Photovoltaic abnormal output prediction method and device based on atmospheric circulation field
By using a photovoltaic anomaly power output prediction method based on atmospheric circulation field, the similarity between the predicted anomaly field and historical data is calculated, and a photovoltaic anomaly power output prediction model is constructed. This solves the problem of insufficient photovoltaic power output prediction in traditional methods and achieves efficient prediction of photovoltaic anomaly power output.
Patent Information
- Application Number
- CN202511265680.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-05
- Publication Date
- 2025-12-05
AI Technical Summary
Traditional photovoltaic power output forecasting methods are unable to fully reflect weather changes, especially lacking the ability to forecast abnormal photovoltaic power output, which restricts the accurate prediction of abnormally high or low power output.
A method for predicting anomalous photovoltaic output based on atmospheric circulation fields constructs a circulation feature set by calculating the differences between numerical weather prediction and climatological data, and uses similarity to predict the anomalousness of photovoltaic output. This includes calculating the similarity between the predicted anomaly field and historical data, and constructing a prediction model for anomalous high and low photovoltaic output.
It improves the accuracy of photovoltaic power output forecasting, especially the ability to predict abnormal power output, and enhances the forecasting effect of future photovoltaic power output and abnormal situations.
Smart Images
Figure CN121077397A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of new energy prediction, in particular to a photovoltaic abnormal output prediction method and device based on atmospheric circulation field. BACKGROUND
[0002] Photovoltaic power generation is significantly affected by weather. Solar radiation is a decisive factor in determining photovoltaic power generation power, and changes in meteorological conditions such as temperature and humidity will also indirectly affect photovoltaic power generation power by affecting the efficiency of the power generation assembly. Therefore, photovoltaic output has obvious randomness and volatility. Traditional photovoltaic output prediction methods often select meteorological data such as temperature and precipitation to construct a photovoltaic output model according to experience, which is difficult to fully reflect the changes in the weather.
[0003] At the same time, atmospheric circulation field (geopotential height field) is closely related to precipitation, wind, and cloud processes, and has been widely used in weather forecasting. Although numerical weather prediction has developed rapidly, and numerous artificial intelligence prediction methods have also emerged, experienced professional forecasters still need to make predictions or corrections based on atmospheric circulation field in the process of weather forecasting by meteorological departments. However, the current traditional photovoltaic output prediction method has obvious deficiencies in the application of atmospheric circulation field, which seriously restricts the prediction ability of photovoltaic output, especially abnormal high and low photovoltaic output. SUMMARY
[0004] In order to overcome the above-mentioned defects, the present application provides a photovoltaic abnormal output prediction method and device based on atmospheric circulation field.
[0005] In a first aspect, a photovoltaic abnormal output prediction method based on atmospheric circulation field is provided, which comprises:
[0006] calculating the difference between the gridded forecast data of the geopotential height of the preset height layer on the preset date in the numerical weather prediction and the climatological data of the preset height layer corresponding to the preset date, to obtain the forecast anomaly field of the preset height layer on the preset date;
[0007] predicting the abnormality of the photovoltaic output on the preset date based on the similarity between the forecast anomaly field of the preset height layer on the preset date and each element in the circulation feature set constructed in advance.
[0008] Preferably, the preset height layer is 850hPa, 700hPa or 500hPa.
[0009] Preferably, the process of obtaining the climatological data of the preset height layer corresponding to the preset date comprises:
[0010] daily average the geopotential height data of the preset height layer in the past X years to obtain the climatological data of the preset height layer corresponding to each date in a year.
[0011] obtaining the climatological data of the preset height layer corresponding to the preset date from the climatological data of the preset height layer corresponding to each date in the year.
[0012] Further, the construction process of the pre-constructed circulation feature set comprises:
[0013] taking the difference between the potential height data of the preset height layer corresponding to each date in the history X years and the climatological data of the preset height layer corresponding to the corresponding date in the year as the anomaly field corresponding to each date in the history X years;
[0014] calculating the similarity between the anomaly field in the pre-obtained high / low output date circulation anomaly field feature set and the anomaly field corresponding to each date in the history X years, arranging them in order from small to large similarity, and obtaining the 80% quantile of the sequence;
[0015] extracting the anomaly field greater than the 80% quantile of the sequence, and calculating the prediction ability of photovoltaic abnormal high / low output corresponding to each anomaly field;
[0016] constructing the circulation feature set by using the anomaly field data of the photovoltaic abnormal high / low output whose prediction ability exceeds 0.8.
