Power load prediction method and system based on VMD and MoE system

Abstract

The invention relates to the technical field of power load prediction, and discloses a power load prediction method and system based on a VMD and MoE system, and the method comprises the steps: inputting original power load data into a trained VMD decomposer, dynamically adjusting a modal number K value and a penalty factor # imgabs0 # value according to a power load fluctuation level, and decomposing the data into K modal components; thirdly, inputting the modal components into the trained MoE system, distributing M groups of expert models and weights for the modal components through a gating network, and adjusting the number of the expert models of the high-frequency and low-frequency modal components according to the fluctuation level; and finally, at least one group of expert models is selected from the M groups of expert models, prediction results are fused through weighted summation based on the selected model and the weight thereof, and a final power load prediction value is obtained. Through the method, the precision and robustness of power load prediction are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power load forecasting, and in particular to a power load forecasting method and system based on VMD and MoE systems. Background Art

[0002] Power load forecasting is the basis and key link of power system planning, dispatching and operation. Accurate load forecasting can provide important support for safe and stable operation of power system, optimized dispatching, demand-side management and power market transactions, and is of great significance to improving the reliability, economy and intelligence level of power system.

[0003] However, power load is affected by many complex factors, such as meteorological conditions, economic activities, social events, holidays, etc., showing significant nonlinear, non-stationary and volatile characteristics. Traditional load forecasting methods, such as statistical models (such as time series model ARIMA, regression model, etc.): It is difficult to effectively capture the nonlinear, non-stationary and volatile characteristics of load data, and the prediction accuracy is poor under complex working conditions. Traditional machine learning models (such as support vector machine SVM, artificial neural network ANN, etc.): Although they have certain nonlinear modeling capabilities, they are prone to overfitting and insufficient generalization when processing high-dimensional, high-noise, and volatile load data, and the model has poor interpretability. Signal decomposition methods (such as empirical mode decomposition EMD, integrated empirical mode decomposition EEMD, etc.) combined with prediction models: Although methods such as EMD / EEMD can decompose complex signals into multiple modal components, they have problems such as modal aliasing, endpoint effects, and lack of rigorous mathematical theoretical support. The decomposition effect is unstable, and the subsequent prediction model still finds it difficult to effectively use the decomposed information, and the prediction accuracy is limited.

[0004] Therefore, how to improve the accuracy and robustness of power load forecasting is an urgent problem to be solved in this field. Summary of the invention

[0005] The object of the present invention is to provide a power load forecasting method, system, medium and electronic device based on VMD and MoE system, so as to at least partially solve the above problems.

[0006] According to one aspect of the present disclosure, a power load forecasting method based on VMD and MoE system is proposed, including: step S1, collecting original power load data, wherein the original power load data at least includes historical power load data, meteorological data and calendar data.

[0007] Step S2, input the original power load data into the trained variational mode decomposer VMD, and decompose the original power load data into K modal components, where K>1.

[0008] Step S3, inputting the K modal components into a trained hybrid expert system MoE, wherein the hybrid expert system MoE includes a gating network and multiple expert models, wherein M groups of expert models and weights corresponding to each model are assigned to the modal components through the gating network, wherein each group of expert models includes at least K expert models, where M>0.

[0009] Step S4, selecting at least one group of expert models from the M groups of expert models, and fusing the prediction results of multiple expert models based on the selected expert models to obtain a final power load prediction result.

[0010] Optionally, the method also includes adjusting the K value and the M value based on the power load fluctuation level in a preset recent period, wherein the higher the fluctuation level, the larger the K value and the M value, and the lower the fluctuation level, the smaller the K value and the M value.

[0011] Optionally, the method also includes that each group of expert models includes at least K expert models, including dividing the K modal components into high-frequency modal components and low-frequency modal components, wherein the higher the level of power load fluctuation, the greater the number of models allocated to the high-frequency modal components.

[0012] Optionally, the method further includes: the higher the power load fluctuation level, the greater the penalty factor corresponding to the variational mode decomposer VMD The larger the value.

[0013] Optionally, the method also includes, based on the selected expert model, fusing the prediction results of multiple expert models to obtain the final power load prediction result, including, based on the selected expert model and its corresponding weights, fusing the prediction results of multiple expert models by weighted summation to obtain the final power load prediction result.

