A regional power consumption intelligent prediction method based on multi-source heterogeneous data

Abstract

The present application relates to the technical field of power consumption prediction, and particularly relates to a regional power consumption intelligent prediction method based on multi-source heterogeneous data, which comprises the following steps: determining the power consumption prediction influence tendency for the next preset monitoring period based on the power change characteristic value; when it is determined that the next preset monitoring period is a strong influence tendency for power consumption prediction, determining that the power consumption is corrected by the elastic influence weight coefficient, and determining whether the prediction for the power consumption is qualified based on the energy change characteristic value; when it is determined that the prediction for the power consumption is abnormal, correcting the prediction parameter for the power consumption in the next preset monitoring period based on the regional industry change quantitative value, including identifying the main analysis data set, determining whether to increase the model training set or separate the prediction model based on the period quantitative value of the main analysis data, and performing targeted analysis and prediction on the power consumption according to the specific situation of the power elasticity coefficient of the region to be predicted, thereby improving the prediction efficiency of the power consumption in the region.

Description

Technical Field

[0001] This invention relates to the field of electricity consumption forecasting technology, and in particular to a regional electricity consumption intelligent forecasting method based on multi-source heterogeneous data. Background Technology

[0002] In today's society, electricity, as a vital energy source, plays a crucial role in ensuring stable economic development, social stability, and the quality of life for the people through its stable supply and rational distribution. Accurate forecasting of regional electricity consumption not only helps power companies rationally plan power generation, optimize grid dispatch, reduce power generation costs, and minimize energy waste, but also provides a scientific basis for governments to formulate energy policies and plan energy infrastructure construction.

[0003] Currently, there are numerous methods for regional electricity consumption forecasting. However, with rapid economic development, continuous adjustments to the energy structure, and increasingly diversified user electricity consumption behaviors, traditional forecasting methods often struggle to fully consider the impact of multi-source heterogeneous data, thus limiting the accuracy and reliability of the forecast results. Multi-source heterogeneous data comes from a wide range of sources and comes in various formats, containing a wealth of information related to electricity consumption.

[0004] In cities, as residents' living standards improve and urban functions are continuously enhanced, residential and commercial electricity demands exhibit different trends. In industrial parks, the upgrading and adjustment of industrial structures directly impact industrial electricity consumption. The complexity and diversity of economic activities, the constantly changing energy consumption structure, and the dynamic adjustment of industrial development pose significant challenges to electricity consumption forecasting.

[0005] Chinese Patent Publication No. CN116108979A discloses a cross-cycle multi-source heterogeneous power data processing system, including a data acquisition module, a data analysis module, a data modeling module, a multi-source heterogeneous information module, a prediction module, a switching module, and a data push module. The data acquisition module, data analysis module, data modeling module, multi-source heterogeneous information module, prediction module, switching module, and data push module are application software installed on a PC. An application method for the cross-cycle multi-source heterogeneous power data processing system includes seven steps. It can derive a power load data model for a power consumption area and perform power consumption prediction based on the obtained model, allowing for advance adjustment before changes in power consumption in the area. However, the above technical solution has the following problems: it does not consider the specific situation of the power elasticity coefficient of the area to be predicted, thus affecting the prediction efficiency of power consumption in the area. Summary of the Invention

[0006] To address this issue, the present invention provides a regional electricity consumption intelligent prediction method based on multi-source heterogeneous data, which overcomes the problem in the prior art that it does not take into account the specific circumstances of the power elasticity coefficient of the region to be predicted, and thus affects the prediction accuracy of electricity consumption in the region.

[0007] To achieve the above objectives, this invention provides a method for intelligent prediction of regional electricity consumption based on multi-source heterogeneous data, comprising:

[0008] Collect electricity consumption data for each historical preset monitoring period within the region to construct an electricity consumption dataset;

[0009] Determine the characteristic value of power change based on historical power elasticity coefficient;

[0010] Based on the power change characterization values, determine the predicted impact of electricity consumption on the next preset monitoring period;

[0011] When it is determined that the electricity consumption forecast has a strong influence on the next preset monitoring period, the electricity consumption is corrected by the elastic influence weighting coefficient, and the qualification of the electricity consumption forecast is determined based on the energy change characterization value.

[0012] When identifying anomalies in electricity consumption forecasts, the forecast parameters for electricity consumption in the next preset monitoring cycle are determined based on the quantitative value of regional industrial changes. This includes identifying the main analysis dataset and determining whether to increase the model training set or separate the forecast model based on the periodic quantitative value of the main analysis data.

[0013] If the next preset monitoring period is determined to be characterized by a weak impact of electricity consumption forecast, the electricity consumption within the next preset monitoring period is determined based on the electricity consumption dataset.

[0014] Furthermore, the process of determining the characteristic value of power change based on historical power elasticity coefficients includes:

[0015] Plot the time-domain curves of the electroelasticity coefficients based on historical data;

[0016] The absolute value of the ratio of the absolute value of the slope of the time-domain curve of the power elasticity coefficient at the current time node to the preset stable slope is obtained to obtain the power change characterization value.