[0017] Further, the obtaining process of the pre-obtained high / low output date circulation anomaly field feature set comprises:
[0018] calculating the moving average and the moving standard deviation of the photovoltaic power plant output data based on the photovoltaic power plant output data in the history year;
[0019] calculating the upper limit and the lower limit of the normal fluctuation threshold of the photovoltaic power plant output data based on the moving average and the moving standard deviation of the photovoltaic power plant output data;
[0020] extracting the dates greater than the upper limit of the normal fluctuation threshold and the dates less than the lower limit of the normal fluctuation threshold from the photovoltaic power plant output data in the history year;
[0021] constructing the high output date circulation anomaly field feature set by using the anomaly field corresponding to the dates greater than the upper limit of the normal fluctuation threshold, and constructing the low output date circulation anomaly field feature set by using the anomaly field corresponding to the dates less than the lower limit of the normal fluctuation threshold.
[0022] Further, the prediction ability of photovoltaic abnormal high / low output corresponding to each anomaly field is as follows:
[0023]
[0024] In the above formula, P kP is the prediction ability of the photovoltaic abnormal high / low output corresponding to the data k in the high / low output date circulation anomaly field feature set k1 P is the number of times that the photovoltaic power plant output data of the date corresponding to the anomaly field greater than the 80th percentile of the sequence of the data k in the high / low output date circulation anomaly field feature set is greater than the upper limit of the normal fluctuation threshold / less than the lower limit of the normal fluctuation threshold k2 P is the amount of data corresponding to the anomaly field greater than the 80th percentile of the sequence of the data k in the high / low output date circulation anomaly field feature set
[0025] Further, the moving average and moving standard deviation of the photovoltaic power plant output data are as follows:
[0026]
[0027] In the above formula, DATA ma(t) P is the moving average of the photovoltaic power plant output data on day t, window is the moving window, and DATA j P is the photovoltaic power plant output data on day j in the past year, DATA msd(t) P is the moving standard deviation of the photovoltaic power plant output data on day t.
[0028] Further, the upper and lower limits of the normal fluctuation threshold of the photovoltaic power plant output data are as follows:
[0029] limit top P is the upper limit of the normal fluctuation threshold of the photovoltaic power plant output data, DATA ma(t) P is the lower limit of the normal fluctuation threshold of the photovoltaic power plant output data, DATA msd(t)
[0030] limit bottom P is the upper limit of the normal fluctuation threshold of the photovoltaic power plant output data, DATA ma(t) P is the lower limit of the normal fluctuation threshold of the photovoltaic power plant output data, DATA msd(t)
[0031] In the above formula, limit top P is the upper limit of the normal fluctuation threshold of the photovoltaic power plant output data, limit bottom P is the lower limit of the normal fluctuation threshold of the photovoltaic power plant output data.
[0032] Further, the similarity between the predicted anomaly field of the preset height layer on the preset date and each element in the pre-constructed circulation feature set predicts the abnormality of the photovoltaic output on the preset date, comprising:
[0033] It is judged whether the predicted anomaly field of the preset height layer on the preset date satisfies:
[0034] res i P is the upper limit of the normal fluctuation threshold of the photovoltaic power plant output data, limit i
[0035] If yes, the preset date photovoltaic output is abnormal, otherwise, the preset date photovoltaic output is not abnormal;
[0036] Wherein, res i is the similarity of the forecast anomaly field of the preset height layer on the preset date and the element i in the pre-constructed circulation feature set, CosSimP i is the 80% quantile corresponding to the element i in the pre-constructed circulation feature set, i=1…I, I is the number of elements in the pre-constructed circulation feature set.
[0037] In a second aspect, a photovoltaic abnormal output prediction device based on an atmospheric circulation field is provided, and the photovoltaic abnormal output prediction device based on the atmospheric circulation field comprises:
[0038] A calculation module is configured to calculate the difference between the gridded forecast data of the potential height of the preset height layer in the numerical weather prediction and the climatological data of the preset height layer corresponding to the preset date, to obtain the forecast anomaly field of the preset height layer on the preset date.