[0014] According to one aspect of the present disclosure, a power load forecasting system based on VMD and MoE system is proposed, including: a collection module for collecting original power load data, wherein the original power load data at least includes historical power load data, meteorological data and calendar data.

[0015] The decomposition module is used to input the original power load data into the trained variational mode decomposer VMD to decompose the original power load data into K modal components, where K>1.

[0016] The allocation module is used to input the K modal components into a trained hybrid expert system MoE, wherein the hybrid expert system MoE includes a gating network and multiple expert models, wherein M groups of expert models and weights corresponding to each model are allocated to the modal components through the gating network, wherein each group of expert models includes at least K expert models, where M>0.

[0017] The prediction module is used to select at least one group of expert models from the M groups of expert models, and based on the selected expert model, fuse the prediction results of multiple expert models to obtain the final power load prediction result.

[0018] Optionally, the system also includes an adjustment module for adjusting the K value and the M value based on the power load fluctuation level in a preset recent period, wherein the higher the fluctuation level, the larger the K value and the M value, and the lower the fluctuation level, the smaller the K value and the M value.

[0019] Optionally, the system further includes an allocation module, further used to divide the K modal components into high-frequency modal components and low-frequency modal components, wherein the higher the level of power load fluctuation is, the greater the number of models allocated to the high-frequency modal components.

[0020] The present disclosure further provides a computer-readable storage medium storing a computer program, including the steps of the method described in any of the above embodiments when the computer program is executed by a processor.

[0021] An embodiment of the present application further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor executes the steps in the method described in any of the above embodiments by calling the computer program stored in the memory.

[0022] The present invention discloses a method for power load forecasting based on variational mode decomposition (VMD) and hybrid expert system (MoE), including inputting the original power load data into a trained VMD decomposer, dynamically adjusting the modal quantity K value and the penalty factor α value according to the power load fluctuation level, and decomposing the data into K modal components. Then, the modal components are input into the trained MoE system, M groups of expert models and weights are assigned to the modal components through a gating network, and the number of expert models of high-frequency and low-frequency modal components is adjusted according to the fluctuation level. Finally, at least one group is selected from the M groups of expert models, and the prediction results are fused by weighted summation based on the selected models and their weights to obtain the final power load prediction value. This method improves the accuracy and robustness of power load forecasting. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A schematic diagram of a power load forecasting method based on VMD and MoE system provided in an embodiment of the present application.

[0024] Figure 2 A schematic diagram of a power load forecasting system based on VMD and MoE systems provided in an embodiment of the present application.

[0025] Figure 3A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] The specific embodiments of the present invention are further described below in conjunction with the accompanying drawings. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0027] It should be noted that, in the present application, "first", "second" and various numerical numbers indicate distinctions made for the convenience of description, and are not used to limit the scope of the embodiments of the present application. For example, different classification results are distinguished, rather than being used to describe a specific order or sequence. It should be understood that the objects described in this way can be interchanged where appropriate, so as to be able to describe solutions other than the embodiments of the present application.

[0028] Specifically, Figure 1 The specific implementation flow chart of a power load forecasting method based on VMD and MoE system provided by an embodiment of the present application is shown. Figure 1 The specific steps are as follows: Step S1, collecting original power load data, the original power load data at least includes historical power load data, meteorological data and calendar data.

[0029] Specifically, the power system data acquisition device is used to collect historical power load data in the target area. These data come from various monitoring nodes in the power grid and are recorded in hours to ensure the continuity and integrity of the data.

[0030] At the same time, meteorological data is obtained from the database of the local meteorological department, including information such as temperature, humidity, wind speed, weather conditions (sunny, cloudy, rainy, etc.). The timestamps of meteorological data and power load data are accurately matched to facilitate subsequent analysis of the relationship between the two.

[0031] Calendar data covers information such as weekdays, weekends, and holidays. Because different date types have significant differences in power load, they are sorted and labeled through a standard calendar library to provide a basis for analyzing the impact of different date types on power load.

[0032] Step S2, step S2, input the original power load data into the trained variational mode decomposer (VMD), and decompose the original power load data into K modal components, where K>1.

[0033] Specifically, the collected original power load data is input into the trained variational mode decomposer VMD. VMD decomposes the original power load data into K modal components by iteratively optimizing the variational problem. These modal components have different frequency characteristics, reflecting the trend terms, periodic terms and some fluctuation details in the power load from low frequency to high frequency. For example, low-frequency modal components may represent long-term power consumption trends, such as load changes caused by seasonal changes; high-frequency modal components may reflect rapid load fluctuations in the short term, such as load changes caused by sudden weather changes or special events.

