Electric quantity prediction method and device for multi-type electric quantity decomposition, equipment and medium

Abstract

The invention relates to the related technical field of electric quantity prediction, in particular to an electric quantity prediction method and device for multi-type electric quantity decomposition, equipment and a medium. The method comprises the following steps: acquiring electric quantity data; decomposing the electric quantity data into multiple types of electric quantities; the electric quantity of the type at least comprises basic electric quantity and temperature control electric quantity; and respectively predicting the electric quantity of each type to obtain a predicted value of the electric quantity of each type: carrying out linear reconstruction on the predicted value of the electric quantity of each type to obtain a total predicted electric quantity. Therefore, the electric quantity can be predicted more accurately.

Description

Technical Field

[0001] The present application relates to the technical field related to electricity forecasting, and specifically to an electricity forecasting method, device, equipment and medium for multi-type electricity decomposition. Background Art

[0002] In today's power system, accurate power forecasting plays a vital role in the stable operation and optimal dispatch of the power grid. With the continuous growth of power demand and the increasing diversification of energy structure, traditional power forecasting methods are gradually unable to adapt to the complex power system needs.

[0003] The change of electricity consumption is affected by a variety of factors, including the level of social and economic development, industrial structure, holidays, and meteorological conditions. However, the existing electricity consumption forecasting technology has some limitations and cannot make accurate electricity consumption forecasts. Summary of the invention

[0004] In view of this, the embodiments of the present application are committed to providing a method, device, equipment and medium for power prediction with multi-type power decomposition to achieve more precise power prediction.

[0005] The present application provides a method for predicting power consumption by decomposing multiple types of power consumption, including:

[0006] Get power data;

[0007] Decomposing the power data into multiple types of power;

[0008] The types of electricity include at least: basic electricity and temperature-controlled electricity;

[0009] Predict each type of electricity separately and get the predicted value of each type of electricity:

[0010] The predicted values ​​of each type of electricity are linearly reconstructed to obtain the total predicted electricity.

[0011] In some embodiments, each type of electricity is predicted respectively to obtain a predicted value of each type of electricity, including:

[0012] Predict the basic electricity quantity and obtain the basic electricity quantity prediction value;

[0013] The temperature control power consumption is predicted to obtain a predicted value of the temperature control power consumption.

[0014] In some embodiments, predicting the basic power to obtain a basic power prediction value includes:

[0015] Based on the Pearson correlation coefficient, the influencing factors that are strongly correlated with the base load power consumption are determined;

[0016] Establishing a basic electricity quantity prediction model based on the influencing factors;

[0017] Based on the basic electricity quantity prediction model, prediction is performed to obtain a basic electricity quantity prediction value.

[0018] In some embodiments, the basic power prediction model is as follows:

[0019] f(x)=a0+a1x1+a2x2+…+a n x n

[0020] f(x) is the objective function, which is used to calculate the basic power forecast value; a0, a1, a2, a n are polynomial fitting coefficients; x0, x1, x2, x n As an influencing factor.

[0021] In some embodiments, predicting the temperature control power to obtain a predicted value of the temperature control power includes:

[0022] Genetic algorithm is used to optimize Xgboost parameters, and the temperature control power consumption is predicted to obtain the predicted value of temperature control power consumption.

[0023] In some embodiments, linearly reconstructing the predicted values ​​of each type of electricity to obtain the total predicted electricity includes:

[0024] The predicted values ​​of each type of electricity are weighted and accumulated to obtain the total predicted electricity.

[0025] y=b1y1+b2y2

[0026] Where y is the total predicted power; b1 and b2 are weight coefficients, y1 and y2 are the basic power prediction value and temperature control power prediction value, respectively.

[0027] The present application also provides a power forecasting device for multi-type power decomposition, including:

[0028] Acquisition module, used to obtain power data;

[0029] A decomposition module, used for decomposing the power data into multiple types of power;

[0030] The types of electricity include at least: basic electricity and temperature-controlled electricity;

[0031] The prediction module is used to predict each type of electricity and obtain the predicted value of each type of electricity:

[0032] The reconstruction module is used to linearly reconstruct the predicted values ​​of each type of electricity to obtain the total predicted electricity.

