CN115619047B - Photovoltaic power prediction method and device and electronic equipment - Google Patents
Photovoltaic power prediction method and device and electronic equipment Download PDFInfo
- Publication number
- CN115619047B CN115619047B CN202211629198.0A CN202211629198A CN115619047B CN 115619047 B CN115619047 B CN 115619047B CN 202211629198 A CN202211629198 A CN 202211629198A CN 115619047 B CN115619047 B CN 115619047B
- Authority
- CN
- China
- Prior art keywords
- weather
- data
- photovoltaic power
- target
- power
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- 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
-
- 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
- H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
- H02J3/381—Dispersed generators
-
- 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
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- 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
- H02J2300/24—The renewable source being solar energy of photovoltaic origin
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Power Engineering (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Operations Research (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Game Theory and Decision Science (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Entrepreneurship & Innovation (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Quality & Reliability (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention provides a photovoltaic power prediction method, a photovoltaic power prediction device and electronic equipment, and relates to the technical field of photovoltaic application, wherein the photovoltaic power prediction method, the photovoltaic power prediction device and the electronic equipment can acquire adjacent time points of a target time point and an extracted target time point and weather data of the adjacent time points; searching target weather data matched with the weather data and historical photovoltaic power corresponding to the target weather data in a pre-stored weather power data table; constructing a weather characteristic sequence according to the target weather data and the historical photovoltaic power; inputting the weather characteristic sequence into a pre-established prediction model, and outputting prediction power corresponding to the weather characteristic sequence through the prediction model; and the predicted power is determined as the predicted photovoltaic power of the target time point, so that the photovoltaic power prediction with smaller time granularity can be realized, the influence of fluctuation in a long period of time is effectively avoided, the prediction error caused by severe vibration in the period of time is reduced, and the accuracy of the photovoltaic power generation power prediction is improved.
Description
Technical Field
The invention relates to the technical field of photovoltaic application, in particular to a photovoltaic power prediction method and device and electronic equipment.
Background
The photovoltaic power generation power is usually affected by weather condition images and has obvious intermittent fluctuation characteristics, so that the safe and stable operation of a power grid is impacted to a certain extent by large-scale photovoltaic power generation access. And with the increase of the proportion of renewable energy sources such as wind power, photovoltaic and the like in various regions, the phenomena of wind abandonment and light abandonment are further increased.
The photovoltaic power generation power prediction is one of key technologies for solving the problem, so that the research on the photovoltaic power station power generation power prediction method and system has important academic and application values. Therefore, how to accurately predict the power of the photovoltaic power station becomes a research hotspot in recent years, and is popular among domestic and foreign scholars.
At present, most of common prediction methods are photovoltaic prediction based on similar days, but the daily fluctuation amplitude of photovoltaic power is large, so that spikes easily appear, especially the shock is severe in the noon and afternoon time period, so that the similarity error calculated by day is large, and the accuracy of photovoltaic power generation power prediction is seriously reduced.
Disclosure of Invention
In view of the above, the present invention provides a method, an apparatus and an electronic device for predicting photovoltaic power, so as to alleviate the above technical problems and improve the accuracy of photovoltaic power prediction.
In a first aspect, an embodiment of the present invention provides a photovoltaic power prediction method, where the method includes: acquiring a target time point, wherein the target time point is a time point at which photovoltaic power needs to be predicted; extracting adjacent time points of the target time point and weather data of the adjacent time points; searching target weather data matched with the weather data and historical photovoltaic power corresponding to the target weather data in a pre-stored weather power data table; constructing a weather characteristic sequence according to the target weather data and the historical photovoltaic power; inputting the weather characteristic sequence into a pre-established prediction model, and outputting prediction power corresponding to the weather characteristic sequence through the prediction model; the prediction model is obtained by training weather characteristic data, wherein the weather characteristic data comprises the weather data and historical photovoltaic power corresponding to the weather data; determining the predicted power as the predicted photovoltaic power of the target time point.
With reference to the first aspect, an embodiment of the present invention provides a first possible implementation manner of the first aspect, where the step of extracting neighboring time points of the target time point includes: determining a target unit time containing the target time point according to a predefined unit time; determining a previous unit time adjacent to the target unit time as the adjacent time point to the target time point.
With reference to the first aspect, an embodiment of the present invention provides a second possible implementation manner of the first aspect, where the weather power data table includes a plurality of historical photovoltaic powers and weather data corresponding to each historical photovoltaic power; the method comprises the following steps of searching target weather data matched with the weather data and historical photovoltaic power corresponding to the target weather data in a pre-stored weather power data table, wherein the steps comprise: calculating the distance between the weather data of the adjacent time points and the weather data in the weather power data table according to a preset distance algorithm; and determining the weather data corresponding to the distance smaller than a preset distance threshold value in the weather power data table as the target weather data, and searching the historical photovoltaic power corresponding to the target weather data in the weather power data table.
