional model of the photovoltaic device is drawn according to the device three-dimensional information using three-dimensional modeling software or tools, such as AutoCAD, sketchUp, and the like, to generate the device three-dimensional model.
The performance and response of the photovoltaic device is then described in terms of the device operational data. The device operating data may include power, battery temperature, etc. From the extracted working data, a mathematical model of the photovoltaic device is built using physical modeling software or tools, such as Matlab, simulink, etc. And then carrying out data fusion on the established mathematical model and the equipment three-dimensional model, namely, carrying out correspondence and matching on variables and parameters in the mathematical model and attributes and characteristics in the three-dimensional model to obtain the fused photovoltaic equipment model. Meanwhile, the equipment working data also comprises a traditional prediction model obtained by training the same type of photovoltaic equipment, namely, initial parameters for initializing the prediction model, such as the same number of convolution kernels, the same step length and the like, can be determined according to the prediction model so as to improve training efficiency.
The three-dimensional visual photovoltaic equipment of above-mentioned scheme directly perceives equipment form and overall arrangement, is convenient for install, debug, overhaul etc.. And the data are fused with the photovoltaic equipment, the performance and response of the equipment are monitored in real time, and then the subsequent photovoltaic power generation prediction model is synchronously updated.
In another implementation, as shown in fig. 2, a visual display may be made of the predicted photovoltaic data based on the visual photovoltaic device model. When the prediction instruction is detected, based on the photovoltaic power generation prediction model, predicting the prediction instruction to obtain predicted photovoltaic data, and then further comprising:
according to the prediction instruction, determining prediction equipment to be predicted;
determining an equipment display model corresponding to the prediction equipment according to the photovoltaic equipment model;
and fusing and displaying the predicted photovoltaic data with the equipment display model.
In particular, the prediction instruction is a command entered by the user for specifying the photovoltaic device and the time of day to be predicted. The equipment to be predicted, i.e. the prediction equipment, is determined according to the prediction instruction. A visual photovoltaic device model has been previously constructed from the photovoltaic device data, and after the prediction device is determined, a device display model that displays the prediction result in the photovoltaic device model may be determined. The device display model refers to a model for directly displaying the appearance and the position of the device display model on a display screen, and is used for prompting the photovoltaic device corresponding to the prediction result.
And fusing and displaying the predicted photovoltaic data with the display model. For example, the system may derive that photovoltaic panel No. 5 of zone A generates 5 kilowatt-hours between 10 and 11 am on tomorrow based on a predictive algorithm. By fusing the data with the display model, the power generation level of the photovoltaic panel can be represented by different colors or intensities on the display screen, so that a user can see the prediction result of the photovoltaic panel at a glance.
In one implementation, the method further comprises:
for each photovoltaic device, when motion information corresponding to the photovoltaic device is detected, calculating updated device information corresponding to the photovoltaic device according to the motion information;
generating an update animation corresponding to the photovoltaic equipment according to the update equipment information; the method comprises the steps of,
and updating the data of the photovoltaic equipment model.
Specifically, in the working process of the photovoltaic device, a plurality of situations such as position adjustment, inclination angle change of the panel and the like may occur. From these parameters, the form of the photovoltaic device after the change, such as the angle of inclination, the placement position, etc., can be calculated. The motion information can be manually input, and can be tracked and collected by a sensor arranged in the photovoltaic equipment.
In order to timely remind a user of the change of the photovoltaic equipment, the embodiment adopts a mode of updating animation. The updating animation is to display visual effects such as position change of the photovoltaic equipment, angle change of the panel, color change of the panel and the like in real time according to the updating equipment information in the three-dimensional model. Updating the animation may help the user intuitively understand the operating state and performance of the photovoltaic device.
Meanwhile, the photovoltaic equipment model is required to be updated, and the data updating means that updated equipment information is stored in a database and compared and analyzed with original equipment information. The data update may help the user monitor and manage the operating conditions and optimization schemes of the photovoltaic device.
