Electric energy demand prediction method and system based on multi-source heterogeneous data fusion
Through the fusion of multi-source heterogeneous data, combined with historical electricity, energy, environment and economic data, a power energy demand prediction model is constructed, which solves the problem of large-scale power demand prediction errors for a single user in the existing technology, and realizes the power energy of large areas Accurate prediction of consumption improves energy management efficiency and system stability.
Patent Information
- Application Number
- CN202410983399.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-08
AI Technical Summary
The existing power demand forecasting methods are mainly aimed at a single user, and only considers one data source, which leads to large errors and inaccurate prediction of urban or regional power demand.
Using a method based on multi-source heterogeneous data fusion, combining historical electricity consumption data, energy data, environmental data and economic data, time correlation, energy correlation, environmental correlation and economic correlation prediction sub-models are constructed, and the power energy demand prediction model is fused to form a power energy demand prediction model. Accurately predict power consumption in large areas.
It improves the accuracy of power demand forecasting and reduces errors, helps power companies and users better plan and manage electricity, improves energy utilization efficiency, reduces energy costs, and ensures balanced and stable operation of the power system.
Smart Images

Figure CN118917878A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power demand forecasting, and in particular to a method and system for forecasting power energy demand based on multi-source heterogeneous data fusion. Background Art
[0002] Current electricity demand forecasting technologies primarily target single users, such as households, factories, or businesses, lacking energy consumption forecasts for industrial parks and cities. Research on urban and regional energy demand is limited, and existing methods primarily use historical energy consumption as input to predict future demand. Numerous factors influence urban and regional energy demand, and using only energy consumption as a guide for forecasting presents significant limitations.
[0003] Therefore, the power energy demand forecast considering only one data source has problems such as large errors and inaccurate predictions.
[0004] In view of this, this application is hereby filed. Summary of the Invention
[0005] The technical problem to be solved by the present invention is that the existing electricity demand forecasting method is mainly aimed at a single user, and the electricity energy demand forecast that only considers one data source has problems such as large errors and inaccurate predictions. The purpose of the present invention is to provide an electricity energy demand forecasting method and system based on multi-source heterogeneous data fusion, which targets the electricity demand of a large area, integrates historical electricity consumption data, energy data, environmental data and economic data to realize integrated prediction of regional electricity consumption, and predicts the electricity consumption in a certain time period in the future, with accurate predictions and small errors. The present invention helps power companies and users to better plan and manage electricity consumption, improve energy utilization efficiency, reduce energy costs, and contribute to the balanced and stable operation of the power system.
[0006] The present invention is achieved through the following technical solutions:
[0007] In a first aspect, the present invention provides a method for predicting electric energy demand based on multi-source heterogeneous data fusion, the method comprising:
[0008] Obtain historical electricity consumption data and build a time-correlated prediction sub-model based on the historical electricity consumption data;
[0009] Based on energy data, an energy correlation prediction sub-model is constructed; energy data includes raw coal, crude oil and natural gas data;
[0010] Based on environmental data, an environmental correlation prediction sub-model is constructed; environmental data includes precipitation, temperature, humidity, sunshine hours and air quality data;
[0011] Based on economic data, an economic correlation prediction sub-model is constructed; the economic data includes the proportion of the primary industry;
[0012] According to the time-related prediction sub-model, energy-related prediction sub-model, environment-related prediction sub-model and economic-related prediction sub-model, an electric energy demand prediction model is obtained by integration;
[0013] The electric energy demand forecasting model is used to forecast the electric energy consumption of the weekly data to be forecasted, and the final forecast value of the electric energy demand is output.
[0014] The present invention combines historical electricity consumption data, energy data, environmental data, and economic data to construct a forecast model for regional electricity consumption. The specific solution is to use one week's electricity consumption as the basic processing unit, and use the consumed electricity data (i.e., historical electricity consumption data) to predict the electricity consumption of the forecasted weekly data (e.g., the next week). At the same time, energy data, environmental data, and economic data are used to predict the electricity demand of the forecasted weekly data (e.g., the next week). The forecast results based on electricity, energy data, environmental data, and economic data are fused and corrected to output the final forecast value of electricity demand. The present invention has accurate predictions and small errors.