With reference to the first aspect, an embodiment of the present invention provides a third possible implementation manner of the first aspect, where the method further includes: acquiring a pre-stored historical weather data table containing a plurality of weather data and a historical power data table corresponding to the historical weather data table and containing a plurality of historical photovoltaic power; assembling the weather data and the historical photovoltaic power according to a preset time interval to obtain weather characteristic data containing the weather data and the historical photovoltaic power; adding the weather characteristic data to a pre-established training set; and training an initial prediction network based on the training set to obtain the prediction model.
With reference to the third possible implementation manner of the first aspect, an embodiment of the present invention provides a fourth possible implementation manner of the first aspect, where the time interval includes a first time interval corresponding to the historical photovoltaic power and a second time interval corresponding to the weather data, and the first time interval is smaller than the second time interval; assembling the weather data and the historical photovoltaic power according to a preset time interval to obtain weather characteristic data comprising the weather data and the historical photovoltaic power, wherein the step comprises the following steps of: and combining the weather data and the historical photovoltaic power according to the time sequence corresponding to the first time interval, so that weather characteristic data comprising the weather data and the historical photovoltaic power is generated every other first time interval, and the weather data of the weather characteristic data in the second time interval are the same.
With reference to the fourth possible implementation manner of the first aspect, an embodiment of the present invention provides a fifth possible implementation manner of the first aspect, where the training the initial prediction network based on the training set includes: for each weather characteristic data, calculating an intermediate characteristic matched with the weather characteristic data in the weather power data table based on the weather data and the historical photovoltaic power; assembling the intermediate features and the weather feature data to obtain feature vectors corresponding to the weather feature data; and training the initial prediction network by using the feature vector to obtain the prediction model.
With reference to the fifth possible implementation manner of the first aspect, an embodiment of the present invention provides a sixth possible implementation manner of the first aspect, where the step of calculating an intermediate feature in the weather power data table, which is matched with the weather feature data, based on the weather data and the historical photovoltaic power includes: acquiring weather data contained in the weather characteristic data; searching at least one target weather data, the distance between which and the weather data is smaller than a preset distance threshold value, in the weather power data table according to a preset distance algorithm; extracting historical photovoltaic power corresponding to at least one target weather data in the weather power data table; and calculating the historical photovoltaic power mean parameter, and determining the mean parameter as an intermediate feature matched with the weather feature data.
In a second aspect, an embodiment of the present invention further provides a photovoltaic power prediction apparatus, where the apparatus includes: the device comprises an acquisition module, a power generation module and a power generation module, wherein the acquisition module is used for acquiring a target time point, and the target time point is a time point needing to predict photovoltaic power; the extraction module is used for extracting adjacent time points of the target time point and weather data of the adjacent time points; the searching module is used for searching target weather data matched with the weather data and historical photovoltaic power corresponding to the target weather data in a pre-stored weather power data table; the building module is used for building a weather characteristic sequence according to the target weather data and the historical photovoltaic power; the forecasting module is used for inputting the weather characteristic sequence into a pre-established forecasting model and outputting forecasting power corresponding to the weather characteristic sequence through the forecasting model; the method comprises the steps that a prediction model is obtained by training weather characteristic data, wherein the weather characteristic data comprise the weather data and historical photovoltaic power corresponding to the weather data; a determination module for determining the predicted power as the predicted photovoltaic power of the target time point.
In a third aspect, an embodiment of the present invention further provides an electronic device, including a processor and a memory, where the memory stores machine-executable instructions capable of being executed by the processor, and the processor executes the machine-executable instructions to implement the method in the first aspect.
In a fourth aspect, embodiments of the present invention also provide a machine-readable storage medium storing machine-executable instructions that, when invoked and executed by a processor, cause the processor to implement the method of the first party.