Referring to fig. 3, the description of the multi-source data-based photovoltaic power generation prediction apparatus provided by the present invention is provided below, and the multi-source data-based photovoltaic power generation prediction apparatus described below and the multi-source data-based photovoltaic power generation prediction method described above may be referred to correspondingly to each other. The photovoltaic power generation prediction device includes: an acquisition module 310, a construction module 320, a training module 330, and a prediction module 340.
The obtaining module 310 is configured to obtain historical data and photovoltaic device data, where the historical data includes historical power generation data and historical environment data, and the photovoltaic device data includes device three-dimensional data and device working data;
The construction module 320 is configured to construct a photovoltaic device model according to the photovoltaic device data, where the photovoltaic device model includes an initial prediction model;
the training module 330 is configured to train the initial prediction model according to the historical power generation data, the historical environment data and the photovoltaic device data until the initial prediction model converges, so as to obtain a photovoltaic power generation prediction model;
the prediction module 340 is configured to predict, when a prediction instruction is detected, the prediction instruction based on the photovoltaic power generation prediction model, so as to obtain predicted photovoltaic data.
Fig. 4 illustrates a physical schematic diagram of an electronic device, as shown in fig. 4, which may include: processor 410, communication interface (Communications Interface) 420, memory 430 and communication bus 440, wherein processor 410, communication interface 420 and memory 430 communicate with each other via communication bus 440. The processor 410 may invoke logic instructions in the memory 430 to perform a photovoltaic power generation prediction method based on multi-source data, the method comprising:
acquiring historical data and photovoltaic equipment data, wherein the historical data comprises historical power generation data and historical environment data, and the photovoltaic equipment data comprises equipment three-dimensional data and equipment working data;
Constructing a photovoltaic equipment model according to the photovoltaic equipment data, wherein the photovoltaic equipment model comprises an initial prediction model;
training the initial prediction model according to the historical power generation data, the historical environment data and the photovoltaic equipment data until the initial prediction model converges to obtain a photovoltaic power generation prediction model;
when a prediction instruction is detected, the prediction instruction is predicted based on the photovoltaic power generation prediction model, and predicted photovoltaic data are obtained.
Further, the logic instructions in the memory 430 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the method for predicting photovoltaic power generation based on multi-source data provided by the above methods, the method comprising:
acquiring historical data and photovoltaic equipment data, wherein the historical data comprises historical power generation data and historical environment data, and the photovoltaic equipment data comprises equipment three-dimensional data and equipment working data;
constructing a photovoltaic equipment model according to the photovoltaic equipment data, wherein the photovoltaic equipment model comprises an initial prediction model;
training the initial prediction model according to the historical power generation data, the historical environment data and the photovoltaic equipment data until the initial prediction model converges to obtain a photovoltaic power generation prediction model;
when a prediction instruction is detected, the prediction instruction is predicted based on the photovoltaic power generation prediction model, and predicted photovoltaic data are obtained.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (8)
1. The photovoltaic power generation prediction method based on the multi-source data is characterized by comprising the following steps of:
acquiring historical data and photovoltaic equipment data, wherein the historical data comprises historical power generation data and historical environment data, and the photovoltaic equipment data comprises equipment three-dimensional data and equipment working data;
constructing a photovoltaic equipment model according to the photovoltaic equipment data, wherein the photovoltaic equipment model comprises an initial prediction model;
training the initial prediction model according to the historical power generation data, the historical environment data and the photovoltaic equipment data until the initial prediction model converges to obtain a photovoltaic power generation prediction model;
when a prediction instruction is detected, predicting the prediction instruction based on the photovoltaic power generation prediction model to obtain predicted photovoltaic data;
training the initial prediction model according to the historical power generation data, the historical environment data and the photovoltaic equipment data until the initial prediction model converges, wherein obtaining the photovoltaic power generation prediction model comprises the following steps:
the characteristic extraction module is used for respectively extracting the historical environmental characteristics corresponding to the historical environmental data, the historical power generation characteristics corresponding to the historical power generation data and the photovoltaic equipment characteristics corresponding to the photovoltaic equipment data based on the initial prediction model;
Predicting based on the historical environmental characteristics to obtain a first predicted value; the method comprises the steps of,
predicting based on the characteristics of the photovoltaic equipment to obtain a second predicted value;
performing feature fusion on the first predicted value and the second predicted value to obtain predicted power generation features;
according to the predicted power generation characteristics and the historical power generation characteristics, adjusting parameters of the initial prediction model until the initial prediction model converges to obtain a photovoltaic power generation prediction model;
based on the photovoltaic power generation prediction model, predicting the prediction instruction, converting power generation characteristics output by the photovoltaic power generation prediction model into numerical values when predicted photovoltaic data are obtained, and obtaining the predicted photovoltaic data, wherein the conversion mode and the characteristic extraction are performed by adopting opposite steps;
and adjusting parameters of the initial prediction model according to the predicted power generation characteristics and the historical power generation characteristics until the initial prediction model converges, wherein the obtaining the photovoltaic power generation prediction model comprises the following steps:
calculating a loss value corresponding to the predicted power generation characteristic according to the historical power generation characteristic;
according to a preset circulation rule and the loss value, performing internal circulation and external circulation on the initial prediction model until the initial prediction model converges to obtain a photovoltaic power generation prediction model, wherein the internal circulation and the external circulation are training circulation processes of two different layers; the inner loop includes:
According to the loss value, carrying out parameter adjustment on a feature extraction module and a feature fusion module in the initial prediction model;
the outer loop is used for adjusting all parameters in the initial prediction model;
the circulation rule is a rule for setting the frequency and interval of the inner circulation and the outer circulation.
2. The method according to claim 1, wherein the feature extraction module based on the initial prediction model extracts the historical environmental feature corresponding to the historical environmental data, the historical power generation feature corresponding to the historical power generation data, and the photovoltaic device feature corresponding to the photovoltaic device data, respectively, including:
convolving the historical environmental data to obtain historical environmental characteristics; the method comprises the steps of,
convolving the photovoltaic equipment data to obtain photovoltaic equipment characteristics;
according to the historical power generation data, calculating the attention weight corresponding to each piece of historical power generation data;
and extracting the historical power generation characteristics corresponding to the historical power generation data according to the attention weight.
3. The method for predicting photovoltaic power generation based on multi-source data according to any one of claims 1-2, wherein constructing a photovoltaic device model from the photovoltaic device data comprises:
Constructing an equipment three-dimensional model according to the equipment three-dimensional data in the photovoltaic equipment data;
and carrying out data fusion on the three-dimensional model of the equipment according to the equipment working data to obtain a photovoltaic equipment model.
4. The method for predicting photovoltaic power generation based on multi-source data according to claim 3, wherein when a prediction instruction is detected, the prediction instruction is predicted based on the photovoltaic power generation prediction model, and after obtaining predicted photovoltaic data, the method further comprises:
according to the prediction instruction, determining prediction equipment to be predicted;
determining an equipment display model corresponding to the prediction equipment according to the photovoltaic equipment model;
and fusing and displaying the predicted photovoltaic data with the equipment display model.
5. A method of photovoltaic power generation prediction based on multi-source data according to claim 3, further comprising:
for each photovoltaic device, when motion information corresponding to the photovoltaic device is detected, calculating updated device information corresponding to the photovoltaic device according to the motion information;
generating an update animation corresponding to the photovoltaic equipment according to the update equipment information; the method comprises the steps of,
And updating the data of the photovoltaic equipment model.
