New energy ultra-short-term power generation prediction method, device, equipment and medium

Abstract

The application discloses a new energy ultra-short-term power generation power prediction method, device, equipment and medium, through acquiring environment sensor data, satellite remote sensing data and aerosol degree data of a target area where a new energy generator set is located; using a preset data assimilation model, based on the satellite remote sensing data and the aerosol degree data, the environment sensor data is corrected to obtain target environment data; the target environment data is subjected to feature extraction to obtain environment feature data of the target area; using a preset ultra-short-term power prediction model, according to the environment feature data, power generation power data of the new energy generator set in a future short time period is predicted, so that the correlation between the satellite remote sensing data and the aerosol degree data and the environment meteorological data is used to correct abnormal data and missing sections in the environment meteorological data, the data quality of the environment sensor data is ensured, and the accuracy of the ultra-short-term prediction is improved.

Description

Technical Field

[0001] This application relates to the field of power forecasting technology, and in particular to a method, device, equipment and medium for predicting the ultra-short-term power generation of new energy sources. Background Technology

[0002] Due to the intermittent and fluctuating nature of renewable energy power grids, real-time adjustments to the power generation plans of the grid dispatching department are necessary to ensure power supply stability. Therefore, ultra-short-term power generation forecasting for renewable energy power grids is essential. Currently, renewable energy power grids mainly consist of photovoltaic and wind power grids. Their generator units are equipped with various sensor components, such as temperature sensors, wind speed sensors, wind direction sensors, light sensors, and humidity sensors. The complex environment and variable weather conditions of the generator units can easily lead to data loss due to long-term use of these sensor components. Furthermore, renewable energy power generation is significantly affected by environmental and weather changes, thus impacting the accuracy of power generation forecasts. Summary of the Invention

[0003] This application provides a method, device, equipment, and medium for predicting ultra-short-term power generation of new energy sources, in order to solve the technical problem that the instability of current power grid data quality affects the accuracy of power generation prediction.

[0004] To address the aforementioned technical problems, firstly, this application provides a method for predicting ultra-short-term power generation from new energy sources, comprising:

[0005] Acquire environmental sensor data, satellite remote sensing data, and aerosol density data for the target area where the new energy generator unit is located;

[0006] Using a pre-defined data assimilation model, environmental sensor data is corrected based on satellite remote sensing data and aerosol density data to obtain target environmental data;

[0007] Feature extraction is performed on the target environmental data to obtain environmental feature data of the target area;

[0008] Using a pre-set ultra-short-term power prediction model, the power generation data of new energy generator units in the future short period of time is predicted based on environmental characteristic data.

[0009] In some implementations of the first aspect, a preset data assimilation model is used to correct environmental sensor data based on satellite remote sensing data and aerosol density data to obtain target environmental data, including:

[0010] Using the environmental simulation model in the preset data assimilation model, the environmental simulation data of the target area is simulated based on satellite remote sensing data and aerosol density data;

[0011] By using the improved Kalman filter in the preset data assimilation model, the environmental sensor data is corrected based on the environmental simulation data to obtain the target environmental data.

[0012] In some implementations of the first aspect, an environmental simulation model within a pre-defined data assimilation model is used to simulate environmental data of the target area based on satellite remote sensing data and aerosol density data, including:

[0013] Feature extraction is performed on satellite remote sensing data and aerosol density data to obtain simulated feature data;

[0014] Using an environmental simulation model, based on simulation characteristic data, environmental simulation data of the target area is simulated. The environmental simulation model is an environmental meteorological inversion model.

[0015] In some implementations of the first aspect, feature extraction is performed on satellite remote sensing data and aerosol density data to obtain simulated feature data, including:

[0016] Based on a preset feature extraction model, remote sensing feature data from satellite remote sensing data and aerosol feature data from aerosol density data are extracted.