[0033] The present application also provides an electronic device, comprising:

[0034] A processor, and a memory for storing a program executable by the processor;

[0035] The processor is used to implement the above-mentioned method for predicting power consumption by decomposing multiple types of power consumption by running the program in the memory.

[0036] The present application discloses a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the processor is enabled to execute the above-mentioned method for predicting electric quantity by decomposing multiple types of electric quantity.

[0037] The present application provides a method for predicting electric quantity by decomposing multiple types of electric quantity, firstly obtaining electric quantity data; decomposing the electric quantity data into multiple types of electric quantity; the types of electric quantity at least include: basic electric quantity and temperature-controlled electric quantity; predicting each type of electric quantity respectively to obtain the predicted value of each type of electric quantity: linearly reconstructing the predicted value of each type of electric quantity to obtain the total predicted electric quantity. In this way, in the scheme provided by the present application, by decomposing the electric quantity data into basic electric quantity and temperature-controlled electric quantity and predicting them respectively, the changing trend and characteristics of different types of electric quantity can be grasped more accurately, thereby improving the accuracy of electric quantity prediction. Independently predicting each type of electric quantity can better consider their respective influencing factors, reduce mutual interference, and make the prediction result more reliable. Linearly reconstructing the predicted value of each type of electric quantity to obtain the total predicted electric quantity can comprehensively consider the influence of multiple factors, make the total predicted electric quantity closer to the actual situation, and provide more powerful support for the planning, scheduling and operation of the power system. Accurate electric quantity prediction helps to optimize the configuration of electric resources, improve the operating efficiency of the power system, and reduce energy waste. In this way, it can help power companies to better formulate production plans and scheduling plans, enhance the stability and reliability of the power grid, and reduce the occurrence of power outages. Provide users with more stable and reliable power supply and improve user satisfaction. Help promote the integration and utilization of renewable energy and promote energy transformation and sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0039] Figure 1 It is a flowchart of a method for predicting electric power with multi-type electric power decomposition provided in an embodiment of the present application.

[0040] Figure 2 It is a partial flow chart of a method provided by an embodiment of the present application.

[0041] Figure 3 It is a structural schematic diagram of an electric power prediction device for multi-type electric power decomposition provided by an embodiment of the present application.

[0042] Figure 4 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0043] 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.

[0044] Application Overview

[0045] In the new power system, accurate power forecasting is crucial to the stable operation and optimal dispatching of the power grid. With the continuous growth of power demand and the diversification of energy structure, traditional forecasting methods have been unable to meet the increasingly complex needs of the power system. Accurately predicting the total power consumption in the region is helpful for the design of power grid dispatching, unit start-up and shutdown plans, and inter-regional transmission plans. Power changes are affected by multiple factors such as the level of social and economic development, industrial structure, holidays, and meteorological conditions. A large number of studies have shown that during the summer and winter, the proportion of temperature-controlled power affected by meteorological factors has exceeded 30% of the total power. The sixth report of the Intergovernmental Panel on Climate Change (IPCC) of the United Nations pointed out that human activities are causing the climate to warm at an unprecedented rate, reflecting the necessity of conducting power forecasting research based on meteorological factors.

[0046] The existing technical solutions mainly include: 1) Using the autoregressive moving average model to predict electricity and analyze the robustness of the model under different noise levels, but it cannot capture nonlinear relationships, and the prediction accuracy of electricity data with seasonal and trend characteristics is low. 2) The seasonal autoregressive moving average model is used to effectively handle seasonal and trend changes in electricity data, but the model is still a linear model, and the effect of handling nonlinear and mutation data is limited, and the model complexity is high. 3) Using nonlinear smoothing functions to capture the impact of weather information on electricity demand to improve prediction accuracy, but the method is essentially nonlinear regression, and an appropriate nonlinear form must be selected to capture nonlinear relationships, and it has high requirements for data volume and computing power. 4) Based on the multi-step least squares support vector machine (LS-SVM) model, it is used to predict short-term electricity demand, but the SVM model has a large amount of calculation and a long training time when processing large-scale data, and the model is sensitive to parameters and needs to be tuned through methods such as cross-validation. 5) Based on the hybrid model of graph convolutional neural network and long short-term memory network, by introducing wavelet transform and correlation analysis, it can process complex time series data and improve prediction accuracy. However, the long short-term memory network model takes a long time to train and is prone to overfitting.

[0047] In order to solve the above problems, the present invention proposes a power forecasting method based on multi-type power decomposition forecasting. Using the proposed forecasting method, a relatively accurate quarterly and annual basic power forecast and temperature control power forecast under extreme weather conditions during summer and winter can be obtained. The linear reconstruction of the forecast results of the two can obtain the total predicted power, which is conducive to the implementation of power grid optimization scheduling, stable power grid operation, unit start and stop plan and cross-regional transmission plan.

[0048] After introducing the basic principles of the present application, various non-limiting embodiments of the present application will be described in detail with reference to the accompanying drawings.

[0049] Exemplary Methods

[0050] Figure 1 FIG. 1 is a flow chart of a method for predicting power consumption by decomposing multiple types of power consumption provided by an embodiment of the present application. Figure 1 As shown, the method includes the following contents.

[0051] Step S110, obtaining power data;

[0052] This step is the starting point of the entire power forecasting method, and it is necessary to obtain data related to power. These data may come from monitoring equipment, sensors or other data sources of the power system, including historical power data, relevant economic indicator data, meteorological data, etc. Obtaining accurate and comprehensive power data is crucial for subsequent power decomposition and forecasting.

[0053] Step S120, decomposing the power data into multiple types of power;

[0054] Specifically, the types of electricity include at least basic electricity and temperature-controlled electricity. In this step, the acquired electricity data is decomposed into multiple types including at least basic electricity and temperature-controlled electricity. Basic electricity usually refers to the stable electricity part that is not affected by meteorological factors or is less affected, while temperature-controlled electricity refers to the part that is greatly affected by meteorological factors, such as the increase or decrease in electricity consumption of air conditioning, heating and other equipment due to temperature changes during summer and winter.

[0055] The purpose of decomposing electricity is to more accurately analyze and predict the changing trends of different types of electricity, thereby improving the accuracy of total electricity forecast. The specific decomposition method may be selected and adjusted according to the actual situation and data characteristics.

[0056] Step S130, predict each type of electricity respectively to obtain the predicted value of each type of electricity:

[0057] This step is to make separate predictions for each type of electricity after decomposition. For the prediction of basic electricity, a method based on the Pearson correlation coefficient may be used to determine the influencing factors that are strongly correlated with the basic load electricity consumption, and then a basic electricity prediction model is established based on these influencing factors, such as using a polynomial fitting method to fit the basic load electricity consumption, and finally obtain the predicted value of basic electricity.

[0058] For the prediction of temperature-controlled power, genetic algorithms may be used to optimize Xgboost parameters. The specific process includes standardizing the original data of temperature-controlled power, setting the relevant parameters of the genetic algorithm, such as the number of iterations, population size, crossover probability, mutation probability, etc., setting the parameter search range of Xgboost, encoding the parameters using binary coding, randomly generating the initial population, calculating individual fitness, performing the selection, crossover and mutation steps in genetic operations, and finally obtaining the optimal parameters of Xgboost to achieve the prediction of temperature-controlled power.

[0059] By adopting targeted forecasting methods for different types of electricity, we can better capture their respective changing patterns and characteristics and improve the accuracy of the forecast.

[0060] Step S140: linearly reconstruct the predicted values ​​of each type of electricity to obtain the total predicted electricity.

[0061] The purpose of linear reconstruction is to reasonably combine the predicted values ​​of basic power and temperature-controlled power to reflect their combined impact on the total power, thereby obtaining a more accurate total power forecast result. This method can effectively reduce the forecast error, improve the reliability of power forecast, and provide important decision-making basis for the optimal dispatch and stable operation of the power grid.

[0062] In some embodiments, the daily electricity consumption decomposition formula can be expressed as:

[0063] L=L b +L w (1)

[0064] Where, L b Is the basic power; L w The temperature control power.

[0065] Specifically, each type of electricity is predicted respectively to obtain a predicted value of each type of electricity, including: predicting basic electricity to obtain a basic electricity prediction value; predicting temperature-controlled electricity to obtain a temperature-controlled electricity prediction value.

[0066] The step of predicting the basic power quantity to obtain a basic power quantity prediction value includes:

[0067] Based on the Pearson correlation coefficient, influencing factors that are strongly correlated with basic load power consumption are determined; a basic power prediction model is established based on the influencing factors; and prediction is performed based on the basic power prediction model to obtain a basic power prediction value.

[0068] Specifically, the Pearson correlation coefficient is a statistical indicator used to measure the linear correlation between two variables. In this step, the Pearson correlation coefficient between the year-on-year growth rate of spring, autumn and annual base load and socioeconomic factors such as the year-on-year growth rate of GDP in each quarter, the year-on-year growth rate of GDP, and the year-on-year growth rate of CPI are calculated to determine which socioeconomic factors have a strong correlation with base load electricity consumption.

[0069] Pearson correlation coefficient ρ XY The calculation formula is:

[0070]

[0071] Where X is the socioeconomic factor that affects base load electricity consumption; Y is base load electricity consumption; C(X,Y) is the covariance; σ X , σ Y is the standard deviation of socioeconomic factors and electricity data.

[0072] After selecting the socio-economic influencing factors that are strongly correlated with the base load electricity consumption, the base load electricity consumption is fitted using a polynomial fitting method to establish a basic electricity prediction model.

[0073] f(x)=a0+a1x1+a2x2+…+a n x n (3)

[0074] Where f(x) is the objective function, which is used to calculate the basic power forecast value; a0, a1, a2, a n are polynomial fitting coefficients; x0, x1, x2, x n As an influencing factor.

[0075] In some embodiments, predicting the temperature-controlled power consumption to obtain a predicted value of the temperature-controlled power consumption includes: predicting the temperature-controlled power consumption to obtain a predicted value of the temperature-controlled power consumption using a genetic algorithm.

[0076] Specifically, after the original data of the temperature control power is calculated, the original data is standardized, and the zero mean standardization method is used to convert the data into a distribution with a mean of 0 and a standard deviation of 1, as follows:

[0077]

[0078] Where, X norm is the standardized data; X pre is the original data; μ is the mean of the data; σ is the standard deviation of the data.

[0079] The five hyperparameters of Xgboost are optimized using genetic algorithm, and the individual fitness function f(x) is established based on the genetic algorithm, as shown in the following formula:

[0080]

[0081] The Xgboost parameters are shown in Table 1:

[0082] Table 1 Xgboost parameter related meaning

[0083]

[0084]

[0085] The specific steps are as follows:

[0086] 1) Set the genetic algorithm parameters (iteration number T = 100, population size M = 3, crossover probability P c =0.8, mutation probability P m =0.1)

[0087] 2) Set the parameter search range of Xgboost.

[0088] 3) Use binary coding method to encode the parameters.

[0089] 4) Randomly generate M individuals to form an initial population pop(t), t=1.

[0090] 5) Calculate the individual fitness of the population pop(t) in the t-th generation according to the individual fitness evaluation function f(x).

[0091] 6) By using the algorithm operator (roulette wheel selection method) in genetic operations, a new population newpop(t) can be obtained.

[0092] 7) Based on the population newpop(t), by using the two-point crossover operator with a probability of P c Generate a new population crosspop(t). Here, an elitist retention strategy is adopted, and the optimal individual does not participate in the crossover operation.

[0093] 8) Based on crosspop(t), by using the single-point mutation operator with a probability of P b Generate a new population mutpop(t). Here, an elitist retention strategy is adopted, and the optimal individual does not participate in the crossover operation.

[0094] 9) Determine whether to stop the optimization. When t < T, then t = t + 1, and the new population pop(t) = mutpop(t - 1), and return to step 5. Otherwise, stop the optimization and enter step 10.

[0095] 10) Calculate the individual fitness of mutpop(t) by using f(x), then select the individual with the best fitness, and after decoding this individual, obtain the optimal parameters of Xgboost to realize the automatic determination of parameters and improve the accuracy of the algorithm.

[0096] The specific process is as Figure 2 shown:[[]]END]]

[0097] In some embodiments, the linear reconstruction of the predicted values of each type of electricity quantity to obtain the total predicted electricity quantity includes:

[0098] Perform weighted accumulation on the predicted values of each type of electricity quantity to obtain the total predicted electricity quantity.

[0099] Perform linear reconstruction on the predicted values of the basic electricity quantity and the temperature control electricity quantity respectively to obtain the total predicted electricity quantity, as shown in the following formula:

[0100] y = b1y1 + b2y2 (7)

[0102] In the formula, y is the total predicted electricity quantity; b1 and b2 are the weight coefficients respectively, and in the present invention, 0.7 and 0.3 are taken respectively; y1 and y2 are the predicted values of the basic electricity quantity and the temperature control electricity quantity respectively.

[0103] In summary, in the solution provided by this application, in order to solve the problem of low prediction accuracy of traditional power prediction methods, a power prediction method based on multi-type power decomposition prediction is proposed. In order to further improve the prediction accuracy, the total power is decomposed into basic power and temperature-controlled power, and the Pearson correlation coefficient method is used in combination with polynomial fitting to predict the basic power, and the genetic algorithm is used to optimize the Xgboost parameters to achieve the prediction of temperature-controlled power and improve the prediction accuracy. The total power is greatly affected by the temperature-controlled power. The method of power decomposition prediction solves the problem of low power prediction accuracy.

[0104] The basic idea of ​​this method is: first, decompose the total electricity into basic electricity and temperature-controlled electricity, calculate the Pearson correlation coefficient between the basic electricity with a large proportion and the relevant economic indicators, select the economic indicators with higher correlation, and fit and predict the basic electricity; secondly, introduce the genetic algorithm to optimize the Xgboost parameters to achieve good prediction of the rapidly changing temperature-controlled electricity; finally, linearly reconstruct the prediction results of the two to obtain the total predicted electricity. With this setting, 1) consider the impact of temperature-controlled electricity on the total electricity, decompose it for prediction, and improve the prediction accuracy. 2) The electricity decomposition method does not rely on mathematical statistical laws. The Pearson correlation coefficient is used to determine the influencing factors that are strongly correlated between basic load electricity consumption and social economy, and the polynomial fitting method is used to fit the basic load electricity consumption, establish a basic electricity prediction model, and avoid the influence of statistical errors. 3) The genetic algorithm is used to optimize the Xgboost parameters. The model is simple and can effectively handle the seasonal and trend changes in electricity data, and it is used for temperature-controlled electricity prediction to further improve the prediction accuracy. 4) The predicted values ​​of basic electricity and temperature-controlled electricity are linearly reconstructed, and the predicted values ​​of the two are reasonably combined to effectively reduce the prediction error.

[0105] Exemplary Devices

[0106] The device embodiments of the present application can be used to execute the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

[0107] Figure 3 FIG. 1 is a block diagram of a device for predicting power consumption by decomposing multiple types of power consumption provided by an embodiment of the present application. Figure 3 As shown, the device comprises:

[0108] An acquisition module 31 is used to acquire power data;

[0109] A decomposition module 32, used for decomposing the power data into multiple types of power;

[0110] The types of electricity include at least: basic electricity and temperature-controlled electricity;

[0111] The prediction module 33 is used to predict each type of electricity and obtain the predicted value of each type of electricity:

[0112] The reconstruction module 34 is used to linearly reconstruct the predicted values ​​of each type of electricity to obtain the total predicted electricity.

[0113] Exemplary Electronic Devices

[0114] Below, reference Figure 4 To describe an electronic device according to an embodiment of the present application. Figure 4 A block diagram of an electronic device according to an embodiment of the present application is illustrated.

[0115] like Figure 4 As shown, electronic device 400 includes one or more processors 410 and memory 420 .

[0116] The processor 410 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 400 to perform desired functions.

[0117] The memory 420 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. The non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 410 may run the program instructions to implement the multi-type power decomposition power prediction method of each embodiment of the present application described above and / or other desired functions. Various contents such as category correspondences may also be stored in the computer-readable storage medium.