[0015] Furthermore, based on the historical electricity consumption data, a time-correlated prediction sub-model is constructed, including:
[0016] Based on the weekly energy consumption of the previous n weeks, where n is a positive integer, calculate the weekly average energy consumption and the deviation value of each energy week;
[0017] According to the deviation values of each electric energy cycle, an electric energy cycle deviation vector is formed;
[0018] A time-correlated prediction sub-model is constructed based on the weekly mean of electric energy consumption, the weekly deviation vector of electric energy and the time-correlated prediction coefficient vector.
[0019] Furthermore, the expression of the time-related prediction sub-model is:
[0020]
[0021] in, is the first predicted value, i.e., the result predicted by the time-related prediction sub-model; W = (W1, W2, ..., W8) is the time-related prediction coefficient vector; ΔE is the electric energy cycle deviation vector, Δe i is the deviation value of each electric energy cycle, Δe i =e i -e, e is the weekly average of electricity consumption, e i The energy consumed in the i-th week among the previous n weeks.
[0022] Furthermore, the solution to the time-dependent prediction coefficient vector is:
[0023] The first objective function is constructed by minimizing the square of the actual value of the electricity consumption historical data in the previous week minus the first predicted value. Where E1 is the actual value of the electricity consumption historical data in the previous week;
[0024] Solve the first objective function and obtain the time-correlated prediction coefficient vector W = ΔE T -eΔE T ΔE -1 .
[0025] Furthermore, the expression of the energy correlation prediction sub-model is:
[0026]
[0027] in, is the second predicted value, i.e. the result predicted by the energy-related prediction sub-model; x1, x2, and x3 are the estimated consumption values of raw coal, crude oil, and natural gas for the next week respectively; a i is the energy correlation prediction coefficient; X is the energy consumption vector X=[x1,x2,x3] T ; A = [a1, a2, a3] is the energy correlation prediction vector, A = X T E1X -1 ; E1 is the actual value of the historical electricity consumption data of the previous week.
[0028] Furthermore, the expression of the environmental association prediction sub-model is:
[0029]
[0030] in, is the third predicted value, i.e. the result predicted by the environmental association prediction sub-model; Y (1) , Y (2) ,…Y (5) The five vectors are the precipitation, temperature, humidity, sunshine hours, and air quality values of the previous week; n=1,2,3,4,5, is the value of the nth variable on the i-th day; is the third predicted value when n=5; B n is the environmental association prediction vector B n =[b1 (n) ,b2 (n) ,…,b7 (n) ] T .
[0031] Furthermore, the expression of the economic correlation prediction sub-model is:
[0032] in, is the fourth prediction value, that is, the result predicted by the economic correlation prediction sub-model.
[0033] Furthermore, the expression of the power energy demand forecasting model is:
[0034]
[0035] in, is the predicted value of the final electricity demand, is the i-th predicted value.
[0036] In a second aspect, the present invention provides a power energy demand forecasting system based on multi-source heterogeneous data fusion, the system comprising:
[0037] The first sub-model construction unit is used to obtain historical power consumption data and construct a time-related prediction sub-model based on the historical power consumption data;
[0038] The second sub-model construction unit is used to construct an energy correlation prediction sub-model based on energy data; the energy data includes raw coal, crude oil and natural gas data;
[0039] The third sub-model construction unit is used to construct an environmental correlation prediction sub-model based on environmental data; the environmental data includes precipitation, temperature, humidity, sunshine hours and air quality data;
[0040] The fourth sub-model construction unit is used to construct an economic correlation prediction sub-model based on economic data; the economic data includes the proportion of the primary industry;
[0041] A fusion unit is used to obtain an electric energy demand prediction model by fusing the time-related prediction sub-model, the energy-related prediction sub-model, the environment-related prediction sub-model and the economic-related prediction sub-model;
[0042] The prediction unit is used to use the electric energy demand prediction model to predict the electric energy consumption of the weekly data to be predicted, and output the final predicted value of the electric energy demand.
