Power load prediction method and system based on intelligent algorithm
Through intelligent algorithms, the power historical data is processed, and the weight adjustment features are extracted and weight adjustment features are combined with adaptive comprehensive training models, the problem of degradation of accuracy and adaptability of traditional power load prediction methods is solved, achieving higher prediction accuracy and stability.
Patent Information
- Application Number
- CN202411673141.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-21
- Publication Date
- 2025-04-11
AI Technical Summary
In the face of complex load modes, multi-dimensional data and new energy power generation, the prediction accuracy and adaptability are gradually declining, which cannot meet the needs of modern power systems.
The power load prediction method based on intelligent algorithm is adopted, and the power historical data is obtained, the dimensionality reduction process is performed using principal component analysis method, the target features are extracted, and the weight coefficient of each feature is determined using the entropy weight method, the target training parameters are generated, and the power load prediction data is output.
It significantly improves the overall accuracy and stability of power load prediction, ensures that the model shows stronger stability when processing new data, and provides a more accurate reference for power grid scheduling.
Smart Images
Figure CN119813151A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of smart grid technology, and in particular to a method and system for predicting electric power load based on an intelligent algorithm. Background Art
[0002] Power load forecasting is one of the core tasks of power system operation and dispatching, and is directly related to the safety, economy and stability of the power system. Accurate load forecasting can help power companies reasonably arrange power generation plans, optimize power market operations, reduce the cost of power supply, and to a certain extent ensure the stable operation of the power grid. With the rapid development of the power system, especially with the widespread application of new energy power generation and smart grid technology, traditional power load forecasting methods have gradually exposed many shortcomings. Although traditional methods such as time series analysis, linear regression and expert systems can provide reliable forecasting results to a certain extent, with the complexity of power load patterns, the increase of multi-dimensional data and the introduction of new energy power generation, the prediction accuracy and adaptability of traditional methods have gradually decreased, and they cannot meet the needs of modern power systems. Summary of the invention
[0003] The purpose of the present invention is to solve the problem that the accuracy and adaptability of current power load forecasting methods are gradually declining and cannot meet the needs of modern power systems. The present invention provides a power load forecasting method and system based on an intelligent algorithm.
[0004] The technical solution of the embodiment of the present application is implemented as follows: A first aspect of an embodiment of the present application provides a method for predicting power load based on an intelligent algorithm, comprising: Obtain historical electricity data; The historical power data is subjected to dimensionality reduction processing by using a principal component analysis method to extract target features; the target features include at least one of historical load values, air temperature and humidity; Determine a first weight coefficient corresponding to each of the target features by using an entropy weight method, and generate a corresponding target training parameter based on the target feature and the first weight coefficient corresponding to the target feature; The target training parameters are input into an adaptive comprehensive training model, and power load forecast data is output; the adaptive comprehensive training model includes at least one of a deep neural network submodule, a long short-term memory network submodule, a random forest submodule, and a gradient boosting decision tree submodule.
[0005] Optionally, the principal component analysis method is used to perform dimensionality reduction processing on the power history data to extract target features, including: Standardizing the power history data to obtain standard data; The principal component analysis method is used to reduce the dimension of the standard data and extract the target features.
[0006] Optionally, the adopting of an entropy weight method to determine a first weight coefficient corresponding to each of the target features, and generating a corresponding target training parameter based on the target feature and the first weight coefficient corresponding to the target feature, includes: Obtaining information entropy of each target feature; Determine a first weight coefficient corresponding to the target feature based on the information entropy; the first weight coefficient represents the degree of influence of the target feature on the power load forecast data; The corresponding target feature is modified using the first weight coefficient to generate corresponding target training parameters.
[0007] Optionally, the inputting the target training parameters into the adaptive comprehensive training model and outputting the power load forecast data includes: Determine the second weight coefficients corresponding to the submodules based on the correlation between the target training parameters and the submodules in the adaptive comprehensive training model; Inputting the target training parameters into the adaptive comprehensive training model, and outputting preliminary power load forecast data for each submodule; The preliminary power load forecast data and the corresponding second weight coefficient are combined to obtain power load forecast data.