[0017] Furthermore, the process of determining the predicted impact of electricity consumption on the next preset monitoring period based on the power change characterization values ​​includes:

[0018] When the power change characterization value is less than or equal to the preset power change characterization value, the next preset monitoring period is determined to be a weak influence tendency of power consumption prediction, and the power consumption in the next preset monitoring period is determined based on the power consumption dataset.

[0019] When the power change indicator value is greater than the preset power change indicator value, if the next preset monitoring period is determined to be when the power consumption forecast has a strong influence tendency, the power consumption will be corrected by the elastic influence weighting coefficient.

[0020] Furthermore, the process of correcting electricity consumption by influencing the weighting coefficient through elasticity includes:

[0021] The selection range of the elasticity coefficient is determined based on the current electric elasticity coefficient;

[0022] Each preset monitoring period in the historical data where the electric elasticity coefficient is within the selected range of the elasticity coefficient is determined as the reference period;

[0023] Obtain the impact ratio for each reference period;

[0024] The elastic force influence weight coefficient is determined based on each influence ratio.

[0025] Furthermore, the process of determining the representative values ​​for energy changes includes:

[0026] The ratio of natural gas consumption in the most recent preset monitoring period to the average natural gas consumption in each preset monitoring period in historical data is used to obtain the natural gas change rate.

[0027] The ratio of coal consumption in the most recent preset monitoring period to the average coal consumption in each preset monitoring period in historical data is used to obtain the coal change rate.

[0028] The ratio of oil consumption in the most recent preset monitoring period to the average oil consumption in each preset monitoring period in historical data is used to obtain the oil change rate.

[0029] The ratio of electricity consumption in the most recent preset monitoring period to the average electricity consumption in each preset monitoring period in historical data is obtained to obtain the electricity consumption change rate.

[0030] Calculate the average of the natural gas change rate, coal change rate, and oil change rate to obtain the change in alternative energy sources;

[0031] The energy change characterization value is obtained by solving the ratio of the absolute value of the difference between the change in alternative energy and the rate of change in electricity consumption to the rate of change in electricity consumption.

[0032] Furthermore, the adequacy of electricity consumption forecasts is determined based on energy change characterization values, including:

[0033] When the energy change characterization value is less than or equal to the preset energy change characterization value, the prediction of electricity consumption is determined to be qualified, and the product of the elasticity influence weighting coefficient and the electricity consumption predicted by the first prediction model is determined as the predicted electricity consumption for the next preset monitoring period.

[0034] When the energy change characterization value is greater than the preset energy change characterization value, an anomaly is identified in the prediction of electricity consumption. The prediction parameters for electricity consumption in the next preset monitoring cycle are then determined based on the quantitative value of regional industrial change.

[0035] Furthermore, the process of determining the quantitative value of regional industrial changes includes:

[0036] Obtain industrial electricity consumption and service electricity consumption within each historical preset monitoring period in the region, so as to obtain industrial electricity consumption datasets and service electricity consumption datasets respectively;

[0037] The ratio of industrial electricity consumption in the most recent preset monitoring period to the average industrial electricity consumption in each historical preset monitoring period is used to obtain the quantitative value of regional industrial changes.

[0038] The ratio of the service electricity consumption in the most recent preset monitoring period to the average service electricity consumption in each historical preset monitoring period is used to obtain the quantitative value of changes in the regional service industry.

[0039] Furthermore, the process of adjusting the predicted parameters for electricity consumption in the next preset monitoring period based on the quantitative value of regional industrial changes includes:

[0040] When the quantitative value of regional industrial sector change is less than or equal to the quantitative value of regional service sector change, the service electricity consumption dataset will be identified as the primary dataset for analysis.

[0041] When the quantitative value of regional industrial sector change is greater than the quantitative value of regional service sector change, the industrial electricity consumption dataset will be selected as the primary dataset for analysis.

[0042] Furthermore, the process of determining whether to increase the model training set or separate the prediction model based on the periodic quantification values ​​of the main analytical data includes:

[0043] Based on the electricity consumption of each preset monitoring period in the main analysis dataset, the main electricity consumption time-domain curves are plotted.

[0044] The signal of the main power consumption time domain curve is converted into a frequency domain signal by Fourier transform to obtain the spectrum.

[0045] Obtain the peak values ​​in the spectrum;

[0046] If a peak value exceeds the preset peak value, the model training set will be increased.

[0047] When all peak values ​​are less than or equal to the preset peak value, the prediction model is separated.

[0048] Furthermore, the process of increasing the model training set includes:

[0049] Based on the maximum peak value, the amount of data in the training set used to train the first prediction model is adjusted to a corresponding value, wherein,

[0050] Obtain each peak in the spectrum to find the maximum peak.

[0051] The increase in data volume is negatively correlated with the maximum peak value.