[0118] In one example, the electronic device 400 may further include: an input device 430 and an output device 440 , and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0119] In addition, the input device 430 may also include, for example, a keyboard, a mouse, an interface, etc. The output device 440 may output various information to the outside, including analysis results, etc. The output device 440 may include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, etc.

[0120] Of course, to simplify, Figure 4Only some of the components in the electronic device related to the present application are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, the electronic device may further include any other appropriate components according to specific application conditions.

[0121] Exemplary computer program products and computer-readable storage media

[0122] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the power prediction method for multi-type power decomposition according to various embodiments of the present application described in the above "Exemplary Method" section of this specification.

[0123] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0124] In addition, an embodiment of the present application may also be a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, causes the processor to execute the steps of the method for predicting electric power with multi-type electric power decomposition according to various embodiments of the present application described in the above “Exemplary Method” section of this specification.

[0125] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable 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.

[0126] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A method for predicting power consumption by decomposing multiple types of power consumption, characterized in that: include: Get power data; Decomposing the power data into multiple types of power; The types of electricity include at least: basic electricity and temperature-controlled electricity; Predict each type of electricity separately and get the predicted value of each type of electricity: The predicted values ​​of each type of electricity are linearly reconstructed to obtain the total predicted electricity.

2. The method for predicting electric quantity by decomposing multiple types of electric quantity according to claim 1, characterized in that: Each type of electricity is predicted separately to obtain the predicted value of each type of electricity, including: Predict the basic electricity quantity and obtain the basic electricity quantity prediction value; The temperature control power consumption is predicted to obtain a predicted value of the temperature control power consumption.

3. The method for predicting electric quantity by decomposing multiple types of electric quantity according to claim 2, characterized in that: The step of predicting the basic power quantity to obtain a basic power quantity prediction value includes: Based on the Pearson correlation coefficient, the influencing factors that have a strong correlation with the base load power consumption are determined; Establishing a basic electricity quantity prediction model based on the influencing factors; Based on the basic electricity quantity prediction model, prediction is performed to obtain a basic electricity quantity prediction value.

4. The method for predicting electric quantity by decomposing multiple types of electric quantity according to claim 3 is characterized in that: The basic electricity prediction model is as follows: f(x)=a0+a1x1+a2x2+…+a n x n f(x) is the objective function, which is used to calculate the basic power forecast value; a0, a1, a2, a n are polynomial fitting coefficients; x0, x1, x2, x n As an influencing factor.

5. The method for predicting electric quantity by decomposing electric quantity into multiple types according to claim 2, characterized in that: The temperature control power consumption is predicted to obtain the predicted value of the temperature control power consumption, including: Genetic algorithm is used to optimize Xgboost parameters, and the temperature control power consumption is predicted to obtain the predicted value of temperature control power consumption.

6. The method for predicting electric quantity by decomposing multiple types of electric quantity according to claim 2, characterized in that: The predicted values ​​of each type of electricity are linearly reconstructed to obtain the total predicted electricity, including: The predicted values ​​of each type of electricity are weighted and accumulated to obtain the total predicted electricity.

7. The method for predicting electric quantity by decomposing multiple types of electric quantity according to claim 6, characterized in that: The formula for total predicted electricity is as follows: y=b1y1+b2y2 Where y is the total predicted power; b1 and b2 are weight coefficients, y1 and y2 are the basic power prediction value and temperature control power prediction value, respectively.

8. A device for predicting power consumption by decomposing multiple types of power consumption, characterized in that: include: Acquisition module, used to obtain power data; A decomposition module, used for decomposing the power data into multiple types of power; The types of electricity include at least: basic electricity and temperature-controlled electricity; The prediction module is used to predict each type of electricity and obtain the predicted value of each type of electricity: The reconstruction module is used to linearly reconstruct the predicted values ​​of each type of electricity to obtain the total predicted electricity.

9. An electronic device, characterized in that: include: A processor, and a memory for storing a program executable by the processor; The processor is used to implement the power prediction method of multi-type power decomposition as described in any one of claims 1 to 7 by running the program in the memory.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the method for predicting electric quantity by decomposing multiple types of electric quantity as claimed in any one of claims 1 to 7.

Images