The embodiment of the invention has the following beneficial effects:
according to the photovoltaic power prediction method, the photovoltaic power prediction device and the electronic equipment, the adjacent time points of the target time point extracted by the target time point and the weather data of the adjacent time points can be obtained; searching target weather data matched with the weather data and historical photovoltaic power corresponding to the target weather data in a pre-stored weather power data table; constructing a weather characteristic sequence according to the target weather data and the historical photovoltaic power; inputting the weather characteristic sequence into a pre-established prediction model, and outputting prediction power corresponding to the weather characteristic sequence through the prediction model; and the predicted photovoltaic power of the target time point is predicted based on the weather data of the adjacent time points and the historical photovoltaic power, so that the prediction of the photovoltaic power of smaller time granularity can be realized, the influence of fluctuation in a long period of time can be effectively avoided, the prediction error caused by severe oscillation in the period of time is reduced, and the accuracy of the prediction of the photovoltaic power generation power is effectively improved.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flowchart of a photovoltaic power prediction method according to an embodiment of the present invention;
fig. 2 is a flowchart of another photovoltaic power prediction method according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a photovoltaic power prediction apparatus according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present invention clearer, 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, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
At present, a common photovoltaic power prediction method includes similar day photovoltaic prediction based on a clustering algorithm, and in this way, historical weather data can be subjected to clustering analysis according to predicted day weather data, a similar day training sample set is divided, then the similar day sample set is trained, and finally, weather prediction data to be predicted is input into a prediction model corresponding to the similar day for prediction. However, the photovoltaic power daily fluctuation range is large, spikes easily occur, especially the oscillation is severe in the noon and afternoon time periods, the similarity error calculated by day is large, and in order to reduce the error of calculating the similarity, a large amount of weather data and historical data are required, so that the calculation amount is increased sharply.
In addition, similar day photovoltaic prediction based on the clustering algorithm and similar day photovoltaic prediction based on wavelet transformation are also provided, the prediction method can select similar day data from historical data of a plurality of days before the day to be predicted and uses the similar day data as a training data set, original photovoltaic power generation data sequences are decomposed into a low-frequency approximate signal and a plurality of high-frequency detail signals by using wavelet transformation, a prediction model is adopted to learn the training data set after wavelet transformation, and prediction models are respectively trained on high-frequency components and low-frequency components. However, the decomposition scale of the method is difficult to control, an optimal value needs to be obtained through multiple experiments, too much or too little decomposition has a large influence on a prediction result, too much decomposition is obtained, the complexity of the algorithm is increased, the prediction time is too long, the decomposition is not enough, the subsequence cannot effectively reflect the characteristics of the original sequence, the decomposed component is wholly similar to the component with the prediction, and the power fluctuates greatly in a certain time period in the component, so that the acquaintance error is also large.
Therefore, in the related art, photovoltaic prediction based on similar subsequences is also available, in this way, the historical power sequence is divided into a plurality of historical power subsequences according to the length of the target power sequence, then a power derivation method is used for determining the similar sequences similar to the target power sequence in change trend from the plurality of historical power subsequences, and the photovoltaic output power value at the next moment is predicted according to the determined similar sequences. However, in this way, it is difficult to find a suitable target power sequence length, the sub-sequence is too long and too short, which has a large influence on the prediction result, and the whole similar sub-sequence is similar to the sub-sequence to be predicted, but the error in a specific time period in the sub-sequence is large, so that it is difficult to improve the accuracy of photovoltaic power generation power prediction.
Based on this, the photovoltaic power prediction method, the photovoltaic power prediction device and the electronic device provided by the embodiment of the invention can effectively alleviate the technical problems.
For the convenience of understanding the embodiment, a method for predicting photovoltaic power disclosed by the embodiment of the present invention will be described in detail first.
In a possible implementation manner, an embodiment of the present invention provides a photovoltaic power prediction method, specifically, a flowchart of a photovoltaic power prediction method as shown in fig. 1, where the method includes the following steps:
step S102, acquiring a target time point;
the target time point in the embodiment of the invention is a time point needing to predict the photovoltaic power;
step S104, extracting adjacent time points of the target time point and weather data of the adjacent time points;
in practical applications, the target time point may be a time, such as nine am, nine tomorrow, or the like, or a time period, such as 11 to 12 pm, or the like, and the adjacent time point may be an adjacent time of the target time point, or an adjacent time period, where there is a preset time interval between the target time point and the adjacent time point, the time interval may be a time interval in minutes, a time interval in hours, or a time interval in days, or the like, and the target time point is a time point at which the photovoltaic power needs to be predicted, so the target time point is usually a future time point. The adjacent time point may be a future time point or an already-passed time point, for example, assuming that the target time point is nine am, the current time is eight and a half am, and the preset time interval is one hour, so that the adjacent time point is 8 am. The specific time point may be set according to actual use conditions.
