Power load prediction method and system based on meteorological prediction model

Abstract

The invention discloses a power load prediction method and system based on a meteorological prediction model, and relates to the technical field of load prediction, and the method comprises the following steps: obtaining historical power load data and corresponding multi-dimensional meteorological data, constructing a meteorological coupling effect analysis model, calculating a dynamic coupling weight matrix among the meteorological parameters through an attention mechanism network; performing feature fusion on the coupling weight matrix and the multi-dimensional meteorological data through a gating circulation unit to generate a space-time coupling feature vector; establishing a power load prediction model, and extracting a short-term fluctuation feature and a long-term trend feature of the power load from the space-time coupling feature vector based on a deep residual network architecture and a parallel convolution channel; the dynamic coupling weight matrix is updated based on prediction error back propagation, and adaptive adjustment of the meteorological parameter coupling relation is achieved. According to the method, dynamic fusion of the spatial-temporal characteristics is realized through the cross gating unit, so that the model can perform power load prediction by using spatial-temporal information, and the prediction precision is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of load forecasting, and more specifically, to a method and system for forecasting electric load based on a meteorological forecasting model. Background Art

[0002] In today's society, the stability and reliability of power supply play a key supporting role in economic development and people's daily lives. With the acceleration of global industrialization and urbanization, the scale of power demand continues to rise and the fluctuation characteristics become more complex. If the power system is to achieve safe and efficient operation, accurate prediction of power load has become an indispensable link, which can provide key basis for power generation planning, grid dispatching and infrastructure planning.

[0003] There is a close connection between meteorological conditions and power load. In terms of temperature, when the temperature deviates from the human comfort zone, whether it is the hot summer that prompts a large number of air-conditioning and refrigeration equipment to operate, or the cold winter that leads to the widespread use of heating equipment, it will cause a significant change in power load. Studies have shown that in some high-temperature areas, the power load may increase by several megawatts for every 1°C increase in summer temperature. Humidity cannot be ignored either. People may use dehumidification equipment in high humidity environments, and humidity will affect the heat dissipation of electrical equipment, indirectly changing the demand for electricity. In addition, meteorological factors such as wind speed and sunshine duration also affect the electricity consumption of places that rely on natural ventilation, as well as the power load of solar photovoltaic power generation connected to the grid.

[0004] For example, the invention patent with announcement number: CN113868817A discloses a method and system for predicting the stability of a power system based on an industrial power load simulation model. The method includes: constructing an equivalent circuit of the industrial equipment based on the voltage and phase current of the industrial equipment; using Kirchhoff's theorem to obtain an industrial power load simulation model based on the equivalent circuit of the industrial equipment; using the industrial power load simulation model to predict the three-phase voltage, three-phase current and power data of the industrial equipment based on the real-time three-phase current data of the industrial equipment; and evaluating whether the power system is stable based on the three-phase voltage, three-phase current and power data of the industrial equipment.

[0005] For example, the invention patent with announcement number CN119442049A announces a method and system for establishing a power load forecasting model based on regional characteristics and meteorology, through the following steps: S1, establishing an influence index system based on regional characteristics and meteorological data; S2, obtaining a historical sample data set based on the influence index system; S3, constructing a machine learning model based on the random forest algorithm, and optimizing the model parameters through the historical sample data set to obtain a power load forecasting model; S4, evaluating the power load forecasting model; S5, forecasting the power load through the power load forecasting model and the influence index system. The prediction is performed using multi-dimensional attribute variables such as regional power building parameters, outdoor meteorological parameters, time parameters, and power consumption parameters at time Tt, and a prediction model is established through a random forest algorithm, which improves the prediction efficiency and prediction accuracy.