[0052] The process of separating the prediction model includes:

[0053] The predicted electricity consumption output by the model trained using the industrial electricity consumption dataset is determined as the predicted industrial electricity consumption.

[0054] The predicted electricity consumption output by the model trained using the service electricity consumption dataset is determined as the service predicted electricity consumption.

[0055] The sum of the industrial forecasted electricity consumption and the service forecasted electricity consumption is determined as the electricity consumption forecast for the next preset monitoring cycle.

[0056] Compared with existing technologies, the advantages of this invention lie in determining the predictive impact of electricity consumption on the next preset monitoring cycle based on the power change characterization value, which reflects the degree of change in the power elasticity coefficient. When the power change characterization value is less than or equal to the preset power change characterization value, the rate of change is similar to the steady state, and the relationship between the economy and electricity consumption is stable. When the power change characterization value is greater than the preset power change characterization value, it indicates drastic changes, and the relationship between the economy and electricity consumption fluctuates. This fluctuation affects the accuracy of the prediction, and the inherent laws governing the relationship between electricity consumption and the economy change rapidly. Using only a model trained based on past stable laws to determine the predicted value carries a high risk. Before directly using electricity consumption data for prediction, the elasticity coefficient is used to judge the actual situation of the prediction environment, providing a decision-making basis for determining whether to initiate a complex correction process. This effectively saves computing resources and further improves the prediction efficiency for electricity consumption.

[0057] Furthermore, when a strong influence is identified, periods with similar elasticity coefficients are identified from historical data as a reference. The elasticity influence weighting coefficient represents the correspondence between historical forecasts and actual values ​​under similar economic conditions, serving as an empirical correction factor. By applying the experience of prediction errors under similar historical scenarios to correct current predictions, the prediction results incorporate historical experience, improving both the accuracy and efficiency of electricity consumption prediction.

[0058] Furthermore, the feasibility of forecasts is verified based on energy change indicator values, which measure the consistency between changes in substitutable energy sources and changes in electricity consumption. When the energy change indicator value exceeds the preset value, electricity consumption behavior differs significantly from other energy sources, and energy substitution will affect electricity consumption forecasts. In this case, regional industrial change quantification values ​​are used to adjust the forecast parameters for electricity consumption in the next preset monitoring period. Electricity consumption forecasts are then validated a second time using multi-source data, forming a more comprehensive evaluation system. This system can more accurately detect risks stemming from changes in the internal energy structure that cannot be revealed by simple electricity consumption or economic data, further improving the efficiency of electricity consumption forecasting.

[0059] Furthermore, when the energy change indicator value exceeds the preset energy change indicator value, an anomaly in energy consumption trends indicates a change in industrial structure. The industrial change quantification value quantifies the extent of the shift in regional industrial focus between industry and services. When a region's industrial orientation changes, its electricity consumption trend will fluctuate. This anomaly can be identified for targeted analysis. Further, by using the industrial change quantification value, industries with larger changes are identified as the primary analysis dataset. Based on the periodic quantification value of the primary analysis data, it is determined whether to increase the model training set or separate the prediction model, specifically analyzing the impact of industrial structure changes on electricity consumption and improving prediction accuracy. Focusing on industries with significant changes better captures electricity consumption trends, further improving the efficiency of electricity consumption prediction.

[0060] Furthermore, when peak values ​​in the spectrum exceed the preset peak value, the periodic fluctuations in electricity consumption are relatively strong, exhibiting complex change patterns. In this case, increasing the model training set allows the model to learn more features and patterns, improving prediction accuracy. When all peak values ​​are less than or equal to the preset peak value, the periodic fluctuations in electricity consumption are relatively weak, and the electricity consumption characteristics of different industries are relatively independent. In this case, separate prediction models are used to make more accurate predictions for different industries. By flexibly adjusting the processing method according to the actual periodic fluctuations in electricity consumption, the model can maintain good performance under different conditions, enhancing its adaptability and further improving the efficiency of electricity consumption prediction.

[0061] Furthermore, based on the periodic quantification values ​​of the main analytical data, the decision is made to either increase the model training set or separate the prediction model. Two different strategies are adopted based on the periodic characteristics of the main industry electricity consumption data. The maximum peak value represents the intensity of the most significant periodic component in electricity consumption changes. When the maximum peak value is small, it indicates that the periodic fluctuations in electricity consumption are weak, and the model needs more data to learn this complex change pattern. Therefore, the increase in data volume is negatively correlated with the maximum peak value; that is, the larger the maximum peak value, the less data needs to be added, because the model can already capture the main change characteristics from the existing data; the smaller the maximum peak value, the more data needs to be added to help the model learn more detailed features. By adjusting the amount of training set data according to the maximum peak value, while ensuring the model's learning effect, the problem of excessively long training time and overfitting caused by excessive data volume is avoided, further improving the prediction efficiency for electricity consumption.