[0043] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for predicting electric power demand based on multi-source heterogeneous data fusion is implemented.
[0044] In a fourth aspect, the present invention further provides a computer-readable storage medium, which stores a computer program, and is characterized in that when the computer program is executed by a processor, it implements the above-mentioned power energy demand forecasting method based on multi-source heterogeneous data fusion.
[0045] Compared with the prior art, the present invention has the following advantages and beneficial effects:
[0046] The present invention is based on a multi-source heterogeneous data fusion method and system for predicting power demand. This system integrates historical electricity consumption data, energy data, environmental data, and economic data to predict regional power consumption for a large region. This system accurately predicts power consumption within a specific time period in the future with minimal error. This system helps power companies and users better plan and manage electricity use, improve energy efficiency, reduce energy costs, and contribute to the balanced and stable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings:
[0048] Figure 1 This is a flow chart of the electric energy demand forecasting method based on multi-source heterogeneous data fusion of the present invention;
[0049] Figure 2 This is a detailed flow chart of the electric energy demand forecasting method based on multi-source heterogeneous data fusion of the present invention;
[0050] Figure 3 Flowchart of the steps for solving the time-related prediction coefficient vector W of the present invention;
[0051] Figure 4 This is a structural block diagram of the power energy demand forecasting system based on multi-source heterogeneous data fusion of the present invention. DETAILED DESCRIPTION
[0052] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with examples and drawings. The exemplary embodiments of the present invention and their descriptions are only used to explain the present invention and are not intended to limit the present invention.
[0053] Example 1
[0054] like Figure 1 As shown, the present invention is based on a method for predicting power energy demand based on multi-source heterogeneous data fusion, the method comprising:
[0055] Step A: acquiring historical electricity consumption data, and constructing a time-correlated prediction sub-model based on the historical electricity consumption data;
[0056] Step A specifically includes:
[0057] Based on the weekly energy consumption e1, e2, ..., e8 of the previous 8 weeks, calculate the weekly average energy consumption e and the deviation value Δe of each energy week i =ei -e, i=1,…,8;
[0058] According to the deviation values of each electric energy cycle, an electric energy cycle deviation vector ΔE is formed;
[0059] Based on the weekly energy consumption mean value e, the weekly energy deviation vector ΔE, and the time-related prediction coefficient vector W, a time-related prediction sub-model is constructed. The expression of the time-related prediction sub-model is:
[0060]
[0061] in, is the first predicted value, i.e., the result predicted by the time-related prediction sub-model; W = (W1, W2, ..., W8) is the time-related prediction coefficient vector; ΔE is the electric energy cycle deviation vector, Δe i is the deviation value of each electric energy cycle, Δe i =e i -e, e is the weekly average of electricity consumption, e i The energy consumed in the i-th week among the previous n weeks.
[0062] In this embodiment, Figure 3 As shown, the steps for solving the time-related prediction coefficient vector W are:
[0063] The first objective function is constructed by minimizing the square of the actual value of the electricity consumption historical data in the previous week minus the first predicted value. Where E1 is the actual value of the electricity consumption historical data in the previous week;
[0064] Solve the first objective function, Obtain the time-related prediction coefficient vector;
[0065] W=ΔE T -eΔE T ΔE -1 (2)
[0066] Thus, W is determined by the existing historical data. Under the premise of determining W, the energy consumption of the next week is calculated using formula (1) Make predictions.
[0067] Step B: constructing an energy correlation prediction sub-model based on energy data; the energy data includes raw coal, crude oil and natural gas data;
[0068] When the consumption of raw coal, crude oil, and natural gas for the next week is announced, the expression of the energy correlation prediction sub-model is:
[0069]
[0070] in, is the second predicted value, i.e. the result predicted by the energy-related prediction sub-model; x1, x2, and x3 are the estimated consumption values of raw coal, crude oil, and natural gas for the next week respectively; a i is the energy correlation prediction coefficient; X is the energy consumption vector X=[x1,x2,x3] T ; A = [a1, a2, a3] is the energy correlation prediction vector.