[0039] A prediction module is configured to predict the abnormality of the photovoltaic output on the preset date based on the similarity between the forecast anomaly field of the preset height layer on the preset date and each element in the pre-constructed circulation feature set.
[0040] In a third aspect, a computer device is provided, comprising one or more processors.
[0041] The processor is configured to store one or more programs.
[0042] When the one or more programs are executed by the one or more processors, the photovoltaic abnormal output prediction method based on the atmospheric circulation field is implemented.
[0043] In a fourth aspect, a computer readable storage medium having a computer program stored thereon is provided, and when the computer program is executed, the photovoltaic abnormal output prediction method based on the atmospheric circulation field is implemented.
[0044] The above one or more technical solutions of the present application have at least one or more of the following beneficial effects:
[0045] The application relates to the technical field of new energy prediction, and particularly provides a photovoltaic abnormal output prediction method and device based on an atmospheric circulation field, which comprises the following steps: calculating the difference between the potential height gridded prediction data of a preset height layer on a preset date in numerical weather prediction and the climate state data of the preset height layer corresponding to the preset date, to obtain the prediction anomaly field of the preset height layer on the preset date; and predicting the abnormality of photovoltaic output on the preset date based on the similarity between the prediction anomaly field of the preset height layer on the preset date and each element in a previously constructed circulation feature set. The technical scheme provided by the application enhances the application level of atmospheric circulation features in photovoltaic output prediction, thereby improving the prediction capability of photovoltaic output, especially abnormal output, can predict abnormal conditions of future photovoltaic output according to the predicted atmospheric circulation features, and can also apply the method to the optimization of abnormal output prediction effect of other irradiance prediction methods. BRIEF DESCRIPTION OF DRAWINGS
[0046] Figure 1 FIG. 1 is a main step flow diagram of a photovoltaic abnormal output prediction method based on an atmospheric circulation field according to an embodiment of the application. DETAILED DESCRIPTION
[0047] The specific embodiments of the application will be further described in detail below with reference to the accompanying drawings.
[0048] To make the objectives, technical schemes and advantages of the embodiments of the application clearer, the technical scheme of the embodiments of the application will be described clearly and completely below with reference to the drawings of the embodiments of the application. Obviously, the described embodiments are some but not all of the embodiments of the application. Based on the embodiments of the application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the application.
[0049] Embodiment 1
[0050] Reference is made to the accompanying drawings Figure 1 , Figure 1 FIG. 1 is a main step flow diagram of a photovoltaic abnormal output prediction method based on an atmospheric circulation field according to an embodiment of the application. As shown in FIG. 1, the photovoltaic abnormal output prediction method based on the atmospheric circulation field according to the embodiment of the application mainly comprises the following steps: Figure 1
[0051] Step S101: calculating the difference between the potential height gridded prediction data of a preset height layer on a preset date in numerical weather prediction and the climate state data of the preset height layer corresponding to the preset date, to obtain the prediction anomaly field of the preset height layer on the preset date;
[0052] Step S102: predicting the abnormality of the photovoltaic output on the preset date based on the similarity between the forecast anomaly field of the preset height layer on the preset date and each element in the pre-constructed circulation feature set.
[0053] In this embodiment, the preset height layer is 850 hPa, 700 hPa or 500 hPa.
[0054] In this embodiment, the process of obtaining the climatological data of the preset height layer corresponding to the preset date includes:
[0055] The potential height data of the preset height layer in the past X years is averaged daily to obtain the climatological data of the preset height layer corresponding to each date in a year; for example, when N is 30, after removing the last day of the leap year, 10950 samples corresponding to 30 years x 365 days can be obtained. For each day in thirty years, 365 samples after 30-year averaging are obtained as the climatological data of the corresponding circulation field
[0056] The climatological data of the preset height layer corresponding to the preset date is obtained from the climatological data of the preset height layer corresponding to each date in the year.
[0057] In one embodiment, the process of constructing the pre-constructed circulation feature set includes:
[0058] The difference between the potential height data of the preset height layer corresponding to each date in the past X years and the climatological data of the preset height layer corresponding to the corresponding date in a year is taken as the anomaly field corresponding to each date in the past X years.