The time interval is also generally referred to as a unit time, and for example, when the target time point is 9 points, the adjacent time point is 8 points and 55 minutes. If one day is taken as a unit time, when the target time point is 9 points, the adjacent time point is 9 points of the previous day, therefore, in the step S104 in the embodiment of the present invention, when the adjacent time point of the target time point is extracted, the target unit time including the target time point may be determined according to the predefined unit time; the previous unit time adjacent to the target unit time is determined as an adjacent time point to the target time point.
Further, when the adjacent time point is a future time point, the weather data thereof may be obtained by a weather forecast system, and if the adjacent time point is an elapsed time point, the weather data of the adjacent time point may be directly obtained from the weather data record, specifically taking an actual use condition as a standard, which is not limited in this embodiment of the present invention.
Step S106, searching target weather data matched with the weather data and historical photovoltaic power corresponding to the target weather data in a pre-stored weather power data table;
step S108, a weather characteristic sequence is constructed according to the target weather data and the historical photovoltaic power;
in practical use, the weather power data table is actually a historical photovoltaic power record data table, and therefore, generally includes historical photovoltaic power, where photovoltaic power in the embodiment of the present invention refers to photovoltaic power generation power. Since the photovoltaic power generation power is generally greatly influenced by weather, an actual weather condition is usually recorded in an actual photovoltaic power recording process, so that the historical photovoltaic power and the weather data corresponding to the historical photovoltaic power in a past period of time are usually recorded in the weather power data table in the embodiment of the present invention, and therefore, in step S106, the target weather data matched with the weather data and the historical photovoltaic power corresponding to the target weather data can be found in the weather power data table, and further, in step S108, the target weather data and the historical photovoltaic power are assembled to construct a weather feature sequence.
Step S110, inputting the weather characteristic sequence into a pre-established prediction model, and outputting prediction power corresponding to the weather characteristic sequence through the prediction model;
the prediction model in the embodiment of the invention is obtained by training weather characteristic data, and the weather characteristic data comprises weather data and historical photovoltaic power corresponding to the weather data;
and step S112, determining the predicted power as the predicted photovoltaic power of the target time point.
According to the photovoltaic power prediction method provided by the embodiment of the invention, the adjacent time points of the target time point extracted by the target time point and the weather data of the adjacent time points can be obtained; searching target weather data matched with the weather data and historical photovoltaic power corresponding to the target weather data in a pre-stored weather power data table; constructing a weather characteristic sequence according to the target weather data and the historical photovoltaic power; inputting the weather characteristic sequence into a pre-established prediction model, and outputting prediction power corresponding to the weather characteristic sequence through the prediction model; and the predicted photovoltaic power of the target time point is predicted based on the weather data of the adjacent time points and the historical photovoltaic power, so that the prediction of the photovoltaic power of smaller time granularity can be realized, the influence of fluctuation in a long period of time can be effectively avoided, the prediction error caused by severe oscillation in the period of time is reduced, and the accuracy of the prediction of the photovoltaic power generation power is effectively improved.
In actual use, the weather data typically includes at least one of the following weather characteristics: the method comprises the following steps of light radiation intensity, weather types (such as sunny days, cloudy days, rainy days and the like), wind direction, wind speed, air temperature, humidity, air pressure, cloud cover, precipitation probability and the like, and when the weather data are recorded, normalization dimension processing can be carried out on a plurality of data so as to normalize the plurality of weather data to the same dimension for calculation.
Further, the weather power data table includes a plurality of historical photovoltaic powers and the weather data corresponding to each historical photovoltaic power, and when the target weather data is searched in the weather power data table, the target weather data is usually searched through a distance parameter, for example, a euclidean distance, a normalized euclidean distance, or a weighted normalized euclidean distance may be calculated, where the normalized euclidean distance is a problem that distribution of each dimension component of data in the euclidean distance is not uniform on the premise that all features are assumed to be equally important, but contribution degrees of different features to a prediction result are very different.
For the convenience of understanding, on the basis of the above fig. 1, the embodiment of the present invention further provides another photovoltaic power prediction method, and the photovoltaic power prediction process is further described, as shown in a flowchart of another photovoltaic power prediction method shown in fig. 2, including the following steps:
step S202, acquiring a target time point;
the target time point in the embodiment of the invention is a time point needing to predict the photovoltaic power;
step S204, extracting adjacent time points of the target time point and weather data of the adjacent time points;
step S206, calculating the distance between the weather data of the adjacent time points and the weather data in the weather power data table according to a preset distance algorithm;
step S208, determining weather data corresponding to the distance smaller than a preset distance threshold value in a weather power data table as target weather data, and searching historical photovoltaic power corresponding to the target weather data in the weather power data table;
the distance calculated in the embodiment of the present invention is the weighted normalized euclidean distance, and in other embodiments, the euclidean distance, the normalized euclidean distance, and the like may also be adopted, which may be specifically set according to an actual use situation, and the embodiment of the present invention is not limited thereto.