[0062] Furthermore, the electricity consumption characteristics of industry and service sectors differ, and their electricity consumption fluctuations are influenced by different factors. When the periodic fluctuations in electricity consumption are relatively weak, training models separately on industrial and service electricity consumption datasets can more accurately capture the electricity consumption patterns of different industries. Finally, the prediction results of the two models are added together to obtain the predicted electricity consumption for the next preset monitoring period. By fully considering the electricity consumption characteristics of different industries and avoiding mutual interference between their electricity consumption features, the accuracy of predictions is improved. By separating the prediction models to model and predict the electricity consumption characteristics of industry and service sectors separately, the electricity consumption patterns of different industries are captured more accurately, providing a more accurate basis for power supply and management, and further improving the efficiency of electricity consumption prediction. Attached Figure Description

[0063] Figure 1 This is a flowchart illustrating the steps of the intelligent regional electricity consumption prediction method based on multi-source heterogeneous data according to an embodiment of the present invention.

[0064] Figure 2 This is a logic diagram for determining the predictive impact of electricity consumption on the next preset monitoring period based on power change characterization values ​​in an embodiment of the present invention.

[0065] Figure 3 This is a logic diagram illustrating the determination of whether a forecast for electricity consumption is qualified based on energy change characterization values ​​in an embodiment of the present invention.

[0066] Figure 4 This is a logic decision diagram for determining whether to add a model training set or separate the prediction model based on the periodic quantization value of the main analysis data in an embodiment of the present invention. Detailed Implementation

[0067] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0068] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0069] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0070] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0071] Please see Figure 1 The diagram shown is a flowchart illustrating the steps of a regional electricity consumption intelligent prediction method based on multi-source heterogeneous data according to an embodiment of the present invention. The method of the present invention includes:

[0072] S1, collect electricity consumption data for each historical preset monitoring period within the region to construct an electricity consumption dataset;

[0073] S2, Determine the power change characterization value based on historical power elasticity coefficient;

[0074] S3, Based on the power change characterization value, determine the predicted impact tendency of electricity consumption for the next preset monitoring period;

[0075] S4, when it is determined that the next preset monitoring period is likely to have a strong influence on electricity consumption forecast, the electricity consumption is corrected by elastic influence weighting coefficient, and the forecast of electricity consumption is determined to be qualified based on the energy change characterization value.

[0076] When identifying anomalies in electricity consumption forecasts, the forecast parameters for electricity consumption in the next preset monitoring cycle are determined based on the quantitative value of regional industrial changes. This includes identifying the main analysis dataset and determining whether to increase the model training set or separate the forecast model based on the periodic quantitative value of the main analysis data.

[0077] If the next preset monitoring period is determined to be characterized by a weak impact of electricity consumption forecast, the electricity consumption within the next preset monitoring period is determined based on the electricity consumption dataset.

[0078] Specifically, the process of determining the characteristic value of power change based on historical power elasticity coefficients includes:

[0079] Plot the time-domain curves of the electroelasticity coefficients based on historical data;

[0080] The absolute value of the ratio of the absolute value of the slope of the time-domain curve of the power elasticity coefficient at the current time node to the preset stable slope is obtained to obtain the power change characterization value.

[0081] Specifically, the process of determining the preset stable slope includes:

[0082] The absolute value of the slope of the preset monitoring period corresponding to the prediction deviation in historical data that is less than or equal to the preset prediction deviation is marked as the stable slope.

[0083] The average value of each stable slope is calculated to obtain the preset stable slope;

[0084] Calculate the absolute value of the difference between the electricity consumption predicted by the first prediction model and the actual electricity consumption, and solve for the ratio of the absolute value of the difference to the actual electricity consumption to obtain the prediction deviation.

[0085] The preset prediction deviation is selected within the range [0.04, 0.08]. Those skilled in the art can determine the preset prediction deviation according to the actual application scenario. It is understood that it can be achieved by dividing the accuracy of the prediction of the first prediction model. In this embodiment, preferably, the preset prediction deviation is 0.06.

[0086] Specifically, the electricity elasticity coefficient within a single preset monitoring period is the ratio of the average annual growth rate of electricity consumption to the average annual growth rate of the national economy.

[0087] Please see Figure 2 As shown, this is a logic diagram illustrating the determination of the predictive impact of electricity consumption on the next preset monitoring period based on power change characterization values, according to an embodiment of the present invention. The process of determining the predictive impact of electricity consumption on the next preset monitoring period based on power change characterization values ​​includes:

[0088] When the power change characterization value is less than or equal to the preset power change characterization value, the next preset monitoring period is determined to be a weak influence tendency of power consumption prediction, and the power consumption in the next preset monitoring period is determined based on the power consumption dataset.

[0089] When the power change indicator value is greater than the preset power change indicator value, the next preset monitoring period is determined to be a strong influence trend of power consumption prediction, and the power consumption is corrected by the elastic influence weighting coefficient.