[0008] The second aspect of the embodiment of the present application provides a method system for predicting power load based on intelligent algorithm, including: a data acquisition module, a feature extraction module, a parameter generation module and a data output module; wherein, A data acquisition module configured to acquire power history data; A feature extraction module is configured to perform dimensionality reduction processing on the power history data using a principal component analysis method to extract target features; the target features include at least one of historical load values, air temperature and humidity; A parameter generation module is configured to determine a first weight coefficient corresponding to each of the target features by using an entropy weight method, and generate a corresponding target training parameter based on the target feature and the first weight coefficient corresponding to the target feature; A data output module is configured to input the target training parameters into an adaptive comprehensive training model and output power load forecast data; the adaptive comprehensive training model includes at least one of a deep neural network submodule, a long short-term memory network submodule, a random forest submodule and a gradient boosting decision tree submodule.
[0009] Optionally, the feature extraction module is specifically configured as follows: Standardizing the power history data to obtain standard data; The principal component analysis method is used to reduce the dimension of the standard data and extract the target features.
[0010] Optionally, the parameter generation module is specifically configured as follows: Obtaining information entropy of each target feature; Determine a first weight coefficient corresponding to the target feature based on the information entropy; the first weight coefficient represents the degree of influence of the target feature on the power load forecast data; The corresponding target feature is modified using the first weight coefficient to generate corresponding target training parameters.
[0011] Optionally, the data output module is specifically configured as follows: Determine the second weight coefficients corresponding to the submodules based on the correlation between the target training parameters and the submodules in the adaptive comprehensive training model; Inputting the target training parameters into the adaptive comprehensive training model, and outputting preliminary power load forecast data for each submodule; The preliminary power load forecast data and the corresponding second weight coefficient are combined to obtain power load forecast data.
[0012] A third aspect of an embodiment of the present application provides an electronic device, comprising a processor and a memory; the memory stores a computer program, wherein the computer program, when executed by the processor, implements the power load forecasting method based on the intelligent algorithm described in the first aspect.
[0013] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect.
[0014] Compared with the prior art, the technical solution provided by this application has the following beneficial effects: The present invention provides a method and system for predicting electric load based on intelligent algorithm, which obtains electric power historical data, uses principal component analysis to reduce the dimension of electric power historical data, extracts target features, and uses entropy weight method to determine the first weight coefficient corresponding to each target feature, generates corresponding target training parameters based on the target features and the first weight coefficient corresponding to the target features, inputs the target training parameters into an adaptive comprehensive training model, and outputs electric load prediction data. The present invention obtains training parameters by using principal component analysis and entropy weight method, so that the training will focus more on features with rich information, thereby improving the generalization performance of the model, ensuring that the model shows stronger stability when processing new data, significantly improving the overall accuracy and stability of electric load prediction, and providing a more accurate reference basis for power grid dispatching. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 A schematic diagram of a flow chart of a power load forecasting method based on an intelligent algorithm provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of a power load forecasting system based on an intelligent algorithm provided in an embodiment of the present application; Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0016] Below, embodiments of the present application will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present application. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present application. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present application. The terms used herein are only for describing specific embodiments and are not intended to limit the present application. The terms "include", "comprising", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components. All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification, and should not be interpreted in an idealized or overly rigid manner.
[0017] Some block diagrams and / or flow charts are shown in the accompanying drawings. It should be understood that some blocks or combinations thereof in the block diagrams and / or flow charts may be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that these instructions, when executed by the processor, may create a device for implementing the functions / operations described in these block diagrams and / or flow charts.
[0018] In some embodiments, see Figure 1 , Figure 1 A schematic diagram of a flow chart of a power load forecasting method based on an intelligent algorithm provided in an embodiment of the present application; the power load forecasting method based on an intelligent algorithm provided in an embodiment of the present application includes: S110, obtaining historical power data.