[0034] VMD is a non-recursive, adaptive signal decomposition method, whose goal is to decompose the original signal into a set of intrinsic mode components (IMFs) with a specific bandwidth, and to ensure that the sum of the decomposed IMF components is equal to the original signal. VMD achieves signal decomposition by constructing and solving constrained variational problems, which can be expressed as: ;in, are the K IMF components obtained by decomposition, is the center frequency of each IMF component, is the Dirac function, represents convolution, represents the time partial derivative, represents the square of the L2 norm, is the penalty factor, is the original power load data, The above variational problem can be solved by an optimization algorithm such as the alternating direction method of multipliers (ADMM) to obtain the decomposed K IMF components, which will not be described in detail in this embodiment.

[0035] Optionally, before the original power load data is input into the trained variational mode decomposer VMD, the original power load data is also preprocessed, including data cleaning (removing outliers, filling missing values), normalization (converting data of different dimensions to the same dimension range for subsequent processing), etc. For example, missing meteorological data can be filled by interpolation method; abnormal mutation values ​​in power load data can be corrected by combining historical data and adjacent data points.

[0036] Step S3, inputting the K modal components into a trained hybrid expert system MoE, wherein the hybrid expert system MoE includes a gating network and multiple expert models, wherein M groups of expert models and weights corresponding to each model are assigned to the modal components through the gating network, wherein each group of expert models includes at least K expert models, where M>0.

[0037] Specifically, the hybrid expert system (MoE) consists of a gating network and multiple expert models. The MoE system contains multiple expert models, each of which is good at processing a specific type of data or predicting loads at a specific time scale. For example, for high-frequency modal components, models that are sensitive to high-frequency signals, such as support vector regression (SVR) and radial basis function neural network (RBFNN), are used to capture rapid fluctuations in load. For low-frequency modal components, models that are sensitive to low-frequency signals, such as long short-term memory networks (LSTM) and gated recurrent unit networks (GRU), are used to capture long-term trends and seasonal changes in loads.

[0038] When K modal components are input into the hybrid expert system MoE, the restricted gating network starts working. In this embodiment, the gating network dynamically assigns M groups of expert models to the modal components according to the frequency characteristics, energy distribution and other information of the modal components, and determines the weight corresponding to each model. Each group of expert models contains at least K expert models. Different expert models can adopt different types of prediction models to adapt to the characteristics of different modal components. In addition, expert models of different time scales can also be set for different prediction time scales, for example, short-term prediction expert models (predicting the next 1-24 hours), medium-term prediction expert models (predicting the next 1-7 days), etc., which are not limited in this embodiment.

[0039] Optionally, the gating network adopts a 3-layer fully connected neural network (MLP). The input layer inputs a feature vector, such as an IMF component feature, multiple hidden layers, the number of neurons can be set according to actual needs, the activation function can be ReLU, the output layer, the number of neurons is the same as the K value, the activation function can be Softmax, and the recommended expert model and its corresponding weight are output.

[0040] Step S4, selecting at least one group of expert models from the M groups of expert models, and fusing the prediction results of multiple expert models based on the selected expert models to obtain a final power load prediction result.

[0041] Specifically, at least one group of expert models with better performance is selected from the M groups of expert models. The selection basis can be indicators such as the prediction accuracy and mean square error of the model on historical data. For example, by comparing the prediction results of each group of expert models on the validation set data, the group with the smallest mean square error is selected.

[0042] Optionally, the prediction results of multiple expert models are integrated based on the selected expert model to obtain the final power load prediction result, including integrating the prediction results of multiple expert models by weighted summation based on the selected expert model and its corresponding weights to obtain the final power load prediction result. Specifically, based on the selected expert model and its corresponding weight, the prediction results of multiple expert models are integrated by weighted summation. Assuming that a group of selected expert models are E1, E2, ..., E6, and the corresponding weights are w1, w2, ..., w6, and the prediction values ​​of each expert model for the power load are p1, p2, ..., p6, then the final power load prediction result P is: In this way, the advantages of multiple expert models are combined to obtain a more accurate final power load forecast result.