[0090] Specifically, the preset power change characterization value is selected within the range [1.07, 1.13]. Those skilled in the art can select and determine the preset power change characterization value according to the actual situation. Historical data can be statistically analyzed to determine the prediction deviation when the power change characterization value exceeds different thresholds, and the threshold that significantly increases the prediction deviation can be selected as the preset power change characterization value. In this embodiment, preferably, the preset power change characterization value is 1.07.

[0091] Specifically, the impact of power consumption changes on the next preset monitoring cycle is determined based on the power change characteristic value, which reflects the degree of change in the power elasticity coefficient. When the power change characteristic value is less than or equal to the preset power change characteristic value, the rate of change is similar to the steady state, and the relationship between the economy and electricity consumption is stable. When the power change characteristic value is greater than the preset power change characteristic value, it indicates drastic changes, and the relationship between the economy and electricity consumption fluctuates. This fluctuation will affect the accuracy of the prediction, and the inherent laws governing the relationship between electricity consumption and the economy are changing rapidly. Using only a model trained based on past stable laws to determine the prediction value is risky. Before directly using electricity consumption data for prediction, the actual situation of the prediction environment is judged by the elasticity coefficient, providing a decision-making basis for determining whether to initiate a complex correction process. This effectively saves computing resources and further improves the efficiency of electricity consumption prediction.

[0092] Specifically, the first prediction model is a time series prediction model LSTM. The specific process for training the first prediction model is not limited and may include:

[0093] Perform data preprocessing on the electricity consumption dataset, including cleaning, handling missing values ​​and outliers;

[0094] Set the number of network layers, the number of neurons, and the learning rate;

[0095] Historical electricity consumption data is divided into training and test sets in chronological order. The training set is used to fit the model, and the test set is used to evaluate the performance. This is an existing technology and will not be elaborated further.

[0096] In a single embodiment, the network has 2 layers, 50 neurons per layer, and a learning rate of 0.001.

[0097] Specifically, the process of determining the electricity consumption in the next preset monitoring period based on the electricity consumption dataset includes determining the electricity consumption in the next preset monitoring period through a first prediction model.

[0098] Specifically, the process of correcting electricity consumption by influencing the weighting coefficient through elasticity includes:

[0099] The selection range of the elasticity coefficient is determined based on the current electric elasticity coefficient;

[0100] Each preset monitoring period in the historical data where the electric elasticity coefficient is within the selected range of the elasticity coefficient is determined as the reference period;

[0101] Obtain the impact ratio for each reference period;

[0102] The elastic force influence weight coefficient is determined based on each influence ratio.

[0103] Specifically, D0±△D is defined as the selection range of the elasticity coefficient, where D0 is the current electric elasticity coefficient and △D is the preset tolerance.

[0104] Specifically, the impact factor for a single reference period is the ratio of the actual electricity consumption during that period to the electricity consumption predicted by the first prediction model.

[0105] Specifically, the process of determining the elasticity influence weight coefficient based on each influence ratio includes determining the elasticity influence weight coefficient based on the median of each influence ratio.

[0106] Specifically, the preset tolerance is selected within the range [0.09T0, 0.14T0], where T0 is the average value of each historical power elasticity coefficient. Those skilled in the art can select and determine the preset tolerance based on the dispersion of historical data. The greater the fluctuation of each historical power elasticity coefficient, the greater the preset tolerance. In this embodiment, preferably, the preset tolerance is 0.1T0.

[0107] Specifically, when a strong influence is identified, periods with similar elasticity coefficients are identified from historical data as a reference. The elasticity influence weighting coefficient represents the correspondence between historical forecasts and actual values ​​under similar economic conditions, serving as an empirical correction factor. By applying the experience of prediction errors under similar historical scenarios to correct current predictions, the prediction results incorporate historical experience, improving both the accuracy and efficiency of electricity consumption prediction.

[0108] Specifically, the process of determining the values ​​representing energy changes includes:

[0109] The ratio of natural gas consumption in the most recent preset monitoring period to the average natural gas consumption in each preset monitoring period in historical data is used to obtain the natural gas change rate.

[0110] The ratio of coal consumption in the most recent preset monitoring period to the average coal consumption in each preset monitoring period in historical data is used to obtain the coal change rate.

[0111] The ratio of oil consumption in the most recent preset monitoring period to the average oil consumption in each preset monitoring period in historical data is used to obtain the oil change rate.

[0112] The ratio of electricity consumption in the most recent preset monitoring period to the average electricity consumption in each preset monitoring period in historical data is obtained to obtain the electricity consumption change rate.

[0113] Calculate the average of the natural gas change rate, coal change rate, and oil change rate to obtain the change in alternative energy sources;

[0114] The energy change characterization value is obtained by solving the ratio of the absolute value of the difference between the change in alternative energy and the rate of change in electricity consumption to the rate of change in electricity consumption.

[0115] Specifically, the unit for natural gas consumption is 10,000 cubic meters, the unit for coal consumption is 10,000 tons, and the unit for oil consumption is 10,000 tons.