6. A photovoltaic power generation prediction device based on multi-source data, comprising:
the photovoltaic device comprises an acquisition module, a storage module and a storage module, wherein the acquisition module is used for acquiring historical data and photovoltaic device data, the historical data comprise historical power generation data and historical environment data, and the photovoltaic device data comprise device three-dimensional data and device working data;
the construction module is used for constructing a photovoltaic equipment model according to the photovoltaic equipment data, wherein the photovoltaic equipment model comprises an initial prediction model;
the training module is used for training the initial prediction model according to the historical power generation data, the historical environment data and the photovoltaic equipment data until the initial prediction model converges to obtain a photovoltaic power generation prediction model;
the prediction module is used for predicting the prediction instruction based on the photovoltaic power generation prediction model when the prediction instruction is detected, so as to obtain predicted photovoltaic data;
training the initial prediction model according to the historical power generation data, the historical environment data and the photovoltaic equipment data until the initial prediction model converges, wherein obtaining the photovoltaic power generation prediction model comprises the following steps:
The characteristic extraction module is used for respectively extracting the historical environmental characteristics corresponding to the historical environmental data, the historical power generation characteristics corresponding to the historical power generation data and the photovoltaic equipment characteristics corresponding to the photovoltaic equipment data based on the initial prediction model;
predicting based on the historical environmental characteristics to obtain a first predicted value; the method comprises the steps of,
predicting based on the characteristics of the photovoltaic equipment to obtain a second predicted value;
performing feature fusion on the first predicted value and the second predicted value to obtain predicted power generation features;
according to the predicted power generation characteristics and the historical power generation characteristics, adjusting parameters of the initial prediction model until the initial prediction model converges to obtain a photovoltaic power generation prediction model;
based on the photovoltaic power generation prediction model, predicting the prediction instruction, converting power generation characteristics output by the photovoltaic power generation prediction model into numerical values when predicted photovoltaic data are obtained, and obtaining the predicted photovoltaic data, wherein the conversion mode and the characteristic extraction are performed by adopting opposite steps;
and adjusting parameters of the initial prediction model according to the predicted power generation characteristics and the historical power generation characteristics until the initial prediction model converges, wherein the obtaining the photovoltaic power generation prediction model comprises the following steps:
Calculating a loss value corresponding to the predicted power generation characteristic according to the historical power generation characteristic;
according to a preset circulation rule and the loss value, performing internal circulation and external circulation on the initial prediction model until the initial prediction model converges to obtain a photovoltaic power generation prediction model, wherein the internal circulation and the external circulation are training circulation processes of two different layers; the inner loop includes:
according to the loss value, carrying out parameter adjustment on a feature extraction module and a feature fusion module in the initial prediction model;
the outer loop is used for adjusting all parameters in the initial prediction model;
the circulation rule is a rule for setting the frequency and interval of the inner circulation and the outer circulation.
7. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the multi-source data based photovoltaic power generation prediction method of any of claims 1 to 5 when the computer program is executed.
8. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the multi-source data based photovoltaic power generation prediction method of any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310984296.4A CN116722545B (en) | 2023-08-07 | 2023-08-07 | Photovoltaic power generation prediction method based on multi-source data and related equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310984296.4A CN116722545B (en) | 2023-08-07 | 2023-08-07 | Photovoltaic power generation prediction method based on multi-source data and related equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116722545A CN116722545A (en) | 2023-09-08 |
| CN116722545B true CN116722545B (en) | 2024-01-16 |
Family
ID=87871941
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310984296.4A Active CN116722545B (en) | 2023-08-07 | 2023-08-07 | Photovoltaic power generation prediction method based on multi-source data and related equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116722545B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117591814A (en) * | 2024-01-19 | 2024-02-23 | 北京志翔科技股份有限公司 | Data restoration method, device and equipment based on photovoltaic envelope |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111210095A (en) * | 2020-03-12 | 2020-05-29 | 深圳前海微众银行股份有限公司 | Power generation amount prediction method, device, equipment and computer readable storage medium |