Step S210, a weather characteristic sequence is constructed according to target weather data and historical photovoltaic power;
step S212, inputting the weather characteristic sequence into a pre-established prediction model, and outputting the prediction power corresponding to the weather characteristic sequence through the prediction model;
step S214, the predicted power is determined as the predicted photovoltaic power of the target time point.
The prediction model in the embodiment of the invention is obtained by training weather characteristic data, wherein the weather characteristic data comprises weather data and historical photovoltaic power corresponding to the weather data;
in practical use, the weather characteristic data is characteristic data obtained by data assembly of weather data and historical photovoltaic power, and therefore, in the embodiment of the invention, when the prediction model is trained, a training set containing the weather characteristic data is usually constructed, and then the prediction model is trained by using the training set.
Therefore, the photovoltaic power prediction method in the embodiment of the present invention further includes the following steps:
(1) Acquiring a pre-stored historical weather data table containing a plurality of weather data and a historical power data table corresponding to the historical weather data table and containing a plurality of historical photovoltaic powers;
(2) Assembling the weather data and the historical photovoltaic power according to a preset time interval to obtain weather characteristic data comprising the weather data and the historical photovoltaic power;
(3) Adding weather characteristic data to a pre-established training set; and training the initial prediction network based on the training set to obtain a prediction model.
In actual use, since the change of the weather data is relatively slow, in the photovoltaic power recording process, the time interval for recording the weather data is greater than the time interval for recording the photovoltaic data, and therefore, the time interval in (2) above includes a first time interval corresponding to the historical photovoltaic power and a second time interval corresponding to the weather data, and the first time interval is smaller than the second time interval.
Moreover, in the above (2), when the weather data and the historical photovoltaic power are assembled, the weather data and the historical photovoltaic power may be combined according to a time sequence corresponding to the first time interval, so that a weather feature data including the weather data and the historical photovoltaic power is generated every other first time interval, and the weather data of the weather feature data in the second time interval is the same.
For example, in the data recording process, one data point in 5 minutes is used for historical photovoltaic power, one data point in 30 minutes is used for weather data, that is, the first time interval is 5 minutes, and the second time interval is 30 minutes, when assembling weather characteristic data, the weather data and the historical photovoltaic power can be assembled according to the minutes, that is, the two data of the weather data and the historical photovoltaic power are connected and combined according to the 5 minute time point, at this time, one weather data can correspond to 6 historical photovoltaic powers, after assembling, in one second time interval, 6 weather characteristic data can be obtained, and the weather data in the 6 weather characteristic data are the same, but the corresponding historical photovoltaic powers are different.
After the weather characteristic data obtained by the method is added to a pre-established training set, the initial prediction network can be trained, and before training, processes such as abnormal value processing and normalization processing can be usually performed on the data of the training set, so that the data of the training set can meet the training requirements.
Further, when the initial prediction network is trained based on the training set, for each weather characteristic data, an intermediate characteristic matched with the weather characteristic data in the weather power data table can be calculated based on the weather data and historical photovoltaic power; then, assembling the intermediate features and the weather feature data to obtain feature vectors corresponding to the weather feature data; and then training the initial prediction network by using the feature vector to obtain a prediction model.
The process of calculating the intermediate features is actually a process of performing data expansion on the weather feature data in the training set, and the weather feature data can be further expanded based on the weather power data table to obtain a more extensive training set.
Specifically, when calculating the intermediate feature, the weather data included in the weather feature data may be acquired; searching at least one target weather data, the distance between which and the weather data is less than a preset distance threshold value, in a weather power data table according to a preset distance algorithm; extracting historical photovoltaic power corresponding to at least one target weather data in a weather power data table; and calculating a historical photovoltaic power mean parameter, and determining the mean parameter as an intermediate feature matched with the weather feature data.
In practical use, the intermediate features are data similar to the target weather data searched by distance, and besides the above manner, the intermediate features may be calculated in a time window manner, for example, in consideration of fluctuation of weather, taking 30 minutes as an example of a unit time, fluctuation rates of weather feature data of a previous unit time and a next unit time, a fluctuation average value, and the like may be calculated as one intermediate feature in the training set.
Further, when assembling the feature vectors corresponding to the weather feature data, assembling and splicing the weather data, the intermediate features and the historical photovoltaic power of the weather data, using the obtained training set as the input of the prediction model, firstly performing prediction model training, and then inputting the current weather feature sequence into the trained prediction model for power prediction.