[0116] Please see Figure 3 The diagram shown illustrates the logic for determining the validity of electricity consumption forecasts based on energy change characterization values, as per an embodiment of the present invention. The process of determining the validity of electricity consumption forecasts based on energy change characterization values ​​includes:

[0117] When the energy change characterization value is less than or equal to the preset energy change characterization value, the prediction of electricity consumption is determined to be qualified, and the product of the elasticity influence weighting coefficient and the electricity consumption predicted by the first prediction model is determined as the predicted electricity consumption for the next preset monitoring period.

[0118] When the energy change characterization value is greater than the preset energy change characterization value, an anomaly is identified in the prediction of electricity consumption. The prediction parameters for electricity consumption in the next preset monitoring cycle are then determined based on the quantitative value of regional industrial change.

[0119] Specifically, the preset energy change characterization value is selected within the range of [0.05, 0.15]. Those skilled in the art can select the preset energy change characterization value themselves, which can be determined by analyzing historical data. It is understood that the situation of whether there is a significant change in the energy structure can be divided. In this embodiment, preferably, the preset energy change characterization value is 0.1.

[0120] Specifically, the feasibility of forecasts is verified based on energy change indicator values, which measure the consistency between changes in substitutable energy sources and changes in electricity consumption. When the energy change indicator value exceeds the preset value, electricity consumption behavior differs significantly from other energy sources, and energy substitution will affect electricity consumption forecasts. In this case, regional industrial change quantification values ​​are used to adjust the forecast parameters for electricity consumption in the next preset monitoring period. Electricity consumption forecasts are then validated a second time using multi-source data, forming a more comprehensive evaluation system. This system can more accurately detect risks stemming from changes in the internal energy structure that cannot be revealed by simple electricity consumption or economic data, further improving the efficiency of electricity consumption forecasting.

[0121] Specifically, the process of determining the quantitative value of regional industrial changes includes:

[0122] Obtain industrial electricity consumption and service electricity consumption within each historical preset monitoring period in the region, so as to obtain industrial electricity consumption datasets and service electricity consumption datasets respectively;

[0123] The ratio of industrial electricity consumption in the most recent preset monitoring period to the average industrial electricity consumption in each historical preset monitoring period is used to obtain the quantitative value of regional industrial changes.

[0124] The ratio of the service electricity consumption in the most recent preset monitoring period to the average service electricity consumption in each historical preset monitoring period is used to obtain the quantitative value of changes in the regional service industry.

[0125] Specifically, the process of determining the predicted parameters for electricity consumption in the next preset monitoring period based on the quantitative value of regional industrial changes includes:

[0126] When the quantitative value of regional industrial sector change is less than or equal to the quantitative value of regional service sector change, the service electricity consumption dataset will be identified as the primary dataset for analysis.

[0127] When the quantitative value of regional industrial sector change is greater than the quantitative value of regional service sector change, the industrial electricity consumption dataset will be selected as the primary dataset for analysis.

[0128] Please see Figure 4 As shown, this is a logic decision diagram for determining whether to increase the model training set or separate the prediction model based on the periodic quantization value of the main analysis data in an embodiment of the present invention. The process of determining whether to increase the model training set or separate the prediction model based on the periodic quantization value of the main analysis data in the present invention includes:

[0129] Based on the electricity consumption of each preset monitoring period in the main analysis dataset, the main electricity consumption time-domain curves are plotted.

[0130] The signal of the main power consumption time domain curve is converted into a frequency domain signal by Fourier transform to obtain the spectrum.

[0131] Obtain the peak values ​​in the spectrum;

[0132] If a peak value exceeds the preset peak value, the model training set will be increased.

[0133] When all peak values ​​are less than or equal to the preset peak value, the prediction model is separated.

[0134] In a single embodiment, the preset peak value is twice the average value of the spectrum.

[0135] Specifically, when the energy change indicator value exceeds the preset energy change indicator value, an anomaly in energy consumption trends is observed, indicating a change in the industrial structure. The industrial change quantification value quantifies the extent of the shift in the regional industrial focus between industry and services. When a region's industrial orientation changes, its electricity consumption trends will fluctuate. This anomaly is identified for targeted analysis. Furthermore, the industries with the largest changes are identified as the primary analysis dataset based on the industrial change quantification value. The periodic quantification value of the primary analysis data determines whether to increase the model training set or separate the prediction model, allowing for targeted analysis of the impact of industrial structure changes on electricity consumption and improving prediction accuracy. Focusing on industries with significant changes better captures electricity consumption trends, further improving the efficiency of electricity consumption prediction.