[0059] The similarity between the anomaly field in the pre-acquired high / low output date circulation anomaly field feature set and the anomaly field corresponding to each date in the past X years is calculated, arranged in order from small to large similarity, and the 80% quantile of the sequence is obtained; one or more of cosine similarity (CosSim), mean absolute error (MAE), structural similarity (SSIM), peak signal-to-noise ratio (PSNR), and root mean square error (RMSE) can be used to calculate the similarity;
[0060] Anomaly fields greater than the 80% quantile of the sequence are extracted, and the prediction ability of each anomaly field corresponding to the photovoltaic abnormal high / low output is calculated.
[0061] The anomaly field feature set is constructed using the anomaly field data whose prediction ability of photovoltaic abnormal high / low output exceeds 0.8.
[0062] In one embodiment, the process of obtaining the pre-acquired high / low output date circulation anomaly field feature set includes:
[0063] calculating a sliding average and a sliding standard deviation of the photovoltaic power plant output data based on historical photovoltaic power plant output data in a year;
[0064] calculating upper and lower normal fluctuation thresholds of the photovoltaic power plant output data based on the sliding average and the sliding standard deviation of the photovoltaic power plant output data;
[0065] extracting dates greater than the upper normal fluctuation threshold and dates less than the lower normal fluctuation threshold from the photovoltaic power plant output data in the historical year;
[0066] constructing a high-output-date circulation anomaly field feature set using the anomaly fields corresponding to the dates greater than the upper normal fluctuation threshold and constructing a low-output-date circulation anomaly field feature set using the anomaly fields corresponding to the dates less than the lower normal fluctuation threshold.
[0067] In one embodiment, the prediction capability of each anomaly field for photovoltaic abnormal high / low output is as follows:
[0068]
[0069] In the above formula, P k is the prediction capability of photovoltaic abnormal high / low output corresponding to data k in the high / low output date circulation anomaly field feature set, P k1 is the number of times that the photovoltaic power plant output data of the date corresponding to the anomaly field greater than the 80th percentile of the sequence is greater than the upper normal fluctuation threshold / less than the lower normal fluctuation threshold, P k2 is the data amount corresponding to the anomaly field greater than the 80th percentile of the sequence.
[0070] In one embodiment, the sliding average and the sliding standard deviation of the photovoltaic power plant output data are as follows:
[0071]
[0072] In the above formula, DATA ma(t) is the sliding average of the photovoltaic power plant output data on day t, window is the sliding window, Data j is the photovoltaic power plant output data on day j in the historical year, DATA msd(t) is the sliding standard deviation of the photovoltaic power plant output data on day t.
[0073] In one embodiment, the upper and lower normal fluctuation thresholds of the photovoltaic power plant output data are as follows:
[0074] limittop = DATA ma(t) + DATA msd(t)
[0075] limit bottom = DATA ma(t) - DATA msd(t)
[0076] In the above formula, limit top is the upper limit of the normal fluctuation threshold of the photovoltaic power station output data, and limit bottom is the lower limit of the normal fluctuation threshold of the photovoltaic power station output data.
[0077] In one embodiment, the similarity between the forecast anomaly field of the preset height layer on the preset date and each element in the pre-constructed circulation feature set predicts the abnormality of the photovoltaic output on the preset date, comprising:
[0078] determining whether the forecast anomaly field of the preset height layer on the preset date satisfies:
[0079] res i > CosSimP i
[0080] If yes, the photovoltaic output on the preset date is abnormal, otherwise, the photovoltaic output on the preset date is not abnormal;
[0081] wherein, res i is the similarity between the forecast anomaly field of the preset height layer on the preset date and the element i in the pre-constructed circulation feature set, CosSimP i is the 80% quantile corresponding to the element i in the pre-constructed circulation feature set, i = 1…I, I is the number of elements in the pre-constructed circulation feature set.
[0082] Embodiment 2
[0083] Based on the same inventive concept, the present application also provides a photovoltaic abnormal output prediction device based on atmospheric circulation field, comprising:
[0084] a calculation module for calculating the difference between the gridded forecast data of the potential height of the preset height layer in the numerical weather prediction on the preset date and the climatological data of the preset height layer corresponding to the preset date, to obtain the forecast anomaly field of the preset height layer on the preset date;
[0085] a prediction module for predicting the abnormality of the photovoltaic output on the preset date based on the similarity between the forecast anomaly field of the preset height layer on the preset date and each element in the pre-constructed circulation feature set.