[0006] The above disclosed technical solutions have at least the following technical problems: there are coupling effects between multiple meteorological factors, such as temperature and humidity, wind speed and air pressure, which interact with each other and affect the power load, but the existing technology often ignores these coupling effects, resulting in inaccurate prediction results. Traditional methods only integrate spatiotemporal features through simple splicing or single gating, without considering the dynamic modulation of the coupling effect of meteorological parameters. In view of the above problems, the present invention proposes a solution. Summary of the invention

[0007] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a power load forecasting method and system based on a meteorological forecasting model. By constructing a meteorological coupling effect analysis model and calculating the dynamic coupling weight matrix between various meteorological parameters through an attention mechanism network, the problem of dynamic modulation without considering the coupling effect of meteorological parameters is solved.

[0008] To achieve the above object, the present invention provides the following technical solutions: The method for predicting power load based on meteorological prediction model comprises the following steps: obtaining historical power load data and corresponding multidimensional meteorological data, wherein the multidimensional meteorological data comprises four meteorological parameters: temperature, humidity, wind speed and air pressure; constructing a meteorological coupling effect analysis model, and calculating the dynamic coupling weight matrix between meteorological parameters through an attention mechanism network; performing feature fusion of the coupling weight matrix and the multidimensional meteorological data through a gated recurrent unit to generate a spatiotemporal coupling feature vector; establishing a power load prediction model, and extracting short-term fluctuation characteristics and long-term trend characteristics of power load from the spatiotemporal coupling feature vector based on a deep residual network architecture and parallel convolution channels; and updating the dynamic coupling weight matrix based on the back propagation of prediction errors to achieve adaptive adjustment of the coupling relationship of meteorological parameters.

[0009] In a preferred embodiment, the meteorological coupling effect analysis model is constructed, and the dynamic coupling weight matrix between each meteorological parameter is calculated through the attention mechanism network, specifically: a dual-stream attention mechanism is adopted, and analysis is performed based on the time series to obtain a first attention weight matrix and a second attention weight matrix; the first attention weight matrix and the second attention weight matrix are cascaded on the feature dimension to obtain a cascaded feature matrix; based on the mapping relationship between the feature matrix and the sigmoid function, a dynamic weight allocation coefficient matrix is ​​obtained; based on the dynamic weight allocation coefficient matrix and the physical constraints of the meteorological parameters, a constrained weight correction function is established to obtain the final coupling weight matrix.

[0010] In a preferred embodiment, the first attention weight matrix and the second attention weight matrix are obtained in the following specific steps: first attention weight matrix: input the multidimensional meteorological data into the first attention flow module based on the time series; calculate the linear correlation coefficient between each meteorological parameter based on each time step in the time series; normalize the linear correlation coefficient through linear transformation and softmax function to obtain the first attention weight matrix based on the linear relationship between each meteorological parameter; Second attention weight matrix: The input meteorological data is feature extracted through the second attention flow module and combined into a nonlinear interaction feature vector; the second attention weight matrix is ​​obtained by calculating the similarity between the nonlinear interaction feature vector and the learnable query vector and normalizing it with the softmax function.

[0011] In a preferred embodiment, the coupling weight matrix and the multidimensional meteorological data are subjected to feature fusion through a gated recurrent unit to generate a spatiotemporal coupling feature vector, specifically: multi-scale time information features are obtained, a meteorological site topology map network is constructed according to the geographic location information of the meteorological site, and spatial correlation features are generated through a dynamic graph attention network; the spatial correlation features are input into the feature interaction layer, and the fusion ratio of the spatiotemporal features is controlled through a cross-gated unit to generate a spatiotemporal coupling feature vector.

[0012] In a preferred embodiment, the specific steps of obtaining multi-scale time information features are as follows: aligning the coupling weight matrix dimension and the multidimensional meteorological data based on the time step to obtain each time step, and based on the dilated causal convolution and the preset convolution kernel size and dilation rate, performing a convolution operation on the meteorological data of the historical time step to obtain the multi-scale time information features.

[0013] In a preferred embodiment, the spatial correlation features are generated through a dynamic graph attention network, specifically: features are obtained through a time dimension feature extraction layer and used as input to a graph attention mechanism; for each node and the graph attention mechanism, the attention weight of the node is calculated; based on the attention weight, the features of each node are weightedly aggregated to obtain spatial correlation features.