[0071] In this embodiment, the steps for solving the energy correlation prediction vector A are:
[0072] Establish the objective function to solve A: Depend on A=X T E1X -1 ;
[0073] E1 is the actual value of the electricity consumption historical data of the previous week; under the premise of determining A, use formula (3) to calculate the electricity consumption of the next week Make predictions.
[0074] Step C, constructing an environmental correlation prediction sub-model based on environmental data; the environmental data includes precipitation, temperature, humidity, sunshine hours and air quality data;
[0075] The precipitation, temperature, humidity, sunshine hours, and air quality AQI values of the previous week were Y (1) , Y (2) ,…Y (5) 5 vectors, among which is the value of the nth variable on the i-th day.
[0076] Establish an environmental correlation prediction sub-model. The expression of the environmental correlation prediction sub-model is:
[0077]
[0078] Among them, f j (Y (j) ) is the jth vector Y of the relationship (j) The contribution function value of .
[0079] When n=5, the predicted value of power consumption of the environment-related prediction sub-model is:
[0080]
[0081] Among them B n is the environmental association prediction vector, B n =[b1 (n) ,b2 (n) ,…,b7 (n) ]T , b1 (n) is the environmental association prediction value on the first day of the week, b2 (n) is the environmental association prediction value for the second day of the week, ..., b7 (n) It is the predicted value of environmental association on the 7th day of the week.
[0082] to get value, then we need to get B in formula (5) n value.
[0083] To solve The value of , establish the objective function:
[0084] right Taylor series expansion:
[0085]
[0086] make: but:
[0087]
[0088] Specific solution process: Let G1 = 1, H1 = 1
[0089] From the above formula (6), we can see that B n The solution can be obtained by the corresponding parameters of the first n-1 variables;
[0090] Therefore, the iteration can be built:
[0091]
[0092] for n=1:4
[0093] According to formula (6), calculate B n+1
[0094] According to formula (6), f n+1 (Y (n+1) )
[0095] According to formula (4),
[0096] end for;
[0097] end if;
[0098] To obtain environmental The prediction expression is:
[0099]
[0100] in, is the third predicted value, i.e. the result predicted by the environmental association prediction sub-model; Y (1) , Y (2) ,…Y (5) The five vectors are the precipitation, temperature, humidity, sunshine hours, and air quality values of the previous week; n=1,2,3,4,5, is the value of the nth variable on the i-th day; is the third predicted value when n=5; B n is the environmental association prediction vector B n =[b1 (n) ,b2 (n) ,…,b7 (n) ] T .
[0101] Step D: construct an economic correlation prediction sub-model based on economic data; the economic data includes the proportion of the primary industry;
[0102] In this embodiment, the expression of the economic correlation prediction sub-model is:
[0103]
[0104] in, is the fourth prediction value, that is, the result predicted by the economic correlation prediction sub-model.
[0105] Step E: Based on the time-related prediction sub-model, the energy-related prediction sub-model, the environment-related prediction sub-model, and the economic-related prediction sub-model, a power energy demand prediction model is obtained by integrating the power energy demand prediction model. The expression of the power energy demand prediction model is:
[0106]
[0107] in, is the predicted value of the final electricity demand, is the i-th predicted value.
[0108] The fused formula (9) of the present invention can effectively reduce E i The influence of large fluctuations in is closer to the true value. This is because the first term in the numerator of formula (9) Can effectively weaken The median has a relatively large impact, while the second term in the numerator Can effectively weaken The median has a relatively small effect.
[0109] Step F: Use the power energy demand forecasting model to forecast the power consumption of the forecasted weekly data and output the final power demand forecast value.