[0019] Here, to ensure the reliability of historical power data, it can be obtained from multiple sources, including data from different dates, different time periods, and different measuring devices. Historical power data can contain multiple parameters, such as power load, temperature, humidity, etc.
[0020] S120, using principal component analysis to perform dimensionality reduction processing on the historical power data to extract target features; the target features include at least one of historical load values, temperature and humidity.
[0021] Principal component analysis (PCA) can retain the main features of the data while removing redundancy and noise by projecting the original data onto new coordinate axes. These new coordinate axes (principal components) are the directions with the largest variance of the original data. For example, the numerical ranges and dimensions of these features, such as power load (unit: MW), temperature (unit: ℃), humidity (unit: %), etc., are different. If they are directly input into the model, the difference in the numerical ranges of different features will cause the model to pay more attention to features with larger values during training and ignore features with smaller values, thus affecting the accuracy of the prediction. Through data standardization, data of different features are converted to the same scale (usually between 0 and 1 or between -1 and 1), eliminating dimensional differences and making each feature equally important during the model training process. This helps the model learn the relationship between each feature more fairly, thereby improving the generalization ability and prediction accuracy of the model.
[0022] In some embodiments, S120, using principal component analysis to reduce the dimension of the power history data and extract target features includes: Standardize the historical power data to obtain standard data; The principal component analysis method is used to reduce the dimension of the standard data and extract the target features.
[0023] In this embodiment, with the help of feature extraction, the features that have the greatest impact on power load prediction can be identified and screened, thereby achieving data dimensionality reduction, eliminating redundant information, streamlining training content, and enabling the model to achieve faster and more efficient training and prediction.
[0024] In an example, the historical power data includes three characteristics: historical load value (unit: MW), temperature (unit: °C), and humidity.
[0025] Sample data: 1. Historical load value: ; 2. Temperature: ; 3. Humidity: ; Calculate the mean and standard deviation: ; ; ; ; ; ; Standardized eigenvalues: For historical load values: ; ; The standardized historical load value is: ; Temperature ; ; The standardized temperature is: ; Humidity: ; ; The standardized humidity is: .
[0026] S130, using an entropy weight method to determine a first weight coefficient corresponding to each target feature, and generating a corresponding target training parameter based on the target feature and the first weight coefficient corresponding to the target feature.
[0027] The entropy weight method is a method to determine the weight according to the amount of information (i.e., entropy value) provided by each indicator data. The entropy weight method uses the information entropy of each feature to evaluate its relative importance and assign a reasonable weight to it. Specifically, the lower the information entropy, the richer the information contained, so it should be given a higher weight in the prediction. With this mechanism, the model will focus more on the features with rich information during training, thereby improving the generalization performance of the model and ensuring that the model shows stronger stability when processing new data.
[0028] In some embodiments, S130, using an entropy weight method to determine a first weight coefficient corresponding to each target feature, and generating a corresponding target training parameter based on the target feature and the first weight coefficient corresponding to the target feature, including: Get the information entropy of each target feature; Determine a first weight coefficient corresponding to the target feature based on the information entropy; the first weight coefficient represents the degree of influence of the target feature on the power load forecast data; The first weight coefficient is used to modify the corresponding target feature to generate the corresponding target training parameter.
[0029] In this embodiment, for each feature, its information entropy is calculated. Information entropy reflects the degree of confusion of feature data. The larger the information entropy, the greater the amount of information of the feature and the greater the uncertainty. Based on the information entropy, the feature redundancy (difference coefficient) can be calculated. The feature redundancy is the complement of the information entropy and reflects the degree of difference of the feature data. The first weight coefficient can be calculated by the ratio of the feature redundancy to the sum of all feature redundancies.
[0030] In one example, using the above standardized data Compute the probability distribution of features.
[0031] Calculate historical load values, ; Temperature: ; humidity: ; Continue to calculate the entropy value: First, the historical load value The entropy value : ; Second, the temperature The entropy value :
[0032] The third is humidity The entropy value ; .
[0033] Correspondingly, the historical load value Temperature and humidity The corresponding first weight coefficients are:
[0034]
[0035] .
[0036] S140, input the target training parameters into the adaptive comprehensive training model, and output the power load forecast data; the adaptive comprehensive training model includes at least one of a deep neural network submodule, a long short-term memory network submodule, a random forest submodule and a gradient boosting decision tree submodule.
[0037] In this embodiment, the adaptive weight adjustment mechanism is used to dynamically adjust the weights of each submodule in the adaptive comprehensive training model according to the prediction performance of each submodule in different scenarios (such as historical error, model accuracy), so that the final comprehensive prediction result can give full play to the advantages of each model. This adaptive adjustment method ensures that the overall performance of the model combination can be optimized regardless of how the data environment changes. When a submodule performs poorly in a specific time period or under specific conditions, the adaptive weight adjustment mechanism will reduce the weight of the model and reduce its impact on the final prediction result; conversely, when a submodule performs well, its weight will be increased, thereby enhancing its contribution to the final prediction result. By dynamically adjusting the weights of each model and integrating the advantages of multiple models, a high-precision and stable prediction result is finally generated, which not only improves the accuracy and robustness of the prediction, but also enhances the system's ability to cope with uncertainty and data changes, ensuring the reliability and practicality of the prediction results.
[0038] In some embodiments, S140, inputting the target training parameters into the adaptive comprehensive training model, and outputting the power load forecast data, includes: Determine the second weight coefficients corresponding to the submodules based on the correlation between the target training parameters and the submodules in the adaptive comprehensive training model; Input the target training parameters into the adaptive comprehensive training model, and output the preliminary power load forecast data of each submodule; The preliminary power load forecast data and the corresponding second weight coefficient are combined to obtain the power load forecast data.
[0039] In an example, suppose that the deep neural network submodule, random forest submodule and LSTM submodule are used for power load forecasting, and the historical errors of each submodule are The second weight coefficients corresponding to each submodule are: ; ; ; Assume that the prediction value of each model at the current moment is , , , the final power load forecast data is: .
[0040] Through the collaborative work of various sub-modules, the complex nonlinear relationships and time series dependencies in the data can be effectively captured, thereby significantly improving the accuracy of the prediction results.
[0041] The embodiment of the present invention obtains historical power data, uses principal component analysis to perform dimensionality reduction processing on the historical power data, extracts target features, and uses entropy weight method to determine the first weight coefficient corresponding to each target feature, generates corresponding target training parameters based on the target features and the first weight coefficient corresponding to the target features, inputs the target training parameters into the adaptive comprehensive training model, and outputs power load forecasting data. The present invention obtains training parameters by using principal component analysis and entropy weight method, so that during training, it will focus more on features with rich information, thereby improving the generalization performance of the model, ensuring that the model shows stronger stability when processing new data, significantly improving the overall accuracy and stability of power load forecasting, and providing a more accurate reference basis for power grid dispatching.
[0042] In some embodiments, see Figure 2 , Figure 2 A schematic diagram of the structure of a power load forecasting system based on an intelligent algorithm provided in an embodiment of the present application; an embodiment of the present application provides a power load forecasting system 200 based on an intelligent algorithm, comprising: a data acquisition module 210, a feature extraction module 220, a parameter generation module 230 and a data output module 240; wherein, A data acquisition module 210 is configured to acquire power history data; The feature extraction module 220 is configured to perform dimensionality reduction processing on the power history data using a principal component analysis method to extract target features; the target features include at least one of historical load values, temperature and humidity; The parameter generation module 230 is configured to determine a first weight coefficient corresponding to each target feature by using an entropy weight method, and generate a corresponding target training parameter based on the target feature and the first weight coefficient corresponding to the target feature; The data output module 240 is configured to input the target training parameters into the adaptive comprehensive training model and output the power load prediction data; the adaptive comprehensive training model includes at least one of a deep neural network submodule, a long short-term memory network submodule, a random forest submodule and a gradient boosting decision tree submodule.
[0043] In some embodiments, the feature extraction module 220 is specifically configured as follows: Standardize the historical power data to obtain standard data; The principal component analysis method is used to reduce the dimension of the standard data and extract the target features.
[0044] In some embodiments, the parameter generation module 230 is specifically configured as follows: Get the information entropy of each target feature; Determine a first weight coefficient corresponding to the target feature based on the information entropy; the first weight coefficient represents the degree of influence of the target feature on the power load forecast data; The first weight coefficient is used to modify the corresponding target feature to generate the corresponding target training parameter.
[0045] In some embodiments, the data output module 240 is specifically configured as follows: Determine the second weight coefficients corresponding to the submodules based on the correlation between the target training parameters and the submodules in the adaptive comprehensive training model; Input the target training parameters into the adaptive comprehensive training model, and output the preliminary power load forecast data of each submodule; The preliminary power load forecast data and the corresponding second weight coefficient are combined to obtain the power load forecast data.
[0046] The power load prediction system based on intelligent algorithm provided in the embodiment of the present application can implement each process in the embodiment corresponding to the above-mentioned power load prediction method based on intelligent algorithm. To avoid repetition, it will not be repeated here.
[0047] It should be noted that the power load forecasting system based on intelligent algorithm provided in the embodiment of the present application and the power load forecasting method based on intelligent algorithm provided in the embodiment of the present application are based on the same application concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned power load forecasting method based on intelligent algorithm, and the repeated parts will not be repeated.
[0048] In some embodiments, see Figure 3 , Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. An electronic device 300 provided in an embodiment of the present application includes a processor 310 and a memory 320; the memory 320 stores a computer program, wherein the computer program implements the above-mentioned power load prediction method based on the intelligent algorithm when executed by the processor. Specifically, the processor 310 may include, for example, a general-purpose microprocessor, an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 310 may also include an onboard memory for cache purposes. The processor 310 may be a single processing unit or multiple processing units for executing different actions of the method flow according to the embodiment of the present application.
[0049] The memory 320 may be any medium capable of containing, storing, conveying, propagating or transmitting instructions. For example, the memory 320 may include, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared or semiconductor system, device, component or propagation medium. Specific examples of the memory 320 include: a magnetic storage device, such as a magnetic tape or a hard disk (HDD); an optical storage device, such as a compact disk (CD-ROM); a random access memory (RAM) or flash memory; and / or a wired / wireless communication link.
[0050] The present application also provides a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, the above-mentioned power load forecasting method based on the intelligent algorithm is implemented. The computer-readable medium may be included in the device / apparatus / system described in the above-mentioned embodiment; or it may exist independently without being assembled into the device / apparatus / system. The above-mentioned computer-readable medium carries one or more programs, and when the above-mentioned one or more programs are executed, the method according to the embodiment of the present application is implemented.
[0051] According to an embodiment of the present application, a computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a computer-readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, optical cable, radio frequency signal, etc., or any suitable combination of the above.
[0052] Those skilled in the art will appreciate that the features described in the various embodiments and / or claims of the present application may be combined and / or combined in a variety of ways, even if such combinations or combinations are not explicitly described in the present application. In particular, without departing from the spirit and teachings of the present application, the features described in the various embodiments and / or claims of the present application may be combined and / or combined in a variety of ways. All of these combinations and / or combinations fall within the scope of the present application. Therefore, the scope of the present application should not be limited to the above-described embodiments, but should be determined not only by the appended claims, but also by the equivalents of the appended claims.