[0136] Specifically, when peak values ​​in the power consumption spectrum exceed the preset peak value, the periodic fluctuations in power consumption are relatively strong, exhibiting complex change patterns. In this case, increasing the model training set allows the model to learn more features and patterns, improving prediction accuracy. When all peak values ​​are less than or equal to the preset peak value, the periodic fluctuations in power consumption are relatively weak, and the power consumption characteristics of different industries are relatively independent. In this case, separate prediction models are used to make more accurate predictions for different industries. Flexibly adjusting the processing method based on the actual periodic fluctuations in power consumption ensures that the model maintains good performance under different conditions, enhancing its adaptability and further improving the efficiency of power consumption prediction.

[0137] Specifically, the process of increasing the model training set includes:

[0138] Based on the maximum peak value, the amount of data in the training set used to train the first prediction model is adjusted to a corresponding value, wherein,

[0139] Obtain each peak in the spectrum to find the maximum peak.

[0140] The increase in data volume is negatively correlated with the maximum peak value.

[0141] The process of separating the prediction model includes:

[0142] The predicted electricity consumption output by the model trained using the industrial electricity consumption dataset is determined as the predicted industrial electricity consumption.

[0143] The predicted electricity consumption output by the model trained using the service electricity consumption dataset is determined as the service predicted electricity consumption.

[0144] The sum of the industrial forecasted electricity consumption and the service forecasted electricity consumption is determined as the electricity consumption forecast for the next preset monitoring cycle.

[0145] Specifically, based on the periodic quantification values ​​of the main analytical data, the decision is made to either increase the model training set or separate the prediction model. Two different strategies are adopted based on the periodic characteristics of electricity consumption data from major industries. The maximum peak value represents the intensity of the most significant periodic component in electricity consumption changes. When the maximum peak value is small, it indicates weaker periodic fluctuations in electricity consumption, and the model needs more data to learn this complex change pattern. Therefore, the increase in data volume is negatively correlated with the maximum peak value; that is, the larger the maximum peak value, the less data is added, because the model can already capture the main change characteristics from the existing data; conversely, the smaller the maximum peak value, the more data is added to help the model learn more detailed features. By adjusting the amount of training set data according to the maximum peak value, the model's learning effect is ensured while avoiding excessively long training times and overfitting problems caused by excessive data volume, further improving the prediction efficiency for electricity consumption.

[0146] In this embodiment, optionally,

[0147] The peak value is compared with the first preset peak value and the second preset peak value;

[0148] If the peak value is less than or equal to the first preset peak value, the amount of data in the training set used to train the first prediction model will be adjusted to 1.31 times the current amount of data.

[0149] If the peak value is less than or equal to the second preset peak value and greater than the first preset peak value, the amount of data in the training set used to train the first prediction model will be adjusted to 1.21 times the current amount of data.

[0150] If the peak value is greater than the second preset peak value, the amount of data in the training set used to train the first prediction model will be adjusted to 1.11 times the current amount of data.

[0151] The first preset peak value is 3.2J0, and the second preset peak value is 3.8J0, where J0 is the average value of the spectrum.

[0152] Specifically, the electricity consumption characteristics of industry and service sectors differ, and their electricity consumption fluctuations are influenced by different factors. When the periodic fluctuations in electricity consumption are relatively weak, training models separately on industrial and service electricity consumption datasets can more accurately capture the electricity consumption patterns of different industries. Finally, the prediction results of the two models are added together to obtain the predicted electricity consumption for the next preset monitoring period. By fully considering the electricity consumption characteristics of different industries and avoiding mutual interference between their electricity consumption features, the accuracy of predictions is improved. By separating the prediction models to model and predict the electricity consumption characteristics of industry and service sectors separately, the electricity consumption patterns of different industries are captured more accurately, providing a more accurate basis for power supply and management, and further improving the efficiency of electricity consumption prediction.