[0014] In a preferred embodiment, the fusion ratio of spatiotemporal features is controlled by a cross-gating unit to generate a spatiotemporal coupling feature vector, specifically: multi-scale time information features and spatial correlation features are fused to obtain a fused feature; an outer product operation is performed on the fused feature to obtain an outer product matrix, and the outer product matrices of all time steps are stacked to obtain an outer product tensor; a preset learning weight matrix and the outer product tensor are tensor multiplied to generate a spatiotemporal interaction gating signal; features are extracted from the coupling weight matrix, each feature is vectorized to obtain a coupling vector, and the coupling vectors of all time steps are stacked to obtain a coupling tensor; the coupling tensor is processed by a multi-layer perceptron to generate a coupling modulation gating signal; the spatiotemporal interaction gating signal and the coupling modulation gating signal are weightedly spliced ​​with the original meteorological data and the coupling weight matrix respectively to obtain a spatiotemporal coupling feature vector.

[0015] In a preferred embodiment, the parallel convolution channel extracts the short-term fluctuation characteristics and long-term trend characteristics of the power load from the spatiotemporal coupling feature vector, specifically: decomposing and preprocessing the input features, and performing mutual compensation of time-frequency domain features to obtain short-term fluctuation flow and long-term trend flow; splicing and fusing the features of the short-term fluctuation flow and the long-term trend flow based on a dual-path residual unit; mapping the fused features to the power load prediction value, compressing the feature dimension, and obtaining the final load prediction value.

[0016] In a preferred embodiment, the dynamic coupling weight matrix is ​​updated based on the back propagation of the prediction error, specifically: calculating the mean square error between the load prediction value and the true value; taking the partial derivative of the mean square error with respect to each element of the coupling weight matrix based on the chain rule to obtain the gradient of the coupling weight matrix; calculating the first-order moment estimate and the second-order moment estimate of the gradient; and performing bias correction on the first-order moment estimate and the second-order moment estimate to obtain an updated coupling weight matrix.

[0017] Technical effects and advantages of the power load forecasting method and system based on the meteorological forecast model of the present invention: 1. The present invention constructs a meteorological coupling effect analysis model and calculates the dynamic coupling weight matrix between various meteorological parameters through an attention mechanism network, fully considering the interaction and coupling relationship between meteorological parameters such as temperature, humidity, wind speed, and air pressure. It is more comprehensive and accurate than considering the impact of each meteorological parameter on the power load separately, and can more accurately capture the comprehensive impact of meteorological factors on the power load, thereby improving the accuracy of the prediction.

[0018] 2. The present invention updates the coupling weight matrix through back propagation of prediction errors, calculates the mean square error between the load prediction value and the true value, obtains the gradient by taking the partial derivative of the mean square error with respect to the coupling weight matrix, and performs deviation correction on the first-order moment estimate and the second-order moment estimate of the gradient to update the coupling weight matrix, so that the model can automatically adjust the coupling relationship between meteorological parameters according to the prediction error. With the continuous input of data and continuous training of the model, the model can adaptively learn the dynamic change relationship between meteorological parameters and power load, thereby maintaining good prediction performance under different meteorological conditions and times. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the flow of the power load forecasting method based on the meteorological forecasting model of the present invention; Figure 2 The diagram is a schematic diagram of the structure of the power load forecasting system based on the meteorological forecasting model of the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] Embodiment 1, Figure 1 The present invention provides a method for predicting power load based on a meteorological prediction model, comprising the following steps: S1, obtaining historical power load data and corresponding multi-dimensional meteorological data, wherein the multi-dimensional meteorological data includes four meteorological parameters: temperature, humidity, wind speed, and air pressure; Meteorological parameters are mutually coupled, and coupling parameters are used to describe the degree of interaction and mutual influence between multiple variables or systems. In the context of meteorology and power load, the four meteorological factors of temperature, humidity, wind speed, and air pressure are used as coupling parameters, and there are mutual relationships between them and between them and power load.

[0022] Coupling of temperature and electrical load: Temperature is an important factor affecting power load. In hot weather, the use of cooling equipment such as air conditioners increases, and the power load rises; in cold weather, the use of heating equipment will also increase the power load. Usually there is a certain functional relationship, which can be described by establishing a load-temperature model. For example, a linear or nonlinear regression model is used to determine the approximate change in power load for every certain degree of temperature change. The coefficient related to this change is a coupling parameter.

[0023] Coupling of humidity and electrical loads: Humidity affects human comfort and the operating environment of equipment. In a high humidity environment, people may use more dehumidification equipment or change the cooling demand for air conditioning, which will affect the power load. At the same time, humidity also affects the heat dissipation and other performance of some electrical equipment, which in turn indirectly affects the power load. The correlation coefficient between humidity and power load can be determined through experiments or data analysis as a coupling parameter to measure the degree of coupling between them.

[0024] Coupling of wind speed and power load: The impact of wind speed on power load is mainly reflected in heat dissipation and ventilation. For some large industrial equipment or outdoor power facilities, the wind speed will affect the heat dissipation effect, which in turn affects the operating power and power consumption of the equipment. In cities, wind speed may also affect the ventilation needs of buildings, indirectly affecting the use of equipment such as air conditioners and power load. By studying the corresponding relationship between wind speed changes and power load changes, we can determine the proportion of power load changes for each increase in wind speed as a coupling parameter.

[0025] Coupling of air pressure and electrical load: Changes in air pressure may be related to changes in weather systems, which in turn affect people's lives and production activities, leading to changes in power load. For example, when the air pressure is low, the weather may be hot and muggy, and the power load may increase due to increased cooling demand. By analyzing the fluctuations in air pressure and power load in historical data, parameters reflecting the degree of mutual influence between them are obtained, such as the covariance between air pressure changes and power load changes as coupling parameters.

[0026] S2, build a meteorological coupling effect analysis model and calculate the coupling weight matrix between meteorological parameters through the attention mechanism network; The meteorological coupling effect analysis model is established, and the dynamic coupling weight matrix between various meteorological parameters is calculated through the attention mechanism network, specifically: Using a two-stream attention mechanism, multi-dimensional meteorological data is input into the first attention stream module based on time series; For each time step in the time series, the linear correlation coefficient between each meteorological parameter is calculated; The linear correlation coefficient is normalized by linear transformation and softmax function to obtain a first attention weight matrix based on the linear relationship between the meteorological parameters; The input meteorological data is feature extracted through the second attention flow module and combined into a nonlinear interaction feature vector; The second attention weight matrix is ​​obtained by calculating the similarity between the nonlinear interaction feature vector and the learnable query vector and normalizing it with the softmax function. Perform a cascade operation on the first attention weight matrix output by the first attention stream and the second attention weight matrix output by the second attention stream in the feature dimension to obtain a cascaded feature matrix; The feature matrix is ​​input into the fully connected layer, and the output is mapped to the interval [0, 1] based on the sigmoid function through linear transformation to obtain a dynamic weight allocation coefficient matrix; The physical constraints of meteorological parameters are obtained, a constrained weight correction function is established, and the final coupling weight matrix is ​​obtained.

[0027] The coupling weight matrix includes temperature-humidity interaction terms and wind speed-pressure gradient terms; The temperature-humidity interaction term is specifically: For each time step, the corresponding theoretical value of saturated water vapor pressure is first calculated based on the temperature data; Then the actual measured humidity data is obtained and the actual saturated water vapor pressure is calculated; The ratio of the temperature change to the saturated water vapor pressure change is calculated to obtain a value reflecting the temperature-humidity interaction relationship, which is used as the element value of the temperature-humidity interaction term in the coupling weight matrix.

[0028] The theoretical value of saturated water vapor pressure is specifically:

[0029] The actual saturated water vapor pressure is specifically:

[0030] in, is the theoretical value of saturated water vapor pressure, is the saturated water vapor pressure under standard conditions, is an exponential function, is the latent heat of vaporization of water vapor, is the specific gas constant of water vapor, is the temperature under standard conditions, is the current temperature, is the actual saturated water vapor pressure, is the relative humidity.

[0031] The wind speed-pressure gradient term is specifically: Get the wind speed value for each time step, and get the pressure gradient by dividing the pressure difference between adjacent meteorological stations by the distance between the stations; The covariance between wind speed and pressure gradient is calculated to obtain the element value of wind speed-pressure gradient term.

[0032] The constraints include: The temperature-humidity interaction term satisfies the Clausius-Clapeyron equation constraint, the wind speed-pressure gradient term complies with the geostrophic wind balance equation constraint, and the weight coefficients of each coupling term maintain the non-negativity constraint.

[0033] The linear correlation coefficient matrix is ​​R. After linear transformation W1*R+b1 (W1 is the weight matrix, b1 is the bias vector), the attention weight matrix A1 is obtained through the softmax function softmax(W1*R+b1).

[0034] A dual-stream attention mechanism is adopted. The first attention stream calculates the linear correlation coefficient between parameters, and the second attention stream extracts nonlinear interaction features. The dual-stream attention output is fused through the feature cascade layer to generate dynamic weight allocation coefficients; Introduce physical constraints on meteorological parameters and establish a weight correction function with constraints to ensure that the coupling relationship conforms to the laws of atmospheric physics. The linear correlation coefficient between the meteorological parameters is calculated as follows:

[0035] in, is the linear correlation coefficient, is the value of meteorological parameter X at time step i, is the mean value of meteorological parameter X, is the value of meteorological parameter Y at time step i, is the mean value of meteorological parameter Y, is the number of time steps.

[0036] S3, feature fusion of the coupling weight matrix and multidimensional meteorological data through a gated recurrent unit to generate a spatiotemporal coupling feature vector; The coupling weight matrix and the multidimensional meteorological data are subjected to feature fusion through a gated recurrent unit to generate a spatiotemporal coupling feature vector, specifically: Align the coupling weight matrix dimensions and multidimensional meteorological data based on time steps; Inputting the original meteorological data into a time dimension feature extraction layer, wherein the time dimension feature extraction layer adopts dilated causal convolution; For each time step, the dilated causal convolution performs a convolution operation on the meteorological data of the historical time step according to the pre-set convolution kernel size and dilation rate to obtain multi-scale time information features; According to the geographical location information of the meteorological stations, a meteorological station topological network is constructed, and spatial correlation features are generated through a dynamic graph attention network; The spatial correlation features are input into the feature interaction layer, and the dynamic fusion of spatiotemporal features is realized through the cross-gating unit; The cross-gating unit controls the fusion ratio of spatiotemporal features by calculating the gating signal between temporal features and spatial features, thereby generating a spatiotemporal coupling feature vector.

[0037] The spatial correlation features are generated by the dynamic graph attention network, specifically: The feature representation obtained by the time dimension feature extraction layer is used as the input of the graph attention mechanism; For each node (weather station), the graph attention mechanism calculates the attention weight of the node based on the characteristics of neighboring nodes and the weight of the edges; Through the graph attention mechanism, the features of each node are weighted and aggregated to obtain spatial correlation features.

[0038] The cross-gating unit controls the fusion ratio of the temporal and spatial features by calculating the gating signal between the temporal features and the spatial features, thereby generating a temporal and spatial coupling feature vector, specifically: The multi-scale temporal information features and spatial correlation features are fused to obtain fusion features; Perform outer product operation on the fused features to obtain the outer product matrix, and stack the outer product matrices of all time steps to obtain the outer product tensor; Perform tensor multiplication of the preset learning weight matrix and the outer product tensor to generate a spatiotemporal interactive gating signal; Extract features from the coupling weight matrix, perform vectorization operations on each feature to obtain a coupling vector, and stack the coupling vectors of all time steps to obtain a coupling tensor; The coupling tensor is processed by a multi-layer perceptron to generate a coupling modulation gating signal; The spatiotemporal interaction gating signal and the coupled modulation gating signal are weightedly concatenated with the original meteorological data and the coupling weight matrix respectively to obtain the spatiotemporal coupling feature vector.

[0039] In the meteorological site topology network, each node represents a meteorological site, and the edges between nodes represent the spatial association between sites. The weight of the edge can be set according to factors such as the distance between sites and terrain; For missing values, linear interpolation or predicted values ​​based on time series models (such as ARIMA models) can be used to fill in the data according to the time correlation; for outliers, statistical methods (such as the 3σ principle) can be used to identify and replace them with reasonable values. Then, the data is normalized, for example, the minimum-maximum normalization method is used to map the values ​​of each meteorological parameter to the [0, 1] interval to ensure that meteorological parameters of different dimensions can participate in subsequent calculations at the same scale.

[0040] S4, establish a power load forecasting model, extract the short-term fluctuation characteristics and long-term trend characteristics of power load from the spatiotemporal coupling feature vector based on the deep residual network architecture and parallel convolution channels; The parallel convolution channel extracts the short-term fluctuation characteristics and long-term trend characteristics of the power load from the spatiotemporal coupling feature vector, specifically: The input features are decomposed into high-frequency short-term fluctuation flow and low-frequency long-term trend flow, which are preprocessed by wavelet packet transform and moving average filtering respectively; The time-frequency domain features are mutually compensated by inverse wavelet transform and fast Fourier transform, that is, the time-domain features are enhanced by using the frequency-domain information; Based on dual-path residual units, short-term fluctuation flow and long-term trend flow are processed separately, and each path adopts cascaded dilated convolution and deconvolution structure; The features of the short-term fluctuation flow and the long-term trend flow processed by the dual-path residual unit are spliced ​​and fused; The output layer maps the fused features to the power load forecast value, and uses the fully connected layer to compress the feature dimension to 1 to obtain the final load forecast value.

[0041] Cascaded dilated convolution: In each channel, cascaded dilated convolution layers are used to capture features of different scales. Dilated convolution can increase the receptive field of the convolution kernel without increasing the number of parameters, thereby better capturing long-term dependencies and short-term fluctuations. For example, different dilation rates (such as 1, 2, 4, etc.) are set to construct multiple dilated convolution layers.

[0042] Deconvolution structure: At the end of each path, a deconvolution layer is used to restore the dimension of the feature to match the requirements of the output layer. Deconvolution can reduce the dimensional feature mapping to a high-dimensional space, which helps improve the accuracy of the prediction.

[0043] S5, based on the back propagation of the prediction error, the dynamic coupling weight matrix is ​​updated to achieve adaptive adjustment of the coupling relationship of meteorological parameters.

[0044] The dynamic coupling weight matrix is ​​updated based on the prediction error back propagation, specifically: Calculate the mean square error between the load forecast value and the actual value; Based on the chain rule, the partial derivative of the mean square error with respect to each element of the coupling weight matrix is ​​calculated to obtain the gradient of the coupling weight matrix; Compute the first moment estimate (mean) and the second moment estimate (uncentered variance) of the gradient; The first-order moment estimation and the second-order moment estimation are bias-corrected to obtain an updated coupling weight matrix.

[0045] The first-order moment estimation and the second-order moment estimation of the calculated gradient are specifically:

[0046] in, is the first moment estimate of the gradient, is the second-order moment estimate of the gradient, and are the attenuation rates, which are 0.9 and 0.99 respectively, is the gradient of the coupling weight matrix, and are the first-order moment estimate and the second-order moment estimate of the gradient of the previous iteration, respectively. is the number of iterations.

[0047] The first-order moment estimation and the second-order moment estimation are subjected to deviation correction to obtain an updated coupling weight matrix, which is specifically:

[0048]

[0049] in, is the bias-corrected first-order moment estimate, is the bias-corrected second-order moment estimate, and are the deviation correction factors, is the updated coupling weight matrix, is the coupling weight matrix before updating, is a small constant, is the learning rate, is the number of iterations.