Claims
1. A method for predicting power load based on intelligent algorithm, characterized in that: include: Obtain historical electricity data; Using principal component analysis to reduce the dimension of the historical power data and extract target features; The target characteristic includes at least one of a historical load value, air temperature, and humidity; Determine a first weight coefficient corresponding to each of the target features by using an entropy weight method, and generate a corresponding target training parameter based on the target feature and the first weight coefficient corresponding to the target feature; The target training parameters are input into an adaptive comprehensive training model, and power load forecast data is output; the adaptive comprehensive training model includes at least one of a deep neural network submodule, a long short-term memory network submodule, a random forest submodule, and a gradient boosting decision tree submodule.
2. The power load forecasting method based on intelligent algorithm according to claim 1 is characterized in that: The principal component analysis method is used to perform dimensionality reduction processing on the power history data to extract target features, including: Standardizing the power history data to obtain standard data; The principal component analysis method is used to reduce the dimension of the standard data and extract the target features.
3. The power load forecasting method based on intelligent algorithm according to claim 1 is characterized in that: The step of using the entropy weight method to determine the first weight coefficient corresponding to each of the target features, and generating the corresponding target training parameters based on the target features and the first weight coefficient corresponding to the target features, includes: Obtaining information entropy of each target feature; Determine a first weight coefficient corresponding to the target feature based on the information entropy; the first weight coefficient represents the degree of influence of the target feature on the power load forecast data; The corresponding target feature is modified using the first weight coefficient to generate corresponding target training parameters.
4. The power load forecasting method based on intelligent algorithm according to claim 1 is characterized in that: The target training parameters are input into the adaptive comprehensive training model, and the output is the power load forecast data, including: Determine the second weight coefficients corresponding to the submodules based on the correlation between the target training parameters and the submodules in the adaptive comprehensive training model; Inputting the target training parameters into the adaptive comprehensive training model, and outputting the preliminary power load forecast data of each submodule; The preliminary power load forecast data and the corresponding second weight coefficient are combined to obtain power load forecast data.
5. A power load forecasting system based on intelligent algorithm, characterized in that: include: Data acquisition module, feature extraction module, parameter generation module and data output module; wherein, A data acquisition module configured to acquire power history data; A feature extraction module is configured to perform dimensionality reduction processing on the power history data using a principal component analysis method to extract target features; the target features include at least one of historical load values, air temperature and humidity; A parameter generation module is configured to determine a first weight coefficient corresponding to each of the target features by using an entropy weight method, and generate a corresponding target training parameter based on the target feature and the first weight coefficient corresponding to the target feature; A data output module is configured to input the target training parameters into an adaptive comprehensive training model and output power load forecast data; the adaptive comprehensive training model includes at least one of a deep neural network submodule, a long short-term memory network submodule, a random forest submodule and a gradient boosting decision tree submodule.
6. The power load forecasting system based on intelligent algorithm according to claim 5 is characterized in that: The feature extraction module is specifically configured as follows: Standardizing the power history data to obtain standard data; The principal component analysis method is used to reduce the dimension of the standard data and extract the target features.
7. The power load forecasting system based on intelligent algorithm according to claim 5 is characterized in that: The parameter generation module is specifically configured as follows: Obtaining information entropy of each target feature; Determine a first weight coefficient corresponding to the target feature based on the information entropy; the first weight coefficient represents the degree of influence of the target feature on the power load forecast data; The corresponding target feature is modified using the first weight coefficient to generate corresponding target training parameters.
8. The power load forecasting system based on intelligent algorithm according to claim 5 is characterized in that: The data output module is specifically configured as follows: Determine the second weight coefficients corresponding to the submodules based on the correlation between the target training parameters and the submodules in the adaptive comprehensive training model; Inputting the target training parameters into the adaptive comprehensive training model, and outputting the preliminary power load forecast data of each submodule; The preliminary power load forecast data and the corresponding second weight coefficient are combined to obtain power load forecast data.
9. An electronic device comprising a processor and a memory; the memory stores a computer program, wherein: When the computer program is executed by the processor, the computer program implements the power load forecasting method based on the intelligent algorithm according to any one of claims 1 to 4.
10. A computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.