[0017] Based on a pre-defined radiative transfer model, the first correlation feature data between satellite remote sensing data and the environmental meteorology of the target area is extracted;

[0018] Based on a pre-defined aerosol model, the second correlation feature data between aerosol intensity data and the environmental meteorology of the target area is extracted. The simulated feature data includes remote sensing feature data, aerosol feature data, first correlation feature data, and second correlation feature data.

[0019] In some implementations of the first aspect, an improved Kalman filter in a pre-defined data assimilation model is used to correct environmental sensor data based on environmental simulation data to obtain target environmental data, including:

[0020] Data cleaning is performed on environmental sensor data to remove abnormal data and identify missing segments in the environmental sensor data after removing abnormal data.

[0021] By using an improved Kalman filter, missing segments in environmental sensor data are corrected to obtain target environmental data.

[0022] In some implementations of the first aspect, an improved Kalman filter is used to correct missing segments in the environmental sensor data to obtain target environmental data, including:

[0023] Based on the current environmental simulation data, environmental sensor data, covariance matrix, and state transition equation, predict the environmental prediction data for the next time step;

[0024] Based on environmental prediction data, update the covariance matrix for the next time step;

[0025] Calculate the Kalman gain matrix based on the covariance matrix at the next time step;

[0026] Based on the Kalman gain matrix, the environmental sensor data at the next time step is updated to correct the environmental sensor data and obtain the target environmental data.

[0027] In some implementations of the first aspect, environmental prediction data for the next time step is predicted based on the current environmental simulation data, environmental sensor data, and covariance matrix, including:

[0028] Calculate the mean data between the current environmental simulation data and the environmental sensor data;

[0029] Based on the covariance matrix, an unscented transformation is performed on the mean data to obtain the sigma point set;

[0030] By inputting the sigma point set into the state transition equation, we obtain the environmental prediction data for the next time step.

[0031] Secondly, this application also provides a new energy ultra-short-term power generation prediction device, comprising:

[0032] The acquisition module is used to acquire environmental sensor data, satellite remote sensing data, and aerosol density data of the target area where the new energy generator set is located;

[0033] The correction module is used to correct environmental sensor data based on satellite remote sensing data and aerosol density data using a preset data assimilation model to obtain target environmental data.

[0034] The extraction module is used to extract features from the target environment data to obtain environmental feature data of the target area;

[0035] The prediction module is used to predict the power generation data of new energy generator sets in the future short period of time by using a preset ultra-short-term power prediction model and based on environmental characteristic data.

[0036] Thirdly, this application also provides a computer device, including a processor and a memory, wherein the memory is used to store a computer program, and when the computer program is executed by the processor, it implements the new energy ultra-short-term power generation prediction method as described in the first aspect.

[0037] Fourthly, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the new energy ultra-short-term power generation prediction method as described in the first aspect.

[0038] This application has at least the following beneficial effects:

[0039] By acquiring environmental sensor data, satellite remote sensing data, and aerosol density data of the target area where the new energy generator sets are located; using a preset data assimilation model, the environmental sensor data is corrected based on the satellite remote sensing data and aerosol density data to obtain target environmental data; feature extraction is performed on the target environmental data to obtain environmental characteristic data of the target area; using a preset ultra-short-term power prediction model, the power generation data of the new energy generator sets in the future short period of time is predicted based on the environmental characteristic data. By utilizing the correlation between satellite remote sensing data and aerosol density data and environmental meteorological data, abnormal data and missing segments in environmental meteorological data are corrected to ensure the data quality of environmental sensor data, thereby improving the accuracy of ultra-short-term prediction. Attached Figure Description

[0040] Figure 1 This is a flowchart illustrating the new energy ultra-short-term power generation prediction method according to an embodiment of this application;

[0041] Figure 2 This is a schematic diagram of the structure of the new energy ultra-short-term power generation prediction device shown in the embodiments of this application;

[0042] Figure 3 This is a schematic diagram of the structure of a computer device shown in an embodiment of this application. Detailed Implementation