| CN115036922A (en) * | 2022-08-10 | 2022-09-09 | 四川中电启明星信息技术有限公司 | Distributed photovoltaic power generation electric quantity prediction method and system |
| CN115618922A (en) * | 2022-09-28 | 2023-01-17 | 国网河北省电力有限公司电力科学研究院 | Photovoltaic power prediction method and device, photovoltaic power generation system and storage medium |
| CN115689062A (en) * | 2022-12-30 | 2023-02-03 | 浙江工业大学 | Photovoltaic output power prediction method based on rapid online migration neural network |
| CN116093932A (en) * | 2023-02-09 | 2023-05-09 | 浪潮云信息技术股份公司 | Photovoltaic power generation prediction method and system based on fusion information |
| CN116435998A (en) * | 2023-04-17 | 2023-07-14 | 石家庄科林电气股份有限公司 | Prediction method of photovoltaic power generation power |
| 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 |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111539550B (en) * | 2020-03-13 | 2023-08-01 | 远景智能国际私人投资有限公司 | Method, device, equipment and storage medium for determining working state of photovoltaic array |
| 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 |
-
2023
- 2023-08-07 CN CN202310984296.4A patent/CN116722545B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111210095A (en) * | 2020-03-12 | 2020-05-29 | 深圳前海微众银行股份有限公司 | Power generation amount prediction method, device, equipment and computer readable storage medium |
| CN115036922A (en) * | 2022-08-10 | 2022-09-09 | 四川中电启明星信息技术有限公司 | Distributed photovoltaic power generation electric quantity prediction method and system |
| CN115618922A (en) * | 2022-09-28 | 2023-01-17 | 国网河北省电力有限公司电力科学研究院 | Photovoltaic power prediction method and device, photovoltaic power generation system and storage medium |
| CN115689062A (en) * | 2022-12-30 | 2023-02-03 | 浙江工业大学 | Photovoltaic output power prediction method based on rapid online migration neural network |
| CN116093932A (en) * | 2023-02-09 | 2023-05-09 | 浪潮云信息技术股份公司 | Photovoltaic power generation prediction method and system based on fusion information |
| 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 |
| CN116435998A (en) * | 2023-04-17 | 2023-07-14 | 石家庄科林电气股份有限公司 | Prediction method of photovoltaic power generation power |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116722545A (en) | 2023-09-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5888640B2 (en) | Photovoltaic power generation prediction apparatus, solar power generation prediction method, and solar power generation prediction program | |
| Ge et al. | A hybrid model for short-term PV output forecasting based on PCA-GWO-GRNN | |
| CN103390116B (en) | Use the photovoltaic power station power generation power forecasting method of stepping mode | |
| CN113128793A (en) | Photovoltaic power combination prediction method and system based on multi-source data fusion | |
| CN110866592B (en) | Model training method, device, energy efficiency prediction method, device and storage medium | |
| CN116722545B (en) | Photovoltaic power generation prediction method based on multi-source data and related equipment | |
| Liu et al. | Heating load forecasting for combined heat and power plants via strand-based LSTM | |
| CN110942205A (en) | Short-term photovoltaic power generation power prediction method based on HIMVO-SVM | |
| Weng et al. | An evolutionary Nelder–Mead slime mould algorithm with random learning for efficient design of photovoltaic models | |
| Aliberti et al. | Forecasting Short-term Solar Radiation for Photovoltaic Energy Predictions. | |
| JP6086875B2 (en) | Power generation amount prediction device and power generation amount prediction method | |
| CN112862274A (en) | WRF-LES and Deepar combined wind power prediction method | |
| CN114970952A (en) | Photovoltaic output short-term prediction method and system considering environmental factors | |
| CN112381282A (en) | Photovoltaic power generation power prediction method based on width learning system | |
| CN113762591B (en) | Short-term electric quantity prediction method and system based on GRU and multi-core SVM countermeasure learning | |
| CN112836876B (en) | Power distribution network line load prediction method based on deep learning | |
| CN117039894B (en) | Photovoltaic power short-term prediction method and system based on improved dung beetle optimization algorithm | |
| Kumar et al. | Forecasting indoor temperature for smart buildings with ARIMA, SARIMAX, and LSTM: A fusion approach | |
| CN116307287B (en) | Prediction method, system and prediction terminal for effective period of photovoltaic power generation | |
| CN117313795A (en) | Intelligent building energy consumption prediction method based on improved DBO-LSTM | |
| CN116796194A (en) | IDBO-KELM-BiGRU neural network-based active power virtual collection method for distributed photovoltaic power station | |
| CN115796327A (en) | Wind power interval prediction method based on VMD (vertical vector decomposition) and IWOA-F-GRU (empirical mode decomposition) -based models | |
| Singh et al. | Grey Wolf Optimization Based CNN-LSTM Network for the Prediction of Energy Consumption in Smart Home Environment | |
| CN112821456B (en) | Distributed source-storage-load matching method and device based on transfer learning | |
| CN115034432A (en) | Wind speed prediction method for wind generating set of wind power plant |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |