Electricity consumption prediction method, device, equipment, readable storage medium and program product

Abstract

The application relates to a power consumption prediction method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: determining a power consumption sequence based on historical power consumption data, and determining initial predicted power consumption information of each industry in each region in a preset future time period based on the power consumption sequence, wherein the power consumption sequence is used to indicate power consumption information of each industry in each region in a historical preset time period; obtaining weather information and historical prediction error information in the preset future time period, and determining initial correction information based on the weather information and the historical prediction error information; performing adjustment processing on the initial correction information to obtain target correction information, and performing correction processing on the initial predicted power consumption information based on the target correction information to obtain target predicted power consumption information of each industry in each region in the preset future time period. The method can improve the accuracy of power consumption prediction.

Description

Technical Field

[0001] This application relates to the field of power technology, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for predicting electricity consumption. Background Technology

[0002] With the large-scale development of the power system and the continuous growth of social electricity demand, regional electricity load fluctuations are frequent and industry-specific electricity consumption varies significantly, making the regulation and control of electricity supply and demand increasingly difficult. Electricity consumption forecasting allows for advance understanding of future electricity load changes, assisting the power grid in rationally arranging dispatch plans and allocating power resources, thereby ensuring the safe and stable operation of the power grid and a dynamic balance between electricity supply and demand.

[0003] In existing technologies, most electricity consumption forecasting methods employ statistical models. However, these methods suffer from poor accuracy. Summary of the Invention

[0004] Therefore, it is necessary to provide a highly accurate electricity consumption forecasting method, device, computer equipment, computer-readable storage medium, and computer program product to address the aforementioned technical problems.

[0005] In a first aspect, this application provides a method for predicting electricity consumption, including:

[0006] Based on historical electricity consumption data, an electricity consumption sequence is determined, and based on the electricity consumption sequence, the initial predicted electricity consumption information for each industry in each region within a future preset time period is determined. The electricity consumption sequence is used to indicate the electricity consumption information of each industry in each region within a historical preset time period.

[0007] Obtain weather information and historical forecast error information for a future preset time period, and determine initial correction information based on the weather information and historical forecast error information;

[0008] The initial correction information is adjusted to obtain the target correction information, and the initial predicted electricity consumption information is corrected based on the target correction information to obtain the target predicted electricity consumption information for each industry in each region within a preset time period in the future.

[0009] In one embodiment, determining initial correction information based on weather information and historical forecast error information includes: determining a weather correction component based on weather information, determining a residual correction component based on historical forecast error information, and determining a trend correction component based on historical electricity consumption data; and determining initial correction information based on the weather correction component, residual correction component, and trend correction component.

[0010] In one embodiment, determining a weather correction component based on weather information includes: determining cooling load pressure, heating load pressure, and target temperature days based on weather information, and determining a temperature correction component based on the cooling load pressure, heating load pressure, and target temperature days; identifying extreme climate events in the weather information, and determining an extreme climate correction component based on the identification results; and determining a weather correction component based on the temperature correction component and the extreme climate correction component.

[0011] In one embodiment, the initial correction information is adjusted to obtain target correction information, including: determining the climate scene type corresponding to a future preset time period based on weather information, and determining an adjustment strategy based on the climate scene type; and adjusting the initial correction information based on the adjustment strategy to obtain target correction information.

[0012] In one embodiment, the initial predicted electricity consumption information is corrected based on the target correction information to obtain the target predicted electricity consumption information for each industry in each region within a future preset time period. This includes: determining the weight values ​​of each industry in each region based on the initial predicted electricity consumption information and industry sensitivity; and using the weight values, correcting the initial predicted electricity consumption information based on the target correction information to obtain the target predicted electricity consumption information.

[0013] In one embodiment, determining the initial predicted electricity consumption information for each industry in each region within a preset time period based on the electricity consumption sequence includes: inputting the electricity consumption sequence into a pre-trained prediction model to obtain the initial predicted electricity consumption information output by the prediction model.

[0014] Secondly, this application also provides an electricity consumption forecasting device, comprising:

[0015] The determination module is used to determine the electricity consumption sequence based on historical electricity consumption data, and to determine the initial predicted electricity consumption information of each industry in each region within a future preset time period based on the electricity consumption sequence. The electricity consumption sequence is used to indicate the electricity consumption information of each industry in each region within a historical preset time period.

[0016] The acquisition module is used to acquire weather information and historical prediction error information for a future preset time period, and to determine initial correction information based on the weather information and historical prediction error information.

[0017] The execution module is used to adjust the initial correction information to obtain the target correction information, and to correct the initial predicted electricity consumption information based on the target correction information to obtain the target predicted electricity consumption information for each industry in each region within a preset time period in the future.

[0018] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in any of the embodiments of the first aspect above.

[0019] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any of the embodiments of the first aspect above.

[0020] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in any of the embodiments of the first aspect above.

[0021] The aforementioned electricity consumption forecasting method, apparatus, computer equipment, computer-readable storage medium, and computer program product first determine an electricity consumption sequence based on historical electricity consumption data, and then determine initial forecasted electricity consumption information for each industry in each region within a preset future time period based on the electricity consumption sequence. The electricity consumption sequence indicates the electricity consumption information of each industry in each region within the preset historical time period. Next, weather information and historical forecast error information for the preset future time period are acquired, and initial correction information is determined based on the weather information and historical forecast error information. Finally, the initial correction information is adjusted to obtain target correction information, and the initial forecasted electricity consumption information is corrected based on the target correction information to obtain the target forecasted electricity consumption information for each industry in each region within the preset future time period. The electricity consumption forecasting method provided in this application, after obtaining the initial forecasted electricity consumption information for each industry in each region within the preset future time period, further corrects the initial forecasted electricity consumption information based on weather information and historical forecast error information within the preset future time period. This ensures that the final target forecasted electricity consumption information takes into account the influence of weather and historical forecast deviations, thereby effectively improving the accuracy of the forecasted electricity consumption information. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart illustrating an electricity consumption prediction method in one embodiment;

[0024] Figure 2 This is a flowchart illustrating a method for determining initial correction information in one embodiment;

[0025] Figure 3 This is a flowchart illustrating a method for determining weather correction components in one embodiment;

[0026] Figure 4 This is a flowchart illustrating a method for obtaining target correction information in one embodiment;

[0027] Figure 5 This is a flowchart illustrating a method for obtaining target predicted electricity consumption information for various industries in different regions within a preset time period, as shown in one embodiment.

[0028] Figure 6 This is a flowchart illustrating the electricity consumption prediction method in another embodiment;

[0029] Figure 7 This is a structural block diagram of an electricity consumption prediction device in one embodiment;

[0030] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0032] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0033] With the large-scale development of the power system and the continuous growth of social electricity demand, regional electricity load fluctuations are frequent and industry-specific electricity consumption varies significantly, making the regulation and control of electricity supply and demand increasingly difficult. Electricity consumption forecasting allows for advance understanding of future electricity load changes, assisting the power grid in rationally arranging dispatch plans and allocating power resources, thereby ensuring the safe and stable operation of the power grid and a dynamic balance between electricity supply and demand.

[0034] Traditional methods often employ statistical models, machine learning models, or rule-based experience to predict electricity consumption. However, under conditions of extreme temperatures, holiday shifts, changes in industrial structure, the commissioning of major projects, and the growth of charging and battery swapping services, single models are easily affected by historical inertia and struggle to reflect future changes in weather and operational conditions in a timely manner. In recent years, extreme weather events such as extreme high temperatures, sustained low temperatures, heavy rainfall, typhoons, cold waves, and droughts have increasingly disrupted electricity load and monthly electricity consumption. Especially during periods of high summer temperatures and low winter temperatures, electricity consumption in sectors such as residential life, commercial services, data centers, and charging and battery swapping services is affected by changes in air conditioning load, heating load, and production activities, resulting in significant nonlinear fluctuations in monthly electricity consumption and leading to poor accuracy in electricity consumption forecasting methods.

[0035] In view of this, this application provides an electricity consumption forecasting method. First, an electricity consumption sequence is determined based on historical electricity consumption data. Then, based on the electricity consumption sequence, initial forecast electricity consumption information for each industry in each region within a preset future time period is determined. The electricity consumption sequence indicates the electricity consumption information of each industry in each region within the historical preset time period. Next, weather information and historical forecast error information for the preset future time period are obtained, and initial correction information is determined based on the weather information and historical forecast error information. Finally, the initial correction information is adjusted to obtain target correction information, and the initial forecast electricity consumption information is corrected based on the target correction information to obtain the target forecast electricity consumption information for each industry in each region within the preset future time period. The electricity consumption forecasting method provided in this application, after obtaining the initial forecast electricity consumption information for each industry in each region within the preset future time period, also corrects the initial forecast electricity consumption information based on the weather information and historical forecast error information within the preset future time period. This ensures that the final target forecast electricity consumption information takes into account the influence of weather and historical forecast deviations, thereby effectively improving the accuracy of the forecast electricity consumption information.

[0036] The electricity forecasting method provided in this application can be deployed on the power grid's electricity forecasting platform, energy data middleware, grid-level power energy data platform, or load forecasting software system. Its execution entity can be a server, cloud computing platform, edge computing server, or electronic device with a processor and memory.

[0037] In one exemplary embodiment, such as Figure 1 As shown, an electricity consumption forecasting method is provided, which includes the following steps:

[0038] Step 101: Determine the electricity consumption sequence based on historical electricity consumption data, and determine the initial predicted electricity consumption information for each industry in each region within a preset time period based on the electricity consumption sequence.

[0039] Optionally, historical electricity consumption data may include daily and monthly actual electricity consumption data for each province, city, and industry within a historical time period. Electricity consumption sequences can be used to indicate electricity consumption information for each industry in each region within a preset historical time period, and can also be used to indicate the monthly electricity consumption changes for each industry in each region within a historical time period.

[0040] In some exemplary embodiments, the computer device may first acquire historical electricity consumption data and determine the electricity consumption sequence based on the historical electricity consumption data.

[0041] Specifically, computer equipment can aggregate daily actual electricity consumption data from historical electricity consumption data by province, city, industry, and month to obtain electricity consumption sequences. For example, the computer equipment can first read the daily actual electricity consumption data and perform data deduplication, missing value filling, and abnormal electricity consumption removal on the daily actual electricity consumption data. Then, it can perform monthly aggregation calculations on the daily actual electricity consumption data according to multiple dimensions such as province, city, industry, and month to unify the data statistical time caliber. Optionally, after obtaining the electricity consumption sequence, the computer equipment can also verify the data completeness of each natural month sequence by sequence, and remove invalid time series with time series breaks or missing data ratios exceeding the threshold.

[0042] Furthermore, after determining the electricity consumption sequence, the computer equipment can determine the initial predicted electricity consumption information for each industry in each region within a preset time period based on the electricity consumption sequence.

[0043] Specifically, computer equipment can use statistical models to determine the initial predicted electricity consumption information for each industry in each region within a preset time period based on electricity consumption sequences; computer equipment can also use machine learning models to determine the initial predicted electricity consumption information for each industry in each region within a preset time period based on electricity consumption sequences; computer equipment can also use rule experience to determine the initial predicted electricity consumption information for each industry in each region within a preset time period based on electricity consumption sequences.

[0044] Step 102: Obtain weather information and historical prediction error information for the future preset time period, and determine the initial correction information based on the weather information and historical prediction error information.

[0045] Optionally, the future preset time period can be determined based on actual needs, such as one month. Weather information can be used to indicate the weather conditions within the future preset time period. For example, if the future preset time period is one month, the weather information can include daily forecast temperature, humidity, and rainfall data for that month. Historical prediction error information, also known as the historical monthly electricity consumption prediction residual dataset, contains the deviation ratio between historical electricity consumption prediction values ​​and actual electricity consumption values ​​for each region and industry, and can be used to reflect the long-term systematic prediction bias of the model.

[0046] In some exemplary embodiments, the computer device may first obtain weather information and historical prediction error information for a future preset time period, and then determine initial correction information based on the weather information and historical prediction error information.

[0047] Specifically, the computer equipment can determine the first correction information based on weather information, determine the second correction information based on historical forecast error information, and determine the initial correction information based on the first and second correction information.

[0048] Step 103: Adjust the initial correction information to obtain target correction information, and correct the initial predicted electricity consumption information based on the target correction information to obtain the target predicted electricity consumption information for each industry in each region within a preset time period in the future.

[0049] In some exemplary embodiments, after obtaining initial correction information, the computer device may adjust the initial correction information to obtain target correction information.

[0050] Specifically, computer equipment can adjust the initial correction information according to preset adjustment rules to obtain the target correction information.

[0051] Furthermore, after obtaining the target correction information, the computer equipment can correct the initial predicted electricity consumption information based on the target correction information to obtain the target predicted electricity consumption information for each industry in each region within a preset time period in the future.

[0052] The aforementioned electricity consumption forecasting method first determines the electricity consumption sequence based on historical electricity consumption data, and then determines the initial forecasted electricity consumption information for each industry in each region within a preset future time period based on the electricity consumption sequence. The electricity consumption sequence is used to indicate the electricity consumption information of each industry in each region within the preset historical time period. Next, it acquires weather information and historical forecast error information for the preset future time period, and determines initial correction information based on the weather information and historical forecast error information. Finally, it adjusts the initial correction information to obtain target correction information, and then corrects the initial forecasted electricity consumption information based on the target correction information to obtain the target forecasted electricity consumption information for each industry in each region within the preset future time period. The electricity consumption forecasting method provided in this application, after obtaining the initial forecasted electricity consumption information for each industry in each region within the preset future time period, further corrects the initial forecasted electricity consumption information based on weather information and historical forecast error information within the preset future time period. This ensures that the final target forecasted electricity consumption information takes into account the influence of weather and historical forecast deviations, thereby effectively improving the accuracy of the forecasted electricity consumption information.

[0053] In some exemplary embodiments, determining the initial predicted electricity consumption information for each industry in each region within a preset time period based on the electricity consumption sequence includes: inputting the electricity consumption sequence into a pre-trained prediction model to obtain the initial predicted electricity consumption information output by the prediction model.

[0054] Optionally, the prediction model can be an Extreme Gradient Boosting (XGBoost) model, a Light Gradient Boosting Machine (LightGBM) model, or a fusion model based on the above two models. In an optional embodiment, the prediction model can also be a Categorical Boosting (CatBoost) model, a Random Forest model, a Ridge Regression model, a Temporal Neural Network model, a Transformer model, or a multi-model weighted fusion model.

[0055] In some exemplary embodiments, a computer device can input an electricity consumption sequence into a pre-trained prediction model to obtain initial predicted electricity consumption information output by the prediction model.

[0056] Specifically, computer equipment can first perform feature engineering on the electricity consumption sequence, and extract multi-dimensional input features based on the electricity consumption sequence. These multi-dimensional input features can include underlying historical electricity consumption lag features, multi-period rolling statistical features, regional and industry-level aggregation features, year-on-year and month-on-month electricity consumption trend features, meteorological features such as temperature, humidity and cooling / heating day-to-day, calendar and macro business features; the prediction model then outputs initial predicted electricity consumption information based on the multi-dimensional input features.

[0057] In optional embodiments of this application, strict data leakage prevention control rules are set for the model training and feature construction stages of the prediction model. The origin month is defined as the prediction base month, and the origin month is the month before the preset future time period T to be predicted. Only historical data with timestamps no later than the preset future time period T are allowed to construct all input features. Real electricity data of the preset future time period T is not allowed to participate in training and feature splicing, so as to avoid forward data leakage and ensure that the prediction logic conforms to the real business pre-deduction scenario.

[0058] The underlying historical electricity lag characteristics specifically include lag0, lag1, lag2, lag3, lag6, lag12, and lag24; multi-period rolling statistical characteristics include the rolling mean and standard deviation of electricity for 3 months, 6 months, and 12 months; regional and industry-level aggregation characteristics include provincial total, municipal total, province-industry combination, and electricity lag items corresponding to the entire network industry; year-on-year and month-on-month electricity trend characteristics include month-on-month, year-on-year, two-year year-on-year, and the ratio of short-term trend to long-term trend; meteorological characteristics include monthly average temperature, daily maximum temperature, daily minimum temperature, Cooling Degree Day 24 (CDD24), Cooling Degree Day 26 (CDD28), Cooling Degree Day 28 (CDD28), and Heating Degree Day 18 (CDD28). 18, HDD18), number of high-temperature days, number of low-temperature days, number of rainy days; calendar and macro business characteristics include month identifier, quarter identifier, total number of days in the current month, Spring Festival offset identifier, macroeconomic lagging indicators, project commissioning plan, etc.

[0059] During the model training phase, the computer device performs a log1p logarithmic transformation on the monthly real electricity consumption labels corresponding to the samples, thereby reducing the dominance of high electricity consumption extreme samples on the model loss function. After the model completes inference and outputs the original predicted values, the expm1 function is called to perform an inverse transformation to restore the magnitude of the real electricity consumption values. At the same time, all predicted values ​​less than 0 are truncated to zero to eliminate negative values ​​that do not conform to the actual electricity consumption logic, thus obtaining the underlying basic prediction value base_pred corresponding to a single electricity consumption sequence.

[0060] After completing the single-sequence prediction calculation, the computer equipment performs bottom-up hierarchical aggregation processing: summing the underlying basic prediction values ​​of all cities and sub-sectors under the same province to obtain the initial predicted electricity consumption information at the provincial level; then accumulating the initial predicted electricity consumption information of multiple provinces to obtain the initial predicted electricity consumption information at the multi-province aggregate level; and finally integrating and outputting complete initial predicted electricity consumption information covering all levels of cities, provinces, multi-province aggregate, and sub-sectors.

[0061] In one exemplary embodiment, such as Figure 2 As shown, the initial correction information is determined based on weather information and historical forecast error information, including the following steps:

[0062] Step 201: Determine the weather correction component based on weather information, determine the residual correction component based on historical forecast error information, and determine the trend correction component based on historical electricity consumption data.

[0063] In some exemplary embodiments, the computer device may determine a weather correction component based on weather information.

[0064] Specifically, computer equipment can determine temperature correction components and extreme climate correction components based on weather information, and then determine weather correction components based on temperature correction components and extreme climate correction components.

[0065] Furthermore, computer equipment can also determine residual correction components based on historical prediction error information.

[0066] Specifically, computer equipment can pre-build and store historical prediction error information for the entire domain. This historical prediction error information records the deviation ratio between the monthly electricity consumption prediction values ​​and actual electricity consumption values ​​for each province and industry over the years. The formula for calculating the residual ratio can be defined as follows: ,in, The proportion of predicted residuals for province p and historical month m. This represents the actual electricity consumption during the same period. These are the baseline predictions from the model during the same period. It is a very small constant used to prevent the denominator from being zero; the computer equipment can match and search the climate type, industrial structure, and year-on-year deviation characteristics of the future preset time period with historical time, screen the similar historical time periods with the highest similarity of meteorological and electricity consumption patterns, extract the corresponding residual ratio, and complete the fitting operation by associating the degree of weather anomaly and regional error fluctuation patterns, so as to determine the residual correction component that is suitable for the future preset time period.

[0067] Furthermore, computer equipment can also determine trend correction components based on historical electricity consumption data.

[0068] Specifically, computer equipment can calculate the year-on-year, month-on-month, and two-year comparison ratios of electricity consumption for a future preset time period based on historical electricity consumption data. It can also perform boundary verification by combining two types of trend indicators: historical electricity consumption percentile range, long-term electricity consumption growth, and short-term load fluctuation. By determining the deviation of the current electricity consumption trend from the same period in previous years, a reasonable trend fluctuation range is defined. Based on the degree of trend deviation, a trend correction component is generated to compensate for the prediction bias caused by long-term changes in electricity consumption that are not covered by simple meteorological or residual correction methods.

[0069] Step 202: Determine the initial correction information based on the weather correction component, residual correction component, and trend correction component.

[0070] In some exemplary embodiments, after obtaining the weather correction component, residual correction component, and trend correction component, the computer device can determine initial correction information based on the weather correction component, residual correction component, and trend correction component.

[0071] Specifically, computer equipment can perform a weighted summation of the weather correction component, residual correction component, and trend correction component to determine the initial correction information.

[0072] Furthermore, in an optional embodiment of this application, the computer device can also acquire calendar correction components and macro-business correction components, and then perform a weighted summation of the weather correction components, residual correction components, trend correction components, calendar month shift correction components, and macro-business correction components to determine the initial correction information. The weight values ​​corresponding to each correction component can be preset fixed values ​​or obtained through training and optimization on a historical validation set. The calendar correction components are used to compensate for electricity consumption fluctuations caused by the misalignment of the Spring Festival and changes in the number of holiday days, while the business correction components are used to correct for electricity consumption deviations caused by business factors such as the commissioning of major projects and temporary industrial and commercial electricity use.

[0073] In one exemplary embodiment, such as Figure 3 As shown, determining weather correction components based on weather information includes the following steps:

[0074] Step 301: Determine the cooling load pressure, heating load pressure, and target temperature days based on weather information, and determine the temperature correction component based on the cooling load pressure, heating load pressure, and target temperature days.

[0075] Cooling load pressure can be used to indicate the cumulative amount of air conditioning power load generated when the ambient temperature is higher than the cooling threshold; heating load pressure can be used to indicate the cumulative amount of heating power load generated when the ambient temperature is lower than the heating threshold; target temperature days are the number of days within a future preset time period where the daily maximum temperature reaches the high temperature threshold and the daily minimum temperature reaches the low temperature threshold, including high temperature days, extreme high temperature days, continuous low temperature days, cold wave days, etc.

[0076] In some exemplary embodiments, the computer device can determine the cooling load pressure, heating load pressure, and target temperature days based on weather information.

[0077] Specifically, the computer equipment can determine the daily average temperature, daily maximum temperature, and daily minimum temperature data for a preset time period based on weather information, and calculate multiple cooling degree days: Cooling Degree Day 24 (CDD24), Cooling Degree Day 26 (CDD26), and Cooling Degree Day 28 (CDD28). The calculation formula is as follows: Where θ is the corresponding temperature threshold. The temperature is adjusted for the day, and the accumulated result is the cooling load pressure; the heating degree-day 18 (HDD18) is used to characterize the heating load pressure, and the calculation formula is as follows: Statistical forecasts are made on the daily maximum temperature, consecutive low temperatures, and number of days covered by cold waves within the month, yielding the number of target temperature days, such as the number of high-temperature days, the number of extreme high-temperature days, and the number of low-temperature days. Simultaneously, the meteorological anomaly coefficient is calculated as: (Forecast meteorological characteristics - Historical average for the same month) / max(Historical standard deviation for the same month). ), It is a very small constant used to measure the degree of deviation of the current temperature from the same historical period.

[0078] Furthermore, after determining the cooling load pressure, heating load pressure, and target temperature days based on weather information, the computer equipment can determine the temperature correction component based on the cooling load pressure, heating load pressure, and target temperature days.

[0079] Specifically, computer equipment can combine cooling time-days, heating time-days, number of days at various target temperatures, and meteorological anomaly coefficients to comprehensively determine the current period's cooling and heating load disturbance magnitude, match the temperature sensitivity characteristics of various industries, quantify the increase or decrease in electricity consumption caused by high and low temperatures, compare the corresponding deviation patterns of temperature and electricity consumption in the same period of history, fit the electricity consumption offset in the temperature dimension, and generate temperature correction components.

[0080] Step 302: Identify extreme climate events in the weather information and determine the extreme climate correction component based on the identification results.

[0081] In some exemplary embodiments, the computer device can identify extreme weather events from weather information and determine extreme weather correction components based on the identification results.

[0082] Specifically, computer equipment can first determine data such as temperature, rainfall, wind force, and meteorological warning signs within a preset time period based on weather information, and then identify various extreme climate events such as extreme high temperatures, sustained low temperatures and cold waves, heavy rainfall, typhoons, and combined high temperature and high humidity weather. It then calculates risk scores for each type of event: the extreme high temperature score (HeatScore) is calculated based on a cooling degree day of 28, the number of extreme high temperature days, and the historical high temperature quantile for the same period; the sustained low temperature score (ColdScore) is calculated based on a heating degree day of 18, the number of consecutive low temperature days, and the duration of the cold wave; the heavy rainfall score (RainScore) is determined by the total number of heavy rainfall days, the duration of continuous rainfall, and the rainfall quantile; the typhoon score (TyphoonScore) is calculated by combining typhoon impact indicators, wind force level, rainfall intensity, and the number of impact days; and the compound meteorological score (CompoundScore) is used to characterize scenarios where multiple extreme climates coexist, such as high temperature and high humidity, or low temperature and heavy rainfall.

[0083] Subsequently, a weighted formula is used to calculate the comprehensive extreme climate impact score. The weighting coefficients for each score can be preset or obtained through training with historical validation samples. The computer classifies the risk level based on the extreme climate impact score, labeling the months as no extreme impact, mild extreme impact, moderate extreme impact, and severe extreme impact. Then, by combining the electricity consumption sensitivity of different regions and industries to different extreme climates and matching the electricity consumption deviation patterns of similar extreme months in history, the magnitude of electricity consumption increase or decrease caused by extreme climates is quantified, and finally, an extreme climate correction component is generated.

[0084] Step 303: Determine the weather correction component based on the temperature correction component and the extreme climate correction component.

[0085] In some exemplary embodiments, after determining the temperature correction component and the extreme climate correction component, the computer device can determine the weather correction component based on the temperature correction component and the extreme climate correction component.

[0086] Specifically, the computer equipment superimposes and merges the temperature correction component and the extreme climate correction component. The two components respectively represent the electricity deviation caused by the fluctuation of normal cold and warm temperatures and the impact of extreme weather. There is no duplicate calculation conflict. The temperature correction component is calculated based only on the monthly average cooling and heating load and the number of normal high and low temperature days. It is used to correct the electricity deviation caused by normal temperature changes. The extreme climate correction component separately quantifies and compensates for unconventional and severe meteorological disturbances such as high temperature heat waves, cold waves, typhoons, and continuous heavy rainfall. The two components cover different dimensions of meteorological impact.

[0087] In one exemplary embodiment, such as Figure 4 As shown, the initial correction information is adjusted to obtain the target correction information, including the following steps:

[0088] Step 401: Determine the climate scenario type corresponding to the preset time period in the future based on weather information, and determine the adjustment strategy according to the climate scenario type.

[0089] In some exemplary embodiments, a computer device can determine the type of climate scenario corresponding to a future preset time period based on weather information.

[0090] Specifically, computer equipment can determine the type of climate scenario based on extreme climate impact scores and preset thresholds. When the extreme climate impact score is below the first threshold, it is determined to be a normal climate scenario; when the extreme climate impact score is greater than or equal to the first threshold and less than the second threshold, it is determined to be a mild extreme climate impact scenario; when the extreme climate impact score is greater than or equal to the second threshold and less than the third threshold, it is determined to be a moderate extreme climate impact scenario; and when the extreme climate impact score is greater than or equal to the third threshold, it is determined to be a severe extreme climate impact scenario.

[0091] Furthermore, after determining the climate scenario type corresponding to a preset time period based on weather information, the computer equipment can determine an adjustment strategy according to the climate scenario type.

[0092] Specifically, computer equipment can configure exclusive upper and lower boundary constraints (i.e., pruning thresholds) for each type of climate scenario as a core adjustment strategy: a narrower correction range is set for normal climate scenarios, with the correction range limited to ±2%; the constraints are relaxed for mild extreme climate impact scenarios, with the correction range limited to ±4%; the threshold range is further widened for moderate extreme climate impact scenarios, while anomaly marking is enabled; the maximum correction range is adopted for severe extreme climate impact scenarios, and a manual review process is forcibly triggered; different provinces and industries can make differentiated fine-tuning of the upper and lower limit thresholds based on local annual climate characteristics and industrial electricity sensitivity; this adjustment strategy will serve as the basis for subsequent pruning constraints on the initial correction information, used to limit large corrections without basis and avoid the correction ratio deviating from the actual electricity fluctuation range.

[0093] Step 402: Adjust the initial correction information based on the adjustment strategy to obtain the target correction information.

[0094] In some exemplary embodiments, after determining the adjustment strategy, the computer device can adjust the initial correction information based on the adjustment strategy to obtain the target correction information.

[0095] Specifically, the initial correction information is the original correction ratio without applied amplitude constraints. The adjustment strategy includes a lower and upper correction limit that matches the current climate scenario, region, and industry. Computer devices can call the truncation constraint function to perform pruning operations. If the original correction ratio is less than the preset lower limit, the correction ratio will be forcibly pulled to the lower limit value. If the original correction ratio is greater than the preset upper limit, the correction ratio will be forcibly pulled to the upper limit value. If the original correction ratio is within the upper or lower limit range, the original correction ratio will remain unchanged.

[0096] Simultaneously, the computer equipment can perform year-on-year and month-on-month business boundary verification. If the adjusted ratio after trimming still causes the predicted electricity consumption to deviate from the historical reasonable percentile range year-on-year and month-on-month, the upper and lower limit thresholds will be tightened again and trimming will be performed again. For the correction results that exceed the normal floating range under severe extreme weather scenarios, the risk level will be marked simultaneously and the manual review mark will be retained. The constrained adjustment ratio output after trimming and boundary verification is the target correction information used to correct the initial predicted electricity consumption information.

[0097] In one exemplary embodiment, such as Figure 5 As shown, the initial predicted electricity consumption information is corrected based on the target correction information to obtain the target predicted electricity consumption information for each industry in each region within a preset time period. This includes the following steps:

[0098] Step 501: Determine the weight values ​​of each industry in each region based on the initial predicted electricity consumption information and industry sensitivity.

[0099] In some exemplary embodiments, the computer device can determine the weight values ​​of each industry in each region based on initial predicted electricity consumption information and industry sensitivity.

[0100] Specifically, computer equipment can first collect the initial forecast electricity consumption information of all cities and all sub-sectors within a single province, summarizing the total initial forecast electricity consumption for the province. Combined with the target correction ratio that has already completed boundary constraints, the monthly overall correction amount for the province is calculated, serving as the fixed constraint amount for industry-wide correction allocation. Secondly, pre-trained and fixed industry-specific electricity sensitivity parameters are retrieved. These industry sensitivity parameters are fixed parameters trained based on long-term historical samples and are used to quantify the electricity consumption response of different industries to temperature fluctuations, extreme weather, and meteorological anomalies. Residential and commercial service industries exhibit high temperature sensitivity, while high-energy-consuming industries and basic manufacturing industries have relatively low temperature sensitivity. Simultaneously, by combining the historical meteorological load response patterns of various cities, the characteristics of industry electricity consumption structures, and the fluctuation characteristics of historical forecast residuals, the basic industry sensitivity is regionally fine-tuned to form accurate sensitivity coefficients adapted to different regions and sub-sectors. Finally, by combining the initial forecast electricity volume of each industry with the industry's climate sensitivity, the corrected allocation weight value of a single "province-city-industry" sequence is calculated. The larger the electricity volume and the higher the weather sensitivity of an industry, the higher the weight of the allocated correction amount, ensuring that the correction logic conforms to the actual electricity consumption disturbance pattern and avoiding the fine deviation caused by average allocation.

[0101] Step 502: Using the weight values, the initial predicted electricity consumption information is corrected based on the target correction information to obtain the target predicted electricity consumption information.

[0102] In some exemplary embodiments, the computer device may use weight values ​​to correct the initial predicted electricity consumption information based on the target correction information to obtain the target predicted electricity consumption information.

[0103] Specifically, the computer equipment uses the provincial overall correction total as a rigid constraint. It distributes the fixed correction total for the entire province to the initial predicted electricity consumption information of each city and sub-sector according to differentiated weight values ​​for each industry, obtaining a single-sequence-specific corrected electricity consumption. The distributed corrected electricity consumption is then added to the initial predicted electricity consumption information of the corresponding industry to complete the electricity consumption correction calculation for the single-dimensional sequence, generating the initial corrected prediction value at the lower level. After correction, the system performs a bottom-up hierarchical consistency check: it re-aggregates the corrected predicted electricity consumption of all lower-level industries across the province, verifying that the aggregated total is completely consistent with the provincial corrected total electricity consumption, eliminating hierarchical aggregation deviations, and ensuring self-consistency and uniformity of data at the city, industry, provincial, and multi-province levels. After verification, it finally outputs target predicted electricity consumption information covering all levels, including province, city, sub-sector, and multiple levels. Simultaneously, complete correction and traceability logs are retained, accurately recording key information such as the weight ratio of each industry, the allocated correction amount, climate scenario constraints, correction ratio values, and the impact level of extreme climate. This can support subsequent error review, manual verification, and business traceability, meeting the full-scenario business application needs of power grid monthly power balance, operation indicator calculation, and load trend analysis.

[0104] In an optional embodiment of this application, after completing the calculation of the target predicted electricity consumption information for each industry, a multi-level data self-consistency verification and result fallback constraint mechanism is added to ensure the consistency, rationality, and business availability of the overall prediction data. Specifically, the computer equipment can perform bottom-up hierarchical summary verification, including city-level summary verification, province-level summary verification, and network-wide total level summary verification, to ensure that the cumulative result of the target predicted electricity consumption information after correction for all lower-level sub-industries is completely consistent with the total correction amount for the upper-level region, eliminating data misalignment, total deviation, and hierarchical non-closure problems caused by hierarchical correction, and achieving bidirectional unification of fine-grained industry prediction and macro-regional total prediction.

[0105] In an optional embodiment of this application, a fallback rule for positive and negative deviations in the forecast results is configured. For months with extreme weather, months of industrial policy adjustments, and months of major project commissioning / retirement, a secondary reasonableness constraint is applied. When the year-on-year and month-on-month fluctuations in the forecasted electricity volume for a single industry or city exceed the reasonable percentile range for the same period in history, the equipment automatically triggers a secondary boundary tightening strategy. This strategy combines the extreme values ​​of electricity volume fluctuations in the same month over the past three years, the annual electricity consumption growth rate range, and the industry's capacity utilization level to slightly correct the forecast results. This avoids distortion caused by single weather corrections or residual corrections, ensuring that the forecast fluctuations closely reflect the actual evolution of electricity load.

[0106] In one optional embodiment of this application, a special secondary calibration mechanism is activated for severe extreme weather scenarios, typhoon and heavy rainfall combined scenarios, and special scenarios such as the Spring Festival falling in the wrong month. Conventional weight allocation strategies have certain limitations in these scenarios. The device can additionally introduce the power response elasticity coefficient from similar historical extreme events to slightly strengthen or weaken the correction for highly sensitive industries, while suppressing fluctuations in less sensitive steady-state industries, further improving the prediction accuracy under special operating conditions.

[0107] In one optional embodiment of this application, all correction processes, parameter values, scenario determinations, weight allocation, and boundary trimming results generate traceable audit logs throughout the entire process. The logs specifically record all fields, including the original base forecast values ​​for each month, the values ​​of correction components for each dimension, the climate scenario determination level, the upper and lower correction thresholds, the industry sensitivity weight, the sub-item allocated correction amount, the secondary correction magnitude, and anomaly marker information. This ensures that the prediction results are interpretable, the process is traceable, and the errors are reviewable, meeting the technical audit requirements for power production, operational statistics, and algorithm iteration optimization.