[0050] Embodiment 2, a power load forecasting system based on a meteorological forecasting model, comprises the following modules: Data acquisition module: used to obtain historical power load data and corresponding multi-dimensional meteorological data, the multi-dimensional meteorological data including four meteorological parameters: temperature, humidity, wind speed and air pressure; Meteorological coupling weight calculation module: used to build a meteorological coupling effect analysis model and calculate the dynamic coupling weight matrix between meteorological parameters through the attention mechanism network; Spatiotemporal feature fusion module: used to fuse the coupling weight matrix and multidimensional meteorological data through a gated recurrent unit to generate a spatiotemporal coupling feature vector; Power load forecasting modeling module: used to establish a power load forecasting model, extracting short-term fluctuation characteristics and long-term trend characteristics of power load from spatiotemporal coupling feature vectors based on a deep residual network architecture and parallel convolution channels; Coupling weight adaptive update module: used to update the dynamic coupling weight matrix based on the back propagation of prediction errors to achieve adaptive adjustment of the coupling relationship of meteorological parameters.

[0051] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0052] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0053] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0054] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0055] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0056] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting power load based on a meteorological prediction model, characterized in that: The following steps are involved: Acquire historical power load data and corresponding multi-dimensional meteorological data, wherein the multi-dimensional meteorological data includes four meteorological parameters: temperature, humidity, wind speed, and air pressure; Construct a meteorological coupling effect analysis model and calculate the dynamic coupling weight matrix between meteorological parameters through the attention mechanism network; The coupling weight matrix and multi-dimensional meteorological data are fused through gated recurrent units to generate spatiotemporal coupling feature vectors. Establish a power load forecasting model to extract short-term fluctuation characteristics and long-term trend characteristics of power load from spatiotemporal coupling feature vectors based on a deep residual network architecture and parallel convolution channels; The dynamic coupling weight matrix is ​​updated based on the back propagation of prediction errors to achieve adaptive adjustment of the coupling relationship of meteorological parameters.

2. The method for predicting power load based on a meteorological prediction model according to claim 1, characterized in that: The meteorological coupling effect analysis model is constructed to calculate the dynamic coupling weight matrix between meteorological parameters through the attention mechanism network, specifically: A dual-stream attention mechanism is used, and the first attention weight matrix and the second attention weight matrix are obtained based on time series analysis; Perform a cascade operation on the first attention weight matrix and the second attention weight matrix in the feature dimension to obtain a cascaded feature matrix; Based on the mapping relationship between the feature matrix and the sigmoid function, the dynamic weight allocation coefficient matrix is ​​obtained; Based on the dynamic weight allocation coefficient matrix and the physical constraints of meteorological parameters, a constrained weight correction function is established to obtain the final coupling weight matrix.

3. The method for predicting power load based on a meteorological prediction model according to claim 2, characterized in that: The specific steps of obtaining the first attention weight matrix and the second attention weight matrix are as follows: First attention weight matrix: Input the multidimensional meteorological data into the first attention flow module based on the time series; Based on each time step in the time series, the linear correlation coefficient between each meteorological parameter is calculated; The linear correlation coefficient is normalized by linear transformation and softmax function to obtain a first attention weight matrix based on the linear relationship between the meteorological parameters; Second attention weight matrix: The input meteorological data is feature extracted through the second attention flow module and combined into a nonlinear interaction feature vector; The second attention weight matrix is ​​obtained by calculating the similarity between the nonlinear interaction feature vector and the learnable query vector and normalizing it with the softmax function.

4. The method for predicting power load based on a meteorological prediction model according to claim 3, characterized in that: The coupling weight matrix and the multidimensional meteorological data are subjected to feature fusion through a gated recurrent unit to generate a spatiotemporal coupling feature vector, specifically: Obtain multi-scale temporal information features, build a meteorological station topology network based on the geographical location information of the meteorological station, and generate spatial correlation features through a dynamic graph attention network; The spatial correlation features are input into the feature interaction layer, and the fusion ratio of the spatiotemporal features is controlled by the cross gating unit to generate a spatiotemporal coupling feature vector.