[0086] Preferably, the preset height layer is 850 hPa, 700 hPa or 500 hPa.
[0087] Preferably, the process of obtaining the climate state data of the preset height layer corresponding to the preset date comprises:
[0088] The potential height data of the preset height layer in the past X years is averaged daily to obtain the climate state data of the preset height layer corresponding to each date in a year;
[0089] The climate state data of the preset height layer corresponding to the preset date is obtained from the climate state data of the preset height layer corresponding to each date in the year.
[0090] Further, the process of constructing the pre-constructed circulation feature set comprises:
[0091] The difference between the potential height data of the preset height layer corresponding to each date in the past X years and the climate state data of the preset height layer corresponding to the corresponding date in a year is taken as the anomaly field corresponding to each date in the past X years;
[0092] The similarity between the anomaly field in the pre-obtained high / low output date circulation anomaly field feature set and the anomaly field corresponding to each date in the past X years is calculated, arranged in order from small to large, and the 80% quantile is obtained;
[0093] Anomaly fields greater than the 80% quantile are extracted, and the prediction ability of each anomaly field corresponding to the abnormal high / low output of photovoltaic is calculated;
[0094] The anomaly field data with a prediction ability of photovoltaic abnormal high / low output exceeding 0.8 is used to construct the circulation feature set.
[0095] Further, the process of obtaining the pre-obtained high / low output date circulation anomaly field feature set comprises:
[0096] The moving average and moving standard deviation of the photovoltaic power plant output data are calculated based on the photovoltaic power plant output data in the past year;
[0097] The upper and lower limits of the normal fluctuation threshold of the photovoltaic power plant output data are calculated based on the moving average and moving standard deviation of the photovoltaic power plant output data;
[0098] The dates greater than the upper limit of the normal fluctuation threshold and the dates less than the lower limit of the normal fluctuation threshold are extracted from the photovoltaic power plant output data in the past year;
[0099] The high-power date circulation anomaly field feature set is constructed using the date corresponding to the anomaly field greater than the upper limit of the normal fluctuation threshold, and the low-power date circulation anomaly field feature set is constructed using the date corresponding to the anomaly field less than the lower limit of the normal fluctuation threshold.
[0100] Further, the prediction ability of the abnormal high / low power of photovoltaic of each anomaly field is as follows:
[0101]
[0102] In the above formula, P k is the prediction ability of the abnormal high / low power of photovoltaic of the data k in the high / low power date circulation anomaly field feature set, P k1 is the number of times that the photovoltaic power plant output data of the date corresponding to the anomaly field greater than the 80th percentile of the sequence is greater than the upper limit of the normal fluctuation threshold / less than the lower limit of the normal fluctuation threshold, P k2 is the data amount of the data corresponding to the anomaly field greater than the 80th percentile of the sequence.
[0103] Further, the moving average and the moving standard deviation of the photovoltaic power plant output data are as follows:
[0104]
[0105] In the above formula, DATA ma(t) is the moving average of the photovoltaic power plant output data on day t, window is the moving window, DATA j is the photovoltaic power plant output data on day j in the history year, DATA msd(t) is the moving standard deviation of the photovoltaic power plant output data on day t.
[0106] Further, the upper limit and the lower limit of the normal fluctuation threshold of the photovoltaic power plant output data are as follows:
[0107] limit top = DATA ma(t) + DATA msd(t)
[0108] limit bottom = DATA ma(t) - DATA msd(t)
[0109] In the above formula, limit top is the upper limit of the normal fluctuation threshold of the photovoltaic power plant output data, limit bottom is the lower limit of the normal fluctuation threshold of the photovoltaic power plant output data.
[0110] Further, the similarity between the forecast anomaly field of the preset height layer on the preset date and each element in the pre-constructed circulation feature set predicts the abnormality of the preset date photovoltaic output, including:
[0111] determining whether the forecast anomaly field of the preset height layer on the preset date satisfies:
[0112] res i CosSimP i
[0113] If yes, the preset date photovoltaic output is abnormal, otherwise, the preset date photovoltaic output is not abnormal.