[0110] The present invention combines historical electricity consumption data, energy data, environmental data, and economic data to construct a forecast model for regional electricity consumption. The specific solution is to use one week's electricity consumption as the basic processing unit, and use the consumed electricity data (i.e., historical electricity consumption data) to predict the electricity consumption of the forecasted weekly data (e.g., the next week). At the same time, energy data, environmental data, and economic data are used to predict the electricity demand of the forecasted weekly data (e.g., the next week). The forecast results based on electricity, energy data, environmental data, and economic data are fused and corrected to output the final forecast value of electricity demand. The present invention has accurate predictions and small errors.
[0111] Example 2
[0112] like Figure 4 As shown, the difference between this embodiment and embodiment 1 is that this embodiment provides an electric power demand forecasting system based on multi-source heterogeneous data fusion, which corresponds to the electric power demand forecasting method based on multi-source heterogeneous data fusion in embodiment 1 in a one-to-one manner; the system includes:
[0113] The first sub-model construction unit is used to obtain historical power consumption data and construct a time-related prediction sub-model based on the historical power consumption data;
[0114] The second sub-model construction unit is used to construct an energy correlation prediction sub-model based on energy data; the energy data includes raw coal, crude oil and natural gas data;
[0115] The third sub-model construction unit is used to construct an environmental correlation prediction sub-model based on environmental data; the environmental data includes precipitation, temperature, humidity, sunshine hours and air quality data;
[0116] The fourth sub-model construction unit is used to construct an economic correlation prediction sub-model based on economic data; the economic data includes the proportion of the primary industry;
[0117] A fusion unit is used to obtain an electric energy demand prediction model by fusing the time-related prediction sub-model, the energy-related prediction sub-model, the environment-related prediction sub-model and the economic-related prediction sub-model;
[0118] The prediction unit is used to use the electric energy demand prediction model to predict the electric energy consumption of the weekly data to be predicted, and output the final predicted value of the electric energy demand.
[0119] Among them, the execution process of each unit can be executed according to the process steps of the power energy demand prediction method based on multi-source heterogeneous data fusion in Example 1, and will not be repeated one by one in this embodiment.
[0120] At the same time, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the computer program, the above-mentioned power energy demand prediction method based on multi-source heterogeneous data fusion is implemented.
[0121] At the same time, the present invention also provides a computer-readable storage medium, which stores a computer program, and is characterized in that when the computer program is executed by a processor, it implements the above-mentioned power energy demand prediction method based on multi-source heterogeneous data fusion.
[0122] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.
[0123] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can 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 process. 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.
[0124] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work 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 The function specified in one or more boxes.
[0125] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1A step that specifies a function in one or more boxes.
[0126] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.
Claims
1. Power energy demand forecasting method based on multi-source heterogeneous data fusion, It is characterized in that The method includes: Obtain historical electricity consumption data, and build a time-related prediction sub-model based on the historical electricity consumption data; Based on energy data, an energy correlation prediction sub-model is constructed; the energy data includes raw coal, crude oil and natural gas data; Based on environmental data, an environmental association prediction sub-model is constructed; the environmental data includes precipitation, temperature, humidity, sunshine hours and air quality data; Based on economic data, construct an economic correlation prediction sub-model; the economic data includes the proportion of the primary industry; According to the time-related prediction sub-model, the energy-related prediction sub-model, the environment-related prediction sub-model and the economic-related prediction sub-model, a power energy demand prediction model is obtained by integration; The electric energy demand forecasting model is used to forecast the electric energy consumption of the weekly data to be forecasted, and the forecast value of the final electric energy demand is output.
2. According to claim 1, the method for predicting power demand based on multi-source heterogeneous data fusion, It is characterized in that Based on the historical electricity consumption data, a time-related prediction sub-model is constructed, including: Based on the weekly energy consumption of the previous several weeks, calculate the weekly energy consumption average and the weekly deviation value of each energy; According to each electric energy cycle deviation value, an electric energy cycle deviation vector is formed; A time-related prediction sub-model is constructed according to the weekly electric energy consumption mean, the weekly electric energy deviation vector and the time-related prediction coefficient vector.