In practical use, the prediction model in the embodiment of the invention can adopt an ensemble learning catboost model to train the prediction model, and the prediction model obtained based on the method can be used for calculating similarity according to minutes under the conditions that the daily fluctuation amplitude of photovoltaic power is large and spikes easily appear, especially the oscillation in the noon and afternoon time periods is severe, so that the influence of the fluctuation in a long time period is avoided, and the large error caused by the severe oscillation in the time period is reduced. Meanwhile, a distance calculation mode, such as a weighted standardized Euclidean distance calculation mode, is adopted, so that a prediction method of weighted similar minutes can be realized, meteorological fluctuation rate factors are considered when target weather data are calculated, and the problem of severe photovoltaic power oscillation is solved from the front. Compared with the traditional similar day calculation, the similarity is calculated according to minutes, the granularity is finer, the training data set is easier to obtain, more samples can be obtained in the same time period, and the prediction precision is more accurate.
Further, on the basis of the above embodiments, an embodiment of the present invention further provides a photovoltaic power prediction apparatus, as shown in fig. 3, which includes the following structures:
an obtaining module 30, configured to obtain a target time point, where the target time point is a time point at which photovoltaic power needs to be predicted;
an extracting module 32, configured to extract neighboring time points of the target time point and weather data of the neighboring time points;
the searching module 34 is configured to search, in a pre-stored weather power data table, target weather data matched with the weather data and historical photovoltaic power corresponding to the target weather data;
a building module 36, configured to build a weather feature sequence according to the target weather data and the historical photovoltaic power;
the prediction module 38 is configured to input the weather feature sequence into a pre-established prediction model, and output a prediction power corresponding to the weather feature sequence through the prediction model; the prediction model is obtained by training weather characteristic data, wherein the weather characteristic data comprises the weather data and historical photovoltaic power corresponding to the weather data;
a determining module 39, configured to determine the predicted power as the predicted photovoltaic power of the target time point.
The photovoltaic power prediction device provided by the embodiment of the invention has the same technical characteristics as the photovoltaic power prediction method provided by the embodiment, so that the same technical problems can be solved, and the same technical effects can be achieved.
Further, an embodiment of the present invention further provides an electronic device, which includes a processor and a memory, where the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the method described above.
Further, embodiments of the present invention also provide a machine-readable storage medium storing machine-executable instructions, which when called and executed by a processor, cause the processor to implement the above method.
An embodiment of the present invention further provides an electronic device, referring to a schematic structural diagram of an electronic device shown in fig. 4, where the electronic device includes a processor 40 and a memory 41, the memory 41 stores machine-executable instructions capable of being executed by the processor 40, and the processor 40 executes the machine-executable instructions to implement the method.
Further, the electronic device shown in fig. 4 further includes a bus 42 and a communication interface 43, and the processor 40, the communication interface 43 and the memory 41 are connected through the bus 42.
The Memory 41 may include a high-speed Random Access Memory (RAM) and may also include a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. The communication connection between the network element of the system and at least one other network element is realized through at least one communication interface 43 (which may be wired or wireless), and the internet, a wide area network, a local network, a metropolitan area network, etc. may be used. The bus 42 may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, an EISA (Enhanced Industry Standard Architecture) bus, or the like. The above-mentioned bus may be classified into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one double-headed arrow is shown in FIG. 4, but this does not indicate only one bus or one type of bus.
The processor 40 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware or instructions in the form of software in the processor 40. The Processor 40 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the device can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component. The various methods, steps and logic blocks disclosed in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present invention may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software module may be located in ram, flash memory, rom, prom, or eprom, registers, etc. storage media as is well known in the art. The storage medium is located in a memory 41, and the processor 40 reads the information in the memory 41 and completes the steps of the method of the foregoing embodiment in combination with the hardware thereof.
The photovoltaic power prediction method, the photovoltaic power prediction apparatus, and the computer program product of the electronic device provided in the embodiments of the present invention include a computer-readable storage medium storing a program code, where instructions included in the program code may be used to execute the method described in the foregoing method embodiments, and specific implementations may refer to the method embodiments and are not described herein again.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working process of the apparatus described above may refer to the corresponding process in the foregoing method embodiment, and is not described herein again.
In addition, in the description of the embodiments of the present invention, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meaning of the above terms in the present invention can be understood in specific cases for those skilled in the art.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk, and various media capable of storing program codes.