5. The method for predicting power load based on a meteorological prediction model according to claim 4, characterized in that: The specific steps of obtaining multi-scale time information features are as follows: The coupling weight matrix dimensions and multidimensional meteorological data are aligned based on the time step to obtain each time step, and based on the dilated causal convolution and the preset convolution kernel size and dilation rate, the meteorological data of the historical time step is convolved to obtain the multi-scale time information features.

6. The method for predicting power load based on a meteorological prediction model according to claim 5, characterized in that: The spatial correlation features are generated by the dynamic graph attention network, specifically: Features are obtained through the time dimension feature extraction layer and used as the input of the graph attention mechanism; For each node and graph attention mechanism, calculate the attention weight of the node; Based on the attention weight, the features of each node are weighted and aggregated to obtain spatial correlation features.

7. The method for predicting power load based on a meteorological prediction model according to claim 6, characterized in that: The cross-gating unit is used to control the fusion ratio of the spatiotemporal features and generate a spatiotemporal coupling feature vector, specifically: The multi-scale temporal information features and spatial correlation features are fused to obtain fusion features; Perform outer product operation on the fused features to obtain the outer product matrix, and stack the outer product matrices of all time steps to obtain the outer product tensor; Perform tensor multiplication of the preset learning weight matrix and the outer product tensor to generate a spatiotemporal interactive gating signal; Extract features from the coupling weight matrix, perform vectorization operations on each feature to obtain a coupling vector, and stack the coupling vectors of all time steps to obtain a coupling tensor; The coupling tensor is processed by a multi-layer perceptron to generate a coupling modulation gating signal; The spatiotemporal interaction gating signal and the coupled modulation gating signal are weightedly concatenated with the original meteorological data and the coupling weight matrix respectively to obtain the spatiotemporal coupling feature vector.

8. The method for predicting power load based on a meteorological prediction model according to claim 7, characterized in that: The parallel convolution channel extracts the short-term fluctuation characteristics and long-term trend characteristics of the power load from the spatiotemporal coupling feature vector, specifically: Decompose and preprocess the input features, and compensate the time-frequency domain features to obtain short-term fluctuation flow and long-term trend flow; Based on the dual-path residual unit, the features of short-term fluctuation flow and long-term trend flow are spliced ​​and fused; The fused features are mapped to the power load forecast value, and the feature dimensions are compressed to obtain the final load forecast value.

9. The method for predicting power load based on a meteorological prediction model according to claim 8, characterized in that: The dynamic coupling weight matrix is ​​updated based on the prediction error back propagation, specifically: Calculate the mean square error between the load forecast value and the actual value; Based on the chain rule, the partial derivative of the mean square error with respect to each element of the coupling weight matrix is ​​calculated to obtain the gradient of the coupling weight matrix; Compute first-order and second-order moment estimates of the gradient; The first-order moment estimation and the second-order moment estimation are bias-corrected to obtain an updated coupling weight matrix.

10. A system for the method for power load forecasting based on a meteorological forecast model according to any one of claims 1 to 9, characterized in that: Includes the following modules: Data acquisition module: used to obtain historical power load data and corresponding multi-dimensional meteorological data, the multi-dimensional meteorological data including four meteorological parameters: temperature, humidity, wind speed and air pressure; Meteorological coupling weight calculation module: used to build a meteorological coupling effect analysis model and calculate the dynamic coupling weight matrix between meteorological parameters through the attention mechanism network; Spatiotemporal feature fusion module: used to fuse the coupling weight matrix and multidimensional meteorological data through a gated recurrent unit to generate a spatiotemporal coupling feature vector; Power load forecasting modeling module: used to establish a power load forecasting model, extracting short-term fluctuation characteristics and long-term trend characteristics of power load from spatiotemporal coupling feature vectors based on a deep residual network architecture and parallel convolution channels; Coupling weight adaptive update module: used to update the dynamic coupling weight matrix based on the back propagation of prediction errors to achieve adaptive adjustment of the coupling relationship of meteorological parameters.

Images