[0043] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0044] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a method for predicting ultra-short-term power generation of new energy sources, provided in an embodiment of this application. The method for predicting ultra-short-term power generation of new energy sources in this embodiment can be applied to computer equipment, including but not limited to smartphones, laptops, tablets, desktop computers, physical servers, and cloud servers. Figure 1 As shown, the new energy ultra-short-term power generation prediction method of this embodiment includes steps S101 to S104, which are described in detail below:

[0045] Step S101: Obtain environmental sensor data, satellite remote sensing data, and aerosol density data for the target area where the new energy generator unit is located.

[0046] In this step, the new energy generator set can be a photovoltaic generator set or a wind turbine generator set; environmental sensor data includes, but is not limited to, temperature, humidity, light intensity, light angle, wind speed and wind direction; satellite remote sensing data is data collected by satellites from observing and collecting the Earth's surface, which is stored in the form of multispectral images, radar images, elevation data, etc., and contains rich information about the Earth's surface; aerosol optical depth (AOD) is a parameter used to describe the ability of aerosols in the atmosphere to absorb and scatter solar radiation.

[0047] Step S102: Using a preset data assimilation model, the environmental sensor data is corrected based on the satellite remote sensing data and aerosol density data to obtain target environmental data.

[0048] In this step, data assimilation involves combining observed data with model simulation results to update the model's state, typically achieved by minimizing the difference between observed values ​​and model simulation results. Because environmental sensors may malfunction or experience other anomalies due to factors such as lifespan limitations or environmental damage, leading to data mutations and missing information, this embodiment uses satellite remote sensing data and aerosol density data, which have a certain physical relationship with environmental meteorology, to correct the environmental sensor data and improve its quality.

[0049] Optionally, data assimilation can be implemented using algorithms such as Kalman filtering, variational methods, and particle filtering. In this embodiment, environmental prediction data predicted from satellite remote sensing data and aerosol density data is used as observation data, and environmental sensor data is used as measured data. An improved Kalman filtering algorithm is used for smooth interpolation to ensure that the corrected environmental sensor data better matches the actual environmental conditions, thereby improving the accuracy of subsequent power generation prediction.

[0050] Step S103: Extract features from the target environment data to obtain environmental feature data of the target area.

[0051] In this step, the time-series features of the target environmental data can be extracted based on statistical algorithms such as mean, variance, maximum, minimum, median, and percentile. The correlation features between the target environmental data and the power data can be extracted using algorithms such as Pearson correlation coefficient, Spearman correlation coefficient, determination coefficient, point-bivariate correlation coefficient, mutual information, and machine learning.

[0052] Step S104: Using a preset ultra-short-term power prediction model, predict the power generation data of the new energy generator set in the future short period of time based on the environmental characteristic data.

[0053] In this embodiment, the preset ultra-short-term power prediction model can be a Long Short-Term Memory (LSTM) network or a Convolutional Neural Network (CNN), etc. Optionally, target environmental data from a historical time period prior to the current moment can be used as the model input for the ultra-short-term power prediction model to predict the power generation of the generator set in the future time period.

[0054] In one embodiment, step S102 includes:

[0055] Using the environmental simulation model in the preset data assimilation model, the environmental simulation data of the target area is simulated based on the satellite remote sensing data and aerosol density data;

[0056] Using the improved Kalman filter in the preset data assimilation model, the environmental sensor data is corrected based on the environmental simulation data to obtain the target environmental data.

[0057] In this embodiment, environmental simulation data of the target area is simulated using satellite remote sensing data and aerosol density data, which serves as the basis for correcting the environmental sensor data, making the corrected environmental sensor data more accurate. At the same time, an improved Kalman filter is used to interpolate the environmental sensor data, which makes the corrected environmental sensor data smoother and more consistent with the actual environmental conditions of the target area.

[0058] Optionally, the environmental simulation model can be a machine learning-based inversion model. For example, historical satellite remote sensing data and aerosol density data can be used as model inputs, and historical environmental sensor data can be used as model outputs to train the inversion model until it reaches a preset convergence condition, thus obtaining the environmental simulation model.

[0059] In one embodiment, the environmental simulation data simulating the target area includes:

[0060] Feature extraction is performed on the satellite remote sensing data and aerosol density data to obtain simulated feature data;

[0061] Using the environmental simulation model, based on the simulation feature data, environmental simulation data of the target area is simulated. The environmental simulation model is an environmental meteorological inversion model.

[0062] In this embodiment, the simulated feature data includes the remote sensing feature data, aerosol feature data, first correlation feature data, and second correlation feature data.

[0063] Optionally, feature extraction includes:

[0064] Based on a preset feature extraction model, remote sensing feature data of the satellite remote sensing data and aerosol feature data of the aerosol density data are extracted.

[0065] Based on a preset radiative transfer model, the first correlation feature data between the satellite remote sensing data and the environmental meteorology of the target area is extracted;

[0066] Based on a preset aerosol model, a second correlation feature data between the aerosol density data and the environmental meteorology of the target area is extracted.

[0067] In this optional embodiment, the preset feature extraction model can be a time-series feature extraction model based on statistics or machine learning. The preset radiative transfer model includes an atmospheric transfer model and a surface reflection model; the atmospheric transfer model describes the attenuation and scattering of solar radiation through the atmosphere, and can be the US Standard Atmosphere model, MODTRAN model, 6S model, etc.; the surface reflection model describes the reflection and radiation state of solar radiation on the Earth's surface, and can be the Lambert model, expansion sphere model, scale curve model, etc. As one implementation, the preset radiative transfer model = first weight × output of atmospheric transfer model + second weight × surface reflection model.

[0068] The pre-defined aerosol model can be represented as:

[0069] ;

[0070] in, It is atmospheric path reflectance. and This represents the total transmittance, from the sun to the ground and from the ground to the sensor, respectively. It is the spherical albedo of atmospheric illumination. It is surface reflectivity. It is the zenith angle of the sun. It is the satellite zenith angle. This indicates the relative azimuth angle between the sun and the satellite.

[0071] In one embodiment, data correction includes:

[0072] The environmental sensor data is cleaned to remove abnormal data and to identify missing segments in the environmental sensor data after removing abnormal data.

[0073] The improved Kalman filter is used to correct missing segments in the environmental sensor data to obtain the target environmental data.

[0074] In this embodiment, the abnormal data can be mutation data. By improving the Kalman filter, the mutation data and the original missing segments are corrected, thereby improving the quality of environmental meteorological data.

[0075] In one embodiment, the step of using the improved Kalman filter to correct missing segments in the environmental sensor data to obtain the target environmental data includes:

[0076] Based on the current environmental simulation data, environmental sensor data, covariance matrix, and state transition equation, predict the environmental prediction data for the next time step.

[0077] Based on the environmental prediction data, update the covariance matrix for the next time step;

[0078] Calculate the Kalman gain matrix based on the covariance matrix at the next time step;

[0079] Based on the Kalman gain matrix, the environmental sensor data at the next time step is updated to correct the environmental sensor data and obtain the target environmental data.

[0080] Optionally, the environmental prediction data for the next time step includes:

[0081] Calculate the average value between the current environmental simulation data and the environmental sensor data;

[0082] Based on the covariance matrix, the mean data is subjected to an unscented transformation to obtain the sigma point set;

[0083] The sigma point set is input into the state transition equation to obtain the environmental prediction data for the next time step.

[0084] In this embodiment, the accuracy of the current state is improved by averaging the environmental simulation data and environmental sensor data, and the variable environmental weather characteristics of the generator set area are simulated by using traceless transformation to better match the environmental weather characteristics of the target area.