In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of description and simplicity of description, but do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art will understand that the following embodiments are merely illustrative of the present invention, and not restrictive, and the scope of the present invention is not limited thereto: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present invention, and they should be construed as being included therein. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A method of photovoltaic power prediction, the method comprising:
acquiring a target time point, wherein the target time point is a time point needing to predict photovoltaic power;
extracting adjacent time points of the target time point and weather data of the adjacent time points;
searching target weather data matched with the weather data and historical photovoltaic power corresponding to the target weather data in a pre-stored weather power data table;
constructing a weather characteristic sequence according to the target weather data and the historical photovoltaic power;
inputting the weather characteristic sequence into a pre-established prediction model, and outputting the prediction power corresponding to the weather characteristic sequence through the prediction model; the prediction model is obtained by training weather characteristic data, wherein the weather characteristic data comprises the weather data and historical photovoltaic power corresponding to the weather data;
determining the predicted power as the predicted photovoltaic power of the target time point.
2. The method of claim 1, wherein the step of extracting time points adjacent to the target time point comprises:
determining a target unit time containing the target time point according to a predefined unit time;
determining a previous unit time adjacent to the target unit time as the adjacent time point to the target time point.
3. The method according to claim 1, wherein the weather power data table includes a plurality of historical photovoltaic powers and weather data corresponding to each of the historical photovoltaic powers;
the method comprises the following steps of searching target weather data matched with the weather data and historical photovoltaic power corresponding to the target weather data in a pre-stored weather power data table, wherein the steps comprise:
calculating the distance between the weather data of the adjacent time points and the weather data in the weather power data table according to a preset distance algorithm;
and determining the weather data corresponding to the distance smaller than a preset distance threshold value in the weather power data table as the target weather data, and searching the historical photovoltaic power corresponding to the target weather data in the weather power data table.
4. The method of claim 1, further comprising:
acquiring a pre-stored historical weather data table containing a plurality of weather data and a historical power data table corresponding to the historical weather data table and containing a plurality of historical photovoltaic powers;
assembling the weather data and the historical photovoltaic power according to a preset time interval to obtain weather characteristic data containing the weather data and the historical photovoltaic power;
adding the weather characteristic data to a pre-established training set;
and training an initial prediction network based on the training set to obtain the prediction model.
5. The method of claim 4, wherein the time interval comprises a first time interval corresponding to the historical photovoltaic power and a second time interval corresponding to the weather data, the first time interval being less than the second time interval;
assembling the weather data and the historical photovoltaic power according to a preset time interval to obtain weather characteristic data comprising the weather data and the historical photovoltaic power, wherein the step of assembling the weather data and the historical photovoltaic power comprises the following steps:
and combining the weather data and the historical photovoltaic power according to the time sequence corresponding to the first time interval, so that weather characteristic data comprising the weather data and the historical photovoltaic power is generated every other first time interval, and the weather data of the weather characteristic data in the second time interval are the same.
6. The method of claim 5, wherein the step of training an initial predictive network based on the training set comprises:
for each weather characteristic data, calculating an intermediate characteristic matched with the weather characteristic data in the weather power data table based on the weather data and the historical photovoltaic power;
assembling the intermediate features and the weather feature data to obtain feature vectors corresponding to the weather feature data;
and training the initial prediction network by using the feature vector to obtain the prediction model.
7. The method of claim 6, wherein the step of calculating an intermediate signature in the weather power data table that matches the weather signature data based on the weather data and the historical photovoltaic power comprises:
acquiring weather data contained in the weather characteristic data;
searching at least one target weather data, the distance between which and the weather data is smaller than a preset distance threshold value, in the weather power data table according to a preset distance algorithm;
extracting historical photovoltaic power corresponding to at least one target weather data in the weather power data table;
and calculating the historical photovoltaic power mean parameter, and determining the mean parameter as the intermediate feature matched with the weather feature data.
8. A photovoltaic power prediction apparatus, characterized in that the apparatus comprises:
the device comprises an acquisition module, a power generation module and a power generation module, wherein the acquisition module is used for acquiring a target time point, and the target time point is a time point needing to predict photovoltaic power;
the extraction module is used for extracting adjacent time points of the target time point and weather data of the adjacent time points;
the searching module is used for searching target weather data matched with the weather data and historical photovoltaic power corresponding to the target weather data in a pre-stored weather power data table;
the building module is used for building a weather characteristic sequence according to the target weather data and the historical photovoltaic power;
the forecasting module is used for inputting the weather characteristic sequence into a pre-established forecasting model and outputting forecasting power corresponding to the weather characteristic sequence through the forecasting model; the prediction model is obtained by training weather characteristic data, wherein the weather characteristic data comprises the weather data and historical photovoltaic power corresponding to the weather data;
a determination module, configured to determine the predicted power as the predicted photovoltaic power of the target time point.