[0153] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0154] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for intelligent prediction of regional electricity consumption based on multi-source heterogeneous data, characterized in that, include: Collect electricity consumption data for each historical preset monitoring period within the region to construct an electricity consumption dataset; Determine the characteristic value of power change based on historical power elasticity coefficient; Based on the power change characterization values, determine the predicted impact of electricity consumption on the next preset monitoring period; When it is determined that the electricity consumption forecast has a strong influence on the next preset monitoring period, the electricity consumption is corrected by the elastic influence weighting coefficient, and the qualification of the electricity consumption forecast is determined based on the energy change characterization value. When identifying anomalies in electricity consumption forecasts, the forecast parameters for electricity consumption in the next preset monitoring cycle are determined based on the quantitative value of regional industrial changes. This includes identifying the main analysis dataset and determining whether to increase the model training set or separate the forecast model based on the periodic quantitative value of the main analysis data. Based on the determination that the next preset monitoring period is likely to have a weak impact from electricity consumption forecasts, the electricity consumption within the next preset monitoring period is determined using the electricity consumption dataset. The process of determining the characteristic value of power change based on historical power elasticity coefficients includes: Plot the time-domain curves of the electroelasticity coefficients based on historical data; The absolute value of the ratio of the absolute value of the slope of the time-domain curve of the power elasticity coefficient at the current time node to the preset stable slope is calculated to obtain the power change characterization value; The process of correcting electricity consumption by influencing the weighting coefficient through elasticity includes: The selection range of the elasticity coefficient is determined based on the current electric elasticity coefficient; Each preset monitoring period in the historical data where the electric elasticity coefficient is within the selected range of the elasticity coefficient is determined as the reference period; Obtain the impact ratio for each reference period; The weighting coefficient of elastic force influence is determined based on each influence factor. The process of determining the values ​​representing energy changes includes: The ratio of natural gas consumption in the most recent preset monitoring period to the average natural gas consumption in each preset monitoring period in historical data is used to obtain the natural gas change rate. The ratio of coal consumption in the most recent preset monitoring period to the average coal consumption in each preset monitoring period in historical data is used to obtain the coal change rate. The ratio of oil consumption in the most recent preset monitoring period to the average oil consumption in each preset monitoring period in historical data is used to obtain the oil change rate. The ratio of electricity consumption in the most recent preset monitoring period to the average electricity consumption in each preset monitoring period in historical data is obtained to obtain the electricity consumption change rate. Calculate the average of the natural gas change rate, coal change rate, and oil change rate to obtain the change in alternative energy sources; The ratio of the absolute value of the difference between the change in alternative energy and the rate of change in electricity consumption to the rate of change in electricity consumption is used to obtain the energy change characterization value; The process of determining the quantitative value of regional industrial changes includes: Obtain industrial electricity consumption and service electricity consumption within each historical preset monitoring period in the region, so as to obtain industrial electricity consumption datasets and service electricity consumption datasets respectively; The ratio of industrial electricity consumption in the most recent preset monitoring period to the average industrial electricity consumption in each historical preset monitoring period is used to obtain the quantitative value of regional industrial changes. The ratio of service electricity consumption in the most recent preset monitoring period to the average service electricity consumption in each historical preset monitoring period is used to obtain a quantitative value of changes in the regional service industry. The process of determining the predicted parameters for electricity consumption in the next preset monitoring period based on the quantitative value of regional industrial changes includes: When the quantitative value of regional industrial sector change is less than or equal to the quantitative value of regional service sector change, the service electricity consumption dataset will be identified as the primary dataset for analysis. When the quantitative value of regional industrial sector change is greater than the quantitative value of regional service sector change, the industrial electricity consumption dataset will be selected as the primary dataset for analysis. The process of determining whether to increase the model training set or separate the prediction model based on the periodic quantization values ​​of the main analysis data includes: Based on the electricity consumption of each preset monitoring period in the main analysis dataset, the main electricity consumption time-domain curves are plotted. The signal of the main power consumption time domain curve is converted into a frequency domain signal by Fourier transform to obtain the spectrum. Obtain the peak values ​​in the spectrum; If a peak value exceeds the preset peak value, the model training set will be increased. When all peak values ​​are less than or equal to the preset peak value, the prediction model is separated. The process of increasing the model training set includes: Based on the maximum peak value, the amount of data in the training set used to train the first prediction model is adjusted to a corresponding value, wherein, Obtain each peak in the spectrum to find the maximum peak. The increase in data volume is negatively correlated with the maximum peak value. The process of separating the prediction model includes: The predicted electricity consumption output by the model trained using the industrial electricity consumption dataset is determined as the predicted industrial electricity consumption. The predicted electricity consumption output by the model trained using the service electricity consumption dataset is determined as the service predicted electricity consumption. The sum of the industrial forecasted electricity consumption and the service forecasted electricity consumption is determined as the electricity consumption forecast for the next preset monitoring cycle.

2. The intelligent prediction method for regional electricity consumption based on multi-source heterogeneous data according to claim 1, characterized in that, The process of determining the predicted impact of electricity consumption on the next preset monitoring period based on electricity change characterization values ​​includes: When the power change characterization value is less than or equal to the preset power change characterization value, the next preset monitoring period is determined to be a weak influence tendency of power consumption prediction, and the power consumption in the next preset monitoring period is determined based on the power consumption dataset. When the power change indicator value is greater than the preset power change indicator value, if the next preset monitoring period is determined to be when the power consumption forecast has a strong influence tendency, the power consumption will be corrected by the elastic influence weighting coefficient.

3. The intelligent prediction method for regional electricity consumption based on multi-source heterogeneous data according to claim 2, characterized in that, Determining the adequacy of electricity consumption forecasts based on energy change characterization values ​​includes: When the energy change characterization value is less than or equal to the preset energy change characterization value, the prediction of electricity consumption is determined to be qualified, and the product of the elasticity influence weighting coefficient and the electricity consumption predicted by the first prediction model is determined as the predicted electricity consumption for the next preset monitoring period. When the energy change characterization value is greater than the preset energy change characterization value, an anomaly is identified in the prediction of electricity consumption. The prediction parameters for electricity consumption in the next preset monitoring cycle are then determined based on the quantitative value of regional industrial change.

Images