[0085] For example, the environmental simulation data and environmental sensor data at the current moment are weighted to obtain the mean data. :

[0086] ;

[0087] in, This represents the state at time k+1. This represents the system state at time k. Indicates process noise. This represents the state update function. Indicates measurement noise. This represents the state transition function.

[0088] Unscented transformation of the mean data yields sigma sampling points:

[0089] ;

[0090] in, The i-th sample of the sigma sampling point is obtained by performing an unscented transformation on the mean data Z(k / k); This represents the estimated value (prior value) of the mean data, where n represents the dimension of the state variable. Let represent the covariance matrix, used to represent the uncertainty of the state variable X; λ represents a constant used to control the range and direction of the sigma point distribution. Typically, λ takes the value 3-n, where n is the dimension of the state variable. Let P(k / k) be the standard deviation of the covariance matrix. This represents a measure of how the information from the covariance matrix is ​​transformed into standard deviation, used to expand or contract the matrix. The range.

[0091] Calculate 2n+1 sigma point sets:

[0092] ;in, The i-th sample of the sigma sampling point is obtained by performing an unscented transformation on the mean data Z(k / k); This represents the state update function. Indicates the current moment. This represents the i-th sample in the sigma point set, updated through the state update function. sigma sampling points The result obtained by performing state propagation.

[0093] State quantity prediction:

[0094] ;in, The predicted value of the state variable is obtained by weighted averaging of the sigma point set; 2n represents the total number of sigma points. The weight corresponding to the i-th sampling point is used to perform a weighted average of the state variables to obtain the predicted value of the state variables. This represents the estimated state value of the i-th sample in the sigma point set at the next time step, obtained through the state update function.

[0095] Covariance matrix:

[0096] ;in, This represents the covariance matrix at the next time step. This represents the weight corresponding to the i-th sampling point. The predicted values ​​of the state variables are obtained by weighted averaging of the sigma point set. Let represent the estimated state value of the i-th sample in the sigma point set at the next time step, obtained through the state update function. express transpose, The covariance matrix represents the process noise and describes the uncertainties not modeled in the model.

[0097] Another traceless transformation yields a new set of sigma points:

[0098] ;in, This represents the newly obtained set of sigma points; This represents the predicted value of the state variable. Let P(k / k) be the standard deviation of the covariance matrix P(k / k) at the next time step k+1.

[0099] The new sigma point set is input into the state transition equation to obtain environmental prediction data:

[0100] ;in, This means that in the state transition equation, based on the new sigma point set of the input... The obtained environmental prediction data, Represents the state transition function;

[0101] Calculate the new mean data:

[0102] ;

[0103] Update the predicted covariance matrix:

[0104] ;

[0105] Update of the measurement covariance matrix:

[0106] ;in, This represents the mean data obtained based on new environmental forecast data; Indicates weight; This represents new environmental forecast data; Represents the predicted covariance matrix. Represents the measurement covariance matrix. The covariance of environmental noise. Let i represent the i-th sample in the sigma point set. This represents the mean data obtained from a newly acquired set of sigma points. express The transpose of .

[0107] Calculate the Kalman gain matrix:

[0108] ;in, This represents the Kalman gain matrix at time k+1. This represents the covariance matrix, used to describe the variance and covariance between the state estimate and the sensor measurement; This represents the inverse of the covariance matrix, used to calculate the Kalman gain matrix.

[0109] Update the environmental sensor data for the next time step:

[0110] ;in, This represents the prior value of the state variable at the next time step k+1; This represents the predicted value of the state variable at the next time step k+1. Represents the Kalman gain matrix; This represents the new environmental prediction data obtained based on the sigma point set at time k+1; This represents the mean data obtained based on new environmental forecast data;

[0111] Update the covariance matrix for the next time step:

[0112] ;

[0113] in, Let the covariance matrix of the state estimate at time k+1 be represented. Let Kalman gain matrix be represented at time k+1. Represents the covariance matrix. This represents the transpose of the Kalman gain matrix.

[0114] To implement the new energy ultra-short-term power generation prediction method corresponding to the above method embodiments, and to achieve the corresponding functions and technical effects. See [link / reference needed]. Figure 2 , Figure 2 This diagram illustrates a structural block diagram of a new energy ultra-short-term power generation prediction device according to an embodiment of this application. For ease of explanation, only the parts relevant to this embodiment are shown. The new energy ultra-short-term power generation prediction device provided in this embodiment includes:

[0115] The acquisition module 201 is used to acquire environmental sensor data, satellite remote sensing data and aerosol density data of the target area where the new energy generator set is located;

[0116] The correction module 202 is used to correct the environmental sensor data based on the satellite remote sensing data and aerosol density data using a preset data assimilation model to obtain target environmental data.

[0117] Extraction module 203 is used to extract features from the target environment data to obtain environmental feature data of the target area;

[0118] The prediction module 204 is used to predict the power generation data of the new energy generator set in a short period of time in the future based on the environmental characteristic data using a preset ultra-short-term power prediction model.

[0119] In one embodiment, the correction module 202 includes:

[0120] The simulation submodule is used to simulate the environmental simulation data of the target area based on the satellite remote sensing data and aerosol density data, using the environmental simulation model in the preset data assimilation model.

[0121] The correction submodule is used to correct the environmental sensor data based on the environmental simulation data using the improved Kalman filter in the preset data assimilation model to obtain the target environmental data.

[0122] In one embodiment, the simulation submodule includes:

[0123] An extraction unit is used to extract features from the satellite remote sensing data and aerosol density data to obtain simulated feature data.

[0124] The simulation unit is used to simulate environmental simulation data of the target area using the environmental simulation model and the simulation feature data. The environmental simulation model is an environmental meteorological inversion model.

[0125] In one embodiment, the extraction unit is specifically used for:

[0126] Based on a preset feature extraction model, remote sensing feature data of the satellite remote sensing data and aerosol feature data of the aerosol density data are extracted.

[0127] Based on a preset radiative transfer model, the first correlation feature data between the satellite remote sensing data and the environmental meteorology of the target area is extracted;

[0128] Based on a preset aerosol model, a second correlation feature data is extracted between the aerosol intensity data and the environmental meteorology of the target area. The simulated feature data includes the remote sensing feature data, aerosol feature data, first correlation feature data, and second correlation feature data.

[0129] In one embodiment, the correction submodule includes:

[0130] The cleaning unit is used to clean the environmental sensor data, remove abnormal data from the environmental sensor data, and determine the missing segments in the environmental sensor data after removing abnormal data.

[0131] The correction unit is used to correct missing segments in the environmental sensor data using the improved Kalman filter to obtain the target environmental data.

[0132] In one embodiment, the correction unit includes:

[0133] The prediction subunit is used to predict the environmental prediction data for the next time step based on the environmental simulation data, environmental sensor data, covariance matrix, and state transition equation at the current time step.

[0134] The update sub-unit is used to update the covariance matrix at the next time step based on the environmental prediction data;

[0135] The computational subunit is used to calculate the Kalman gain matrix based on the covariance matrix at the next time step.

[0136] The correction subunit is used to update the environmental sensor data at the next time step based on the Kalman gain matrix, so as to correct the environmental sensor data and obtain the target environmental data.

[0137] In one embodiment, the prediction subunit is specifically used for:

[0138] Calculate the average value between the current environmental simulation data and the environmental sensor data;

[0139] Based on the covariance matrix, the mean data is subjected to an unscented transformation to obtain the sigma point set;

[0140] The sigma point set is input into the state transition equation to obtain the environmental prediction data for the next time step.

[0141] The aforementioned new energy ultra-short-term power generation prediction device can implement the new energy ultra-short-term power generation prediction method of the above method embodiments. The options in the above method embodiments are also applicable to this embodiment, and will not be detailed here. The remaining content of this application embodiment can be referred to the content of the above method embodiments, and will not be repeated in this embodiment.