9. An electronic device comprising a processor and a memory, the memory storing machine executable instructions executable by the processor, the processor executing the machine executable instructions to implement the method of any one of claims 1-7.
10. A machine-readable storage medium having stored thereon machine-executable instructions which, when invoked and executed by a processor, cause the processor to implement the method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211629198.0A CN115619047B (en) | 2022-12-19 | 2022-12-19 | Photovoltaic power prediction method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211629198.0A CN115619047B (en) | 2022-12-19 | 2022-12-19 | Photovoltaic power prediction method and device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115619047A CN115619047A (en) | 2023-01-17 |
| CN115619047B true CN115619047B (en) | 2023-02-28 |
Family
ID=84880558
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211629198.0A Active CN115619047B (en) | 2022-12-19 | 2022-12-19 | Photovoltaic power prediction method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115619047B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117332901B (en) * | 2023-10-17 | 2024-07-19 | 南方电网数字电网研究院股份有限公司 | New energy small time scale power prediction method adopting layered time aggregation strategy |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2339063A1 (en) * | 2001-01-19 | 2002-07-19 | Cognos Incorporated | System and method for business performance management |
| CN112507793A (en) * | 2020-11-05 | 2021-03-16 | 上海电力大学 | Ultra-short-term photovoltaic power prediction method |
| CN112925824A (en) * | 2021-02-25 | 2021-06-08 | 山东大学 | Photovoltaic power prediction method and system for extreme weather type |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9158035B2 (en) * | 2012-04-05 | 2015-10-13 | General Electric Company | System and method of automated acquisition, correlation and display of power distribution grid operational parameters and weather events |
-
2022
- 2022-12-19 CN CN202211629198.0A patent/CN115619047B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2339063A1 (en) * | 2001-01-19 | 2002-07-19 | Cognos Incorporated | System and method for business performance management |
| CN112507793A (en) * | 2020-11-05 | 2021-03-16 | 上海电力大学 | Ultra-short-term photovoltaic power prediction method |
| CN112925824A (en) * | 2021-02-25 | 2021-06-08 | 山东大学 | Photovoltaic power prediction method and system for extreme weather type |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115619047A (en) | 2023-01-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112285807B (en) | Meteorological information prediction method and device | |
| CN117078048B (en) | Digital twinning-based intelligent city resource management method and system | |
| CN115619047B (en) | Photovoltaic power prediction method and device and electronic equipment | |
| CN118153766B (en) | A method, device and medium for predicting wind power output under extreme weather scenarios | |
| CN117291314B (en) | Construction method of energy risk identification model, energy risk identification method and device | |
| CN110675879B (en) | Audio evaluation method, system, equipment and storage medium based on big data | |
| CN118644096A (en) | A short-term load forecasting method and system based on ARIMA and CNN-LSTM combined model | |
| CN113222210A (en) | Typhoon short-term prediction method and device based on multi-mode generation confrontation | |
| CN117933497B (en) | TSA-ARIMA-CNN-based enterprise carbon emission prediction method | |
| CN118133060A (en) | Photovoltaic power generation prediction method, device and storage medium based on time-series generation edge | |
| CN116109007B (en) | Power generation power determination method, server and storage medium | |
| CN119047651A (en) | Community energy consumption prediction method, device and equipment based on Bayesian calibration | |
| CN115826092A (en) | Multi-meteorological-element prediction method and system based on information self-attention model | |
| CN115689055A (en) | A short-term solar irradiance prediction method and device | |
| CN119397173A (en) | Data cleaning method, device, medium and equipment based on time series similarity | |
| CN110942196B (en) | Predicted irradiation correction method and device | |
| CN119272178B (en) | Data processing method and device | |
| CN119375986B (en) | Hail short-time prediction and early warning method, device and equipment based on time convolution network | |
| CN117688367B (en) | Wind power generation ultra-short term power prediction method and device based on instant learning | |
| CN119382103B (en) | TCN-LSTM wind power generation prediction method combining time-frequency analysis and attention mechanism | |
| CN119561047B (en) | A wind power prediction method, system and device based on multi-model fusion | |
| CN119670982B (en) | Method for improving the accuracy of photovoltaic power generation capacity prediction | |
| CN118399402B (en) | A multi-model coupled wind power prediction method, device, equipment and storage medium | |
| CN118761007B (en) | Integrated identification and analysis method, system and medium for photovoltaic power generation system | |
| CN118657243B (en) | Photovoltaic power generation power prediction method based on multivariate time sequence decomposition and multiple models |
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 |