[0114] wherein, res i is the similarity between the forecast anomaly field of the preset height layer on the preset date and the element i in the pre-constructed circulation feature set, CosSimP i is the 80% quantile corresponding to the element i in the pre-constructed circulation feature set, i = 1…I, I is the number of elements in the pre-constructed circulation feature set.
[0115] Embodiment 3
[0116] Based on the same inventive concept, the present application also provides a computer device, which comprises a processor and a memory, the memory is used to store a computer program, the computer program comprises program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, and are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions in the computer storage medium to implement a corresponding method process or a corresponding function, so as to implement the steps of the photovoltaic abnormal output prediction method based on the atmospheric circulation field in the above-mentioned embodiments.
[0117] Embodiment 4
[0118] Based on the same inventive concept, the present application also provides a storage medium, specifically a computer readable storage medium (Memory), which is a memory device in a computer device, used for storing programs and data. It can be understood that the computer readable storage medium here can include the built-in storage medium in the computer device, and of course can also include the extended storage medium supported by the computer device. The computer readable storage medium provides a storage space which stores the operating system of the terminal. Moreover, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer readable storage medium here can be a high-speed RAM memory, or a non-volatile memory such as at least one disk memory. One or more instructions stored in the computer readable storage medium can be loaded and executed by the processor to realize the steps of the above-mentioned atmospheric circulation field-based photovoltaic abnormal output prediction method.
[0119] Those skilled in the art will appreciate that embodiments of the present application can be supplied as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage media, etc.) containing computer-usable program code.
[0120] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks Figure 1 The means for implementing the functions specified in one or more flows and / or blocks.
[0121] These computer program instructions can also be stored in a computer readable memory capable of directing the computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce a product including instruction means, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocksFigure 1 the function specified in the one or more blocks.
[0122] These computer program instructions can also be loaded into computer or other programmable data processing devices, so that a series of operation steps are performed on the computer or other programmable data processing devices to generate computer-implemented processes, thus the instructions executed on the computer or other programmable data processing devices provide a process for implementing the flow Figure 1 the flow or flows and / or blocks Figure 1 the steps of the function specified in the one or more blocks.
[0123] Finally, it should be noted that: the above examples are only used to illustrate the technical solutions of the present application, but not to limit it, although the above embodiments of the present application have been described in detail, those skilled in the art should understand: the specific embodiments of the present application can be modified or replaced by the same, without departing from the spirit and scope of the present application, any modification or equivalent replacement, which should be covered in the protection scope of the claims of the present application.
Claims
1. A photovoltaic abnormal power output prediction method based on atmospheric circulation field, characterized in that, The method comprises: calculating the difference between the gridded forecast data of the geopotential height of a preset height layer on a preset date in a numerical weather prediction and the climatological data of the preset height layer corresponding to the preset date, to obtain a forecast anomaly field of the preset height layer on the preset date; predicting the abnormality of the photovoltaic output on the preset date based on the similarity between the forecast anomaly field of the preset height layer on the preset date and each element in a pre-constructed circulation feature set.
2. The method of claim 1, wherein, The preset height layer is 850 hPa, 700 hPa or 500 hPa.
3. The method of claim 1, wherein, The process of obtaining the climatological data of the preset height layer corresponding to the preset date comprises: daily averaging the geopotential height data of the preset height layer in the past X years to obtain the climatological data of the preset height layer corresponding to each date in a year; obtaining the climatological data of the preset height layer corresponding to the preset date from the climatological data of the preset height layer corresponding to each date in a year.
4. The method of claim 3, wherein, The process of constructing the pre-constructed circulation feature set comprises: taking the difference between the geopotential height data of the preset height layer corresponding to each date in the past X years and the climatological data of the preset height layer corresponding to the corresponding date in a year as the anomaly field corresponding to each date in the past X years; calculating the similarity between the anomaly fields in a pre-obtained high / low output date circulation anomaly field feature set and the anomaly fields corresponding to each date in the past X years, arranging them in order from small to large similarity, and obtaining the 80% quantile of the sequence; extracting the anomaly fields greater than the 80% quantile of the sequence, and calculating the prediction ability of photovoltaic abnormal high / low output corresponding to each anomaly field; constructing the circulation feature set using the anomaly field data whose prediction ability of photovoltaic abnormal high / low output exceeds 0.