[0043] Optionally, based on the power load fluctuation level in a preset recent period, adjust the important parameters K and Value, penalty factor Used to balance data fidelity and sparsity of modal components, The larger the value, the narrower the bandwidth of the decomposed modal component and the better the sparsity. Specifically, the recent period is preset as the past week, and the fluctuation of the power load during this period is calculated. The power load fluctuation level is determined by calculating statistics such as the standard deviation and range of the load data, combined with the fluctuation range of the historical data. For example, the fluctuation level is divided into three levels: low, medium, and high. When the standard deviation is less than a certain threshold and the range is in a small range, it is judged as a low fluctuation level; when the standard deviation and range are in a medium range, it is a medium fluctuation level; when the standard deviation is greater than a larger threshold and the range is large, it is a high fluctuation level.

[0044] According to the power load fluctuation level in the past period of time, determine the modal number K value and Among them, the higher the power load fluctuation level, the larger the penalty factor corresponding to the variational mode decomposer VMD. The larger the value, the lower the volatility. Set to a relatively small value. For example, when the recent power load fluctuation level is high, choose a larger K value (such as K=5) to more finely decompose the modal components of different frequencies and improve the robustness of the algorithm under high power load fluctuation conditions; when the fluctuation level is low, choose a smaller K value (such as K=3). At the same time, adjust the penalty factor of VMD according to the fluctuation level The higher the volatility level, The larger the value, for example, when the volatility level is high, Can be set to a relatively large value, such as 3000. The value makes VMD emphasize the capture of signal details during decomposition, and can better decompose different modal components under complex fluctuations, providing richer and more accurate information for subsequent predictions. It can be set to a relatively small value, such as 1000. In this case, VMD pays more attention to smoothness during decomposition to extract relatively stable modal components.

[0045] Optionally, the M value is adjusted based on the power load fluctuation level in a preset recent period. Specifically, the M value is adjusted according to the power load fluctuation level. When the fluctuation level is low, the M value is reduced, for example, M is adjusted to 2. Because in the case of low fluctuation, the power load data is relatively stable, and there is no need for too many modal components and expert model groups to analyze and predict in order to improve the efficiency of the system. When the fluctuation level is high, the M value is increased, such as M is increased to 4. This is because in the case of high fluctuation, the changes in power load data are complex, and more modal components are needed to decompose the data, and more groups of expert models are needed to analyze and predict the data from different angles.

[0046] Optionally, the K modal components are divided into high-frequency modal components and low-frequency modal components, wherein the higher the power load fluctuation level, the more models are allocated to the high-frequency modal components. Specifically, the K modal components are divided into high-frequency modal components and low-frequency modal components. For low fluctuation levels, since the data is relatively stable and the high-frequency modal components are relatively few, the number of models allocated to the high-frequency modal components is relatively small, such as allocating 1 model to each high-frequency modal component in each group of expert models, and more expert models can be allocated to the low-frequency modal components to enhance the ability to predict long-term trends. When the fluctuation level increases, such as under high fluctuation levels, the power load changes dramatically, the high-frequency modal components increase, and the number of models allocated to the high-frequency modal components increases accordingly. Each group of expert models may allocate 2-3 models to each high-frequency modal component to enhance the modeling ability of short-term fluctuations, so as to better capture the load change information of the high-frequency part.

[0047] Through this embodiment, the original power load data is decomposed into modal components with different frequency characteristics through VMD, which can more finely display the trend items, periodic items and fluctuation details in the power load, and provide rich and accurate information for subsequent predictions. The gated network in MoE allocates multiple groups of expert models to different modal components. Each expert model adopts a different prediction algorithm, and then the prediction results are fused through weighted summation to give full play to the advantages of each model; at the same time, the K value, M value and penalty factor are adjusted according to the power load fluctuation level in the preset recent period. When the load fluctuation is low, the number of modal components and expert model groups is reduced to reduce the computational complexity; when the fluctuation is high, the corresponding number and value to adapt to complex and changing data; in addition, different numbers of expert models are assigned to high-frequency and low-frequency modal components according to the load fluctuation level. When the fluctuation level is high, more models are assigned to the high-frequency modal components to better capture the rapid load fluctuation information, and vice versa when the fluctuation level is low; through the above scheme, the accuracy and robustness of power load forecasting are improved.

[0048] Corresponding to the above embodiment, a power load forecasting method based on VMD and MoE system is provided. Figure 2 A structural block diagram of a power load forecasting system based on VMD and MoE system provided in an embodiment of the present application is shown. For ease of explanation, only the parts related to the embodiment of the present application are shown.

[0049] See also Figure 2 As shown, an electric power load forecasting system 200 based on VMD and MoE system provided in an embodiment of the present application includes: a collection module for collecting original electric power load data, and the original electric power load data at least includes historical electric power load data, meteorological data and calendar data.

[0050] The decomposition module is used to input the original power load data into the trained variational mode decomposer VMD to decompose the original power load data into K modal components, where K>1.

[0051] The allocation module is used to input the K modal components into a trained hybrid expert system MoE, wherein the hybrid expert system MoE includes a gating network and multiple expert models, wherein M groups of expert models and weights corresponding to each model are allocated to the modal components through the gating network, wherein each group of expert models includes at least K expert models, where M>0.

[0052] The prediction module is used to select at least one group of expert models from the M groups of expert models, and based on the selected expert model, fuse the prediction results of multiple expert models to obtain the final power load prediction result.

[0053] Optionally, the system also includes an adjustment module for adjusting the K value and the M value based on the power load fluctuation level in a preset recent period, wherein the higher the fluctuation level, the larger the K value and the M value, and the lower the fluctuation level, the smaller the K value and the M value.

[0054] Optionally, the system further includes an allocation module, further used to divide the K modal components into high-frequency modal components and low-frequency modal components, wherein the higher the level of power load fluctuation is, the greater the number of models allocated to the high-frequency modal components.

[0055] Accordingly, an embodiment of the present application further provides an electronic device, which may be a mobile terminal or a server. Figure 3 As shown, Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. For ease of explanation, only the parts related to the embodiment of the present application are shown.

[0056] The electronic device 300 includes a processor 301 having one or more processing cores, a memory 302 having one or more computer-readable storage media, and a computer program stored in the memory 302 and executable on the processor. The processor 301 is electrically connected to the memory 302. Those skilled in the art will appreciate that the electronic device structure shown in the figure does not constitute a limitation on the electronic device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0057] The processor 301 is the control center of the electronic device 300. It uses various interfaces and lines to connect various parts of the entire electronic device 300, and executes various functions of the electronic device 300 and processes data by running or loading software programs (computer programs) and / or units stored in the memory 302, and calling data stored in the memory 302, thereby monitoring the electronic device 300 as a whole.

[0058] In an embodiment of the present application, the processor 301 in the electronic device 300 will load the instructions corresponding to the processes of one or more applications into the memory 302 according to the following steps, and the processor 301 will run the applications stored in the memory 302 to realize various functions: Step S1, collect original power load data, and the original power load data at least includes historical power load data, meteorological data and calendar data.

[0059] Step S2, input the original power load data into the trained variational mode decomposer VMD, and decompose the original power load data into K modal components, where K>1.

[0060] Step S3, inputting the K modal components into a trained hybrid expert system MoE, wherein the hybrid expert system MoE includes a gating network and multiple expert models, wherein M groups of expert models and weights corresponding to each model are assigned to the modal components through the gating network, wherein each group of expert models includes at least K expert models, where M>0.

[0061] Step S4, selecting at least one group of expert models from the M groups of expert models, and fusing the prediction results of multiple expert models based on the selected expert models to obtain a final power load prediction result.

[0062] The specific implementation of the above operations can be found in the aforementioned embodiments, which will not be described in detail here.

[0063] Optional, such as Figure 3 As shown, the electronic device 300 further includes: a prediction unit 303, a communication unit 304, an input unit 305 and a power supply 306. The processor 301 is electrically connected to the prediction unit 303, the communication unit 304, the input unit 305 and the power supply 306 respectively. Those skilled in the art can understand that Figure 3 The electronic device structure shown in the figure does not constitute a limitation of the electronic device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0064] The prediction unit 303 may be used for power load prediction based on VMD and MoE systems.

[0065] The communication unit 304 may be used to communicate with other devices.