[0108] In one optional embodiment of this application, the final output of the target predicted electricity consumption information is a multi-dimensional structured result, which not only includes monthly predicted electricity consumption values ​​by province, city, and industry, but also outputs auxiliary results such as meteorological impact contribution, residual correction contribution, trend deviation contribution, extreme climate risk level, predicted fluctuation range, and data confidence label, forming a complete prediction output system of "numerical value + correction logic + risk assessment", which can be directly adapted to various business scenarios such as monthly power grid supply and demand balance analysis, electricity operation budget, load trend analysis, abnormal electricity consumption monitoring, and regional power operation analysis.

[0109] In one optional embodiment of this application, the entire prediction process maintains a time-series anti-leakage mechanism throughout the entire chain. Feature construction, model training, sample selection, and parameter fitting correction all use only historical information before the prediction base month. The real data of the target month is isolated throughout the process, ensuring that the overall prediction logic fully fits the pre-engineering simulation scenario and avoids algorithm defects such as post-fitting and data penetration, thus ensuring the model's generalization ability and the reliability of business implementation.

[0110] In one exemplary embodiment, such as Figure 6 As shown, another method for electricity consumption forecasting is provided, which includes the following steps:

[0111] Step 601: Determine the electricity consumption sequence based on historical electricity consumption data, and input the electricity consumption sequence into the pre-trained prediction model to obtain the initial predicted electricity consumption information output by the prediction model. The electricity consumption sequence is used to indicate the electricity consumption information of various industries in various regions within a preset historical time period.

[0112] Step 602: Obtain weather information and historical prediction error information for the future preset time period, determine the cooling load pressure, heating load pressure and target temperature days based on the weather information, and determine the temperature correction component based on the cooling load pressure, heating load pressure and target temperature days.

[0113] Step 603: Identify extreme climate events in the weather information and determine the extreme climate correction component based on the identification results; determine the weather correction component based on the temperature correction component and the extreme climate correction component; determine the residual correction component based on historical forecast error information and the trend correction component based on historical electricity consumption data.

[0114] Step 604: Determine initial correction information based on weather correction components, residual correction components, and trend correction components; determine the climate scenario type corresponding to the future preset time period based on weather information, and determine the adjustment strategy according to the climate scenario type;

[0115] Step 605: Adjust the initial correction information based on the adjustment strategy to obtain the target correction information; determine the weight values ​​of each industry in each region based on the initial predicted electricity consumption information and industry sensitivity; use the weight values ​​to correct the initial predicted electricity consumption information based on the target correction information to obtain the target predicted electricity consumption information.

[0116] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0117] Based on the same inventive concept, this application also provides an electricity forecasting device for implementing the electricity forecasting method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the electricity forecasting device provided below can be found in the limitations of the electricity forecasting method described above, and will not be repeated here.

[0118] In one exemplary embodiment, such as Figure 7As shown, an electricity consumption prediction device 700 is provided, including: a determination module 701, an acquisition module 702, and an execution module 703, wherein:

[0119] The determination module 701 is used to determine the electricity consumption sequence based on historical electricity consumption data, and to determine the initial predicted electricity consumption information of each industry in each region within a future preset time period based on the electricity consumption sequence. The electricity consumption sequence is used to indicate the electricity consumption information of each industry in each region within a historical preset time period.

[0120] The acquisition module 702 is used to acquire weather information and historical prediction error information within a future preset time period, and to determine initial correction information based on the weather information and historical prediction error information.

[0121] The execution module 703 is used to adjust the initial correction information to obtain the target correction information, and to correct the initial predicted electricity consumption information based on the target correction information to obtain the target predicted electricity consumption information of each industry in each region within a future preset time period.

[0122] In one embodiment, the acquisition module 702 is specifically used to determine the weather correction component based on weather information, determine the residual correction component based on historical forecast error information, and determine the trend correction component based on historical electricity consumption data; and determine the initial correction information based on the weather correction component, the residual correction component, and the trend correction component.

[0123] In one embodiment, the acquisition module 702 is specifically used to determine the cooling load pressure, heating load pressure, and target temperature days based on weather information, and to determine the temperature correction component based on the cooling load pressure, heating load pressure, and target temperature days; to identify extreme climate events in the weather information, and to determine the extreme climate correction component based on the identification results; and to determine the weather correction component based on the temperature correction component and the extreme climate correction component.

[0124] In one embodiment, the execution module 703 is specifically used to determine the climate scene type corresponding to a future preset time period based on weather information, and to determine an adjustment strategy based on the climate scene type; and to adjust the initial correction information based on the adjustment strategy to obtain the target correction information.

[0125] In one embodiment, the execution module 703 is specifically used to determine the weight values ​​of each industry in each region based on the initial predicted electricity consumption information and industry sensitivity; and to use the weight values ​​to correct the initial predicted electricity consumption information based on the target correction information to obtain the target predicted electricity consumption information.

[0126] In one embodiment, the determining module 701 is specifically used to input the electricity consumption sequence into a pre-trained prediction model to obtain the initial predicted electricity consumption information output by the prediction model.

[0127] Each module in the aforementioned electricity consumption prediction device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0128] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 8 As shown, this computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When executed by the processor, the computer program implements a power consumption prediction method.

[0129] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0130] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the methods described in any of the above embodiments.

[0131] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the methods described in any of the above embodiments.

[0132] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the methods described in any of the above embodiments.

[0133] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