8.
5. The method of claim 4, wherein, The process of obtaining the pre-obtained high / low output date circulation anomaly field feature set comprises: calculating the moving average and moving standard deviation of the photovoltaic power plant output data based on the photovoltaic power plant output data in the past year; calculating the upper and lower limits of the normal fluctuation threshold of the photovoltaic power plant output data based on the moving average and moving standard deviation of the photovoltaic power plant output data; extracting dates greater than the upper limit of the normal fluctuation threshold and dates less than the lower limit of the normal fluctuation threshold from the photovoltaic power plant output data in the past year; constructing a high output date circulation anomaly field feature set using the anomaly fields corresponding to the dates greater than the upper limit of the normal fluctuation threshold, and constructing a low output date circulation anomaly field feature set using the anomaly fields corresponding to the dates less than the lower limit of the normal fluctuation threshold.
6. The method of claim 5, wherein, The prediction ability of photovoltaic abnormal high / low output corresponding to each anomaly field is as follows: In the above formula, P k is the prediction capability of the photovoltaic abnormal high / low output corresponding to the data k in the high / low output date circulation anomaly field characteristic set, P k1 is the number of times that the photovoltaic power plant output data of the date corresponding to the anomaly field greater than the 80% quantile of the sequence of data k in the high / low output date circulation anomaly field characteristic set is greater than the upper limit of the normal fluctuation threshold / less than the lower limit of the normal fluctuation threshold, P k2 is the data amount corresponding to the anomaly field greater than the 80% quantile of the sequence of data k in the high / low output date circulation anomaly field characteristic set.
7. The method of claim 6, wherein, The moving average and moving standard deviation of the photovoltaic power plant output data are as follows: In the above formula, DATA ma(t) is the moving average of the output data of the photovoltaic power plant on day t, weindow is the moving window, Data j is the output data of the photovoltaic power plant on day j in the past year, DATA msd(t) is the moving standard deviation of the output data of the photovoltaic power plant on day t.
8. The method of claim 7, wherein, The upper and lower limits of the normal fluctuation threshold of the photovoltaic power plant output data are as follows: limit top = DATA ma(t) + DATA msd(t) limit bottom = DATA ma(t) -DATA msd(t) In the above formula, limit top is the upper limit of the normal fluctuation threshold of the output data of the photovoltaic power plant, limit bottom is the lower limit of the normal fluctuation threshold of the output data of the photovoltaic power plant.
9. The method of claim 6, wherein, The prediction of the abnormality of the photovoltaic output on the preset date based on the similarity between the forecast anomaly field of the preset height layer on the preset date and each element in a pre-constructed circulation feature set comprises: determining whether the forecast anomaly field of the preset height layer on the preset date satisfies: res i >CosSimP i if yes, the photovoltaic output on the preset date is abnormal, otherwise, the photovoltaic output on the preset date is not abnormal; wherein res i is the similarity between the forecast anomaly field of the preset height layer for the preset date and the element i in the pre-constructed circulation feature set, CosSimP i is the 80% quantile of the sequence corresponding to the element i in the pre-constructed circulation feature set, i = 1…I, I is the number of elements in the pre-constructed circulation feature set.
10. A device for predicting photovoltaic abnormal power output based on the atmospheric circulation field-based photovoltaic abnormal power output prediction method according to any one of claims 1 to 9, characterized by, The device comprises: The computing module is configured to calculate a difference between the gridded forecast data of the geopotential height at the preset height layer on the preset date and the climatological data of the preset height layer corresponding to the preset date, to obtain a forecast anomaly field of the preset height layer on the preset date. The prediction module is configured to predict abnormality of photovoltaic output on the preset date based on a similarity between the forecast anomaly field of the preset height layer on the preset date and each element in the set of circulation features.
11. A computer device, comprising: The method comprises: one or more processors; the processor is configured to execute one or more programs; when the one or more programs are executed by the one or more processors, the method for predicting abnormal output of photovoltaic power based on atmospheric circulation field is realized.
12. A computer-readable storage medium, characterized in that, The computer program is stored on the computer readable medium, and when the computer program is executed, the method for predicting abnormal output of photovoltaic power based on atmospheric circulation field is realized.