[0142] Figure 3 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Figure 3 As shown, the computer device 3 of this embodiment includes: at least one processor 30 ( Figure 3(Only one is shown in the diagram), memory 31, and computer program 32 stored in said memory 31 and executable on said at least one processor 30, wherein said processor 30 executes said computer program 32 to implement the steps in any of the above method embodiments.

[0143] The computer device 3 can be a smartphone, tablet, desktop computer, cloud server, or other computing device. This computer device may include, but is not limited to, a processor 30 and a memory 31. Those skilled in the art will understand that... Figure 3 The computer device 3 is merely an example and does not constitute a limitation on the computer device 3. It may include more or fewer components than shown in the figure, or combine certain components, or different components, such as input / output devices, network access devices, etc.

[0144] The processor 30 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0145] In some embodiments, the memory 31 may be an internal storage unit of the computer device 3, such as a hard disk or memory of the computer device 3. In other embodiments, the memory 31 may be an external storage device of the computer device 3, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the computer device 3. Furthermore, the memory 31 may include both internal and external storage units of the computer device 3. The memory 31 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory 31 can also be used to temporarily store data that has been output or will be output.

[0146] In addition, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in any of the above method embodiments.

[0147] This application provides a computer program product that, when run on a computer device, enables the computer device to execute the steps described in the various method embodiments above.

[0148] In the several embodiments provided in this application, it will be understood that each block in the flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the figures. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved.

[0149] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0150] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above descriptions are merely specific embodiments of this application and are not intended to limit the scope of protection of this application. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application for those skilled in the art.

Claims

1. A new energy ultra-short-term power generation power prediction method, characterized in that, The method comprises the following steps: Obtain environmental sensor data, satellite remote sensing data and aerosol degree data of a target area where a new energy generator set is located; Use a preset data assimilation model to correct the environmental sensor data based on the satellite remote sensing data and the aerosol degree data, and obtain target environmental data, which comprises the following steps: Extract remote sensing feature data of the satellite remote sensing data and aerosol feature data of the aerosol degree data based on a preset feature extraction model; Extract first correlation feature data between the satellite remote sensing data and the environmental meteorology of the target area based on a preset radiation transmission model; Extract second correlation feature data between the aerosol degree data and the environmental meteorology of the target area based on a preset aerosol model, and the simulation feature data comprises the remote sensing feature data, the aerosol feature data, the first correlation feature data and the second correlation feature data; use an environmental simulation model to simulate environmental simulation data of the target area according to the simulation feature data, and the environmental simulation model is an environmental meteorology inversion model; The preset aerosol model is as follows: ； wherein, is the atmospheric path reflectance, and denotes the total transmittance from the sun to the ground and from the ground to the sensor, respectively, is the spherical albedo of the atmospheric illumination, is the surface reflectance, is the solar zenith angle, is the satellite zenith angle, denotes the solar / satellite relative azimuth angle; Use an improved Kalman filter in the preset data assimilation model to correct the environmental sensor data based on the environmental simulation data, and obtain the target environmental data; Predict environmental prediction data at the next moment based on the environmental simulation data, the environmental sensor data, a covariance matrix and a state transition equation at the current moment; Update the covariance matrix at the next moment based on the environmental prediction data; Calculate a Kalman gain matrix based on the covariance matrix at the next moment; Update the environmental sensor data at the next moment based on the Kalman gain matrix to correct the environmental sensor data, and obtain the target environmental data; Extract environmental feature data of the target area by extracting features of the target environmental data; Use a preset ultra-short-term power prediction model to predict power generation data of the new energy generator set in a future short time period according to the environmental feature data.