3. The method for predicting electric power demand based on multi-source heterogeneous data fusion according to claim 2, It is characterized in that The expression of the time-related prediction sub-model is: in, is the first prediction value, i.e., the result predicted by the time-related prediction sub-model; W = (W 1 , W 2 , …, W 8 ) is the time-related prediction coefficient vector; ΔE is the electric energy cycle deviation vector, Δe i is the deviation value of each electric energy cycle, Δe i =e i -e, e is the weekly average of electricity consumption, e i The energy consumed in the ith week among the previous n weeks.
4. The method for predicting electric power demand based on multi-source heterogeneous data fusion according to claim 3, It is characterized in that The solution method of the time-related prediction coefficient vector W is: The first objective function is constructed by taking the square of the actual value of the electricity consumption historical data in the previous week minus the first predicted value as the minimum. Where E 1 The actual value of the historical data of electricity consumption in the previous week; Solve the first objective function to obtain the time-related prediction coefficient vector W = ΔE T -eΔE T ΔE -1 .
5. The method for predicting electric power demand based on multi-source heterogeneous data fusion according to claim 1, It is characterized in that The expression of the energy correlation prediction sub-model is: in, is the second predicted value, i.e., the result predicted by the energy-related prediction sub-model; x 1 , x 2 , x 3 The estimated consumption values of raw coal, crude oil and natural gas for the next week are announced respectively; a i is the energy correlation prediction coefficient; X is the energy consumption vector X = [x 1 ,x 2 ,x 3 ] T ; A=[a 1 ,a 2 ,a 3 ] is the energy related prediction vector, A=X T E 1 X -1 ; E 1 It is the actual value of the historical data of electricity consumption in the previous week.
6. The method for predicting electric power demand based on multi-source heterogeneous data fusion according to claim 1, It is characterized in that The expression of the environmental association prediction sub-model is: in, is the third prediction value, i.e., the result predicted by the environment-related prediction sub-model; Y (1) , Y (2) , ...Y (5) The five vectors are the precipitation, temperature, humidity, sunshine hours, and air quality values of the previous week; n=1,2,3,4,5, is the value of the nth variable on the ith day; is the third predicted value when n=5; B n is the environmental association prediction vector B n =[b 1 (n) ,b 2 (n) ,…,b 7 (n) ] T .
7. The method for predicting electric power demand based on multi-source heterogeneous data fusion according to claim 1, It is characterized in that The expression of the electric energy demand prediction model is: in, is the predicted value of the final electricity demand, is the i-th predicted value.
8. Power energy demand forecasting system based on multi-source heterogeneous data fusion, It is characterized in that The system includes: The first sub-model construction unit is used to obtain historical power consumption data and construct a time-related prediction sub-model based on the historical power consumption data; A second sub-model construction unit is used to construct an energy correlation prediction sub-model based on energy data; the energy data includes raw coal, crude oil and natural gas data; A third sub-model construction unit is used to construct an environment-related prediction sub-model based on environmental data; the environmental data includes precipitation, temperature, humidity, sunshine hours and air quality data; The fourth sub-model construction unit is used to construct an economic correlation prediction sub-model based on economic data; the economic data includes the proportion of the primary industry; A fusion unit, used for fusing the time-related prediction sub-model, the energy-related prediction sub-model, the environment-related prediction sub-model and the economic-related prediction sub-model to obtain an electric energy demand prediction model; The prediction unit is used to use the electric energy demand prediction model to predict the electric energy consumption of the weekly data to be predicted, and output the final predicted value of the electric energy demand.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the method for predicting electric power demand based on multi-source heterogeneous data fusion as described in any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium storing a computer program. It is characterized in that When the computer program is executed by a processor, the method for predicting electric power demand based on multi-source heterogeneous data fusion as described in any one of claims 1 to 7 is implemented.
Citation Information
Cited By
Energy demand prediction system and method based on data fusion and deep learning
CN120278427A