[0066] The input unit 305 may be used to receive input numbers, character information or user feature information (such as fingerprint, iris, facial information, etc.), and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0067] The power supply 306 is used to supply power to various components of the electronic device 300. Optionally, the power supply 306 can be logically connected to the processor 301 through a power management system, so that the power management system can manage charging, discharging, power consumption, and other functions. The power supply 306 can also include one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0068] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0069] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0070] To this end, an embodiment of the present application provides a computer-readable storage medium, in which multiple computer programs are stored. The computer programs can be loaded by a processor to execute the steps of a power load forecasting method based on VMD and MoE system provided in an embodiment of the present application.

[0071] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0072] The computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0073] Since the computer program stored in the storage medium can execute the steps of any one of the power load forecasting methods based on VMD and MoE systems provided in the embodiments of the present application, the beneficial effects of any one of the power load forecasting methods based on VMD and MoE systems provided in the embodiments of the present application can be achieved. For details, please see the previous embodiments and will not be repeated here.

[0074] The present application is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present application. These computer program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0075] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0076] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0077] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, various changes, modifications, substitutions and variations of these embodiments are made without departing from the principles and spirit of the present invention, and still fall within the scope of protection of the present invention.

Claims

1. A power load forecasting method based on VMD and MoE system, characterized in that: include: Step S1, collecting original power load data, wherein the original power load data at least includes historical power load data, meteorological data and calendar data; Step S2, input the original power load data into the trained variational mode decomposer VMD, and decompose the original power load data into K modal components, where K>1; Step S3, input the K modal components into the trained hybrid expert system MoE, the hybrid expert system MoE includes a gating network and multiple expert models, wherein M groups of expert models and weights corresponding to each model are assigned to the modal components through the gating network, wherein each group of expert models includes at least K expert models, where M>0; Step S4, select at least one group of expert models from the M groups of expert models, and based on the selected expert model, fuse the prediction results of multiple expert models to obtain the final power load prediction result.

2. The power load forecasting method based on VMD and MoE system according to claim 1 is characterized in that: Based on the power load fluctuation level in a preset recent period, the K value and the M value are adjusted, wherein the higher the fluctuation level, the larger the K value and the M value, and the lower the fluctuation level, the smaller the K value and the M value.

3. The power load forecasting method based on VMD and MoE system according to claim 2 is characterized in that: Each group of expert models includes at least K expert models, including: dividing the K modal components into high-frequency modal components and low-frequency modal components, wherein the higher the power load fluctuation level, the more models are allocated to the high-frequency modal components.

4. The power load forecasting method based on VMD and MoE system according to claim 2 is characterized in that: The higher the power load fluctuation level, the larger the penalty factor corresponding to the variational mode decomposer VMD. The larger the value.

5. The power load forecasting method based on VMD and MoE system according to claim 3 is characterized in that: The method of fusing the prediction results of multiple expert models based on the selected expert model to obtain the final power load prediction result includes fusing the prediction results of multiple expert models by weighted summation based on the selected expert model and its corresponding weights to obtain the final power load prediction result.

6. A power load forecasting system based on VMD and MoE system, characterized in that: include: A collection module, used for collecting original power load data, wherein the original power load data at least includes historical power load data, meteorological data and calendar data; A decomposition module is used to input the original power load data into a trained variational mode decomposer VMD, and decompose the original power load data into K modal components, where K>1; an allocation module is used to input the K modal components into a trained hybrid expert system MoE, and the hybrid expert system MoE includes a gating network and multiple expert models, wherein M groups of expert models and weights corresponding to each model are allocated to the modal components through the gating network, wherein each group of expert models includes at least K expert models, where M>0; a prediction module is used to select at least one group of expert models from the M groups of expert models, and based on the selected expert model, fuse the prediction results of multiple expert models to obtain a final power load prediction result.

7. The power load forecasting system based on VMD and MoE system according to claim 6, characterized in that: The system also includes: an adjustment module for adjusting the K value and the M value based on the power load fluctuation level in a preset recent period, wherein the higher the fluctuation level, the larger the K value and the M value, and the lower the fluctuation level, the smaller the K value and the M value.

8. The power load forecasting system based on VMD and MoE system according to claim 7, characterized in that: The system further includes: an allocation module, further used to divide the K modal components into high-frequency modal components and low-frequency modal components, wherein the higher the level of power load fluctuation, the greater the number of models allocated to the high-frequency modal components.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is performed.

10. An electronic device, characterized in that: It comprises a memory storing executable program code and a processor coupled to the memory; wherein the processor calls the executable program code stored in the memory to execute the method according to any one of claims 1-5.

Images