2. The new energy ultra-short-term power generation power prediction method of claim 1, wherein, The method comprises the following steps: Use an environmental simulation model in the preset data assimilation model to simulate environmental simulation data of the target area according to the satellite remote sensing data and the aerosol degree data; Use an improved Kalman filter in the preset data assimilation model to correct the environmental sensor data based on the environmental simulation data, and obtain the target environmental data.

3. The new energy ultra-short-term power generation power prediction method of claim 2, wherein, The method comprises the following steps: Extract simulation feature data by extracting features of the satellite remote sensing data and the aerosol degree data; Use an environmental simulation model to simulate environmental simulation data of the target area according to the simulation feature data, and the environmental simulation model is an environmental meteorology inversion model.

4. The new energy ultra-short-term power generation power prediction method of claim 2, wherein, The improved Kalman filter in the preset data assimilation model is used to correct the environment sensor data based on the environment simulation data, to obtain the target environment data, including: The environment sensor data is data cleaned, and the abnormal data in the environment sensor data is removed, and the missing section in the environment sensor data after removing the abnormal data is determined; The improved Kalman filter is used to correct the missing section in the environment sensor data, to obtain the target environment data.

5. The new energy ultra-short-term power generation power prediction method of claim 1, wherein, The environment prediction data at the next moment is predicted based on the environment simulation data, environment sensor data and covariance matrix at the current moment, including: The mean value data between the environment simulation data and environment sensor data at the current moment is calculated; The sigma point set is obtained by performing unscented transformation on the mean value data based on the covariance matrix; The sigma point set is input into the state transition equation to obtain the environment prediction data at the next moment.

6. A new energy ultra-short-term power generation prediction device characterized by comprising: It includes: The environment sensor data, satellite remote sensing data and aerosol degree data of the target area where the new energy generator set is located are acquired by the acquisition module; The environment sensor data is corrected based on the satellite remote sensing data and aerosol degree data by using the preset data assimilation model to obtain the target environment data, and the remote sensing feature data of the satellite remote sensing data and the aerosol feature data of the aerosol degree data are extracted based on the preset feature extraction model; The first correlation feature data between the satellite remote sensing data and the environment meteorology of the target area is extracted based on the preset radiation transmission model; The second correlation feature data between the aerosol degree data and the environment meteorology of the target area is extracted based on the preset aerosol model, and the simulation feature data includes the remote sensing feature data, aerosol feature data, first correlation feature data and second correlation feature data; the environment simulation data of the target area is simulated according to the simulation feature data by using the environment simulation model, and the environment simulation model is an environment meteorology inversion model; The preset aerosol model is: The improved Kalman filter in the preset data assimilation model is used to correct the environment sensor data based on the environment simulation data, to obtain the target environment data; ； wherein, is the atmospheric path reflectance, and denotes the total transmittance, respectively from the sun to the ground and from the ground to the sensor, is the spherical albedo of the atmospheric illumination, is the surface reflectance, is the solar zenith angle, is the satellite zenith angle, denotes the solar / satellite relative azimuth angle; The environment prediction data at the next moment is predicted based on the environment simulation data, environment sensor data, covariance matrix and state transition equation at the current moment; The covariance matrix at the next moment is updated based on the environment prediction data; The Kalman gain matrix is calculated based on the covariance matrix at the next moment; The environment sensor data at the next moment is updated based on the Kalman gain matrix to correct the environment sensor data, to obtain the target environment data; The feature extraction module is used to extract features from the target environment data to obtain the environment feature data of the target area; The prediction module is used to predict the power generation data of the new energy generator set in the future short period of time according to the environment feature data by using the preset ultra-short-term power prediction model. ​ 7. A computer device, characterized by The application relates to a computer program product, comprising a processor and a memory for storing a computer program, wherein the computer program is executed by the processor to realize the new energy ultra-short-term power generation power prediction method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The application relates to a computer program product, comprising a processor and a memory for storing a computer program, wherein the computer program is executed by the processor to realize the new energy ultra-short-term power generation power prediction method according to any one of claims 1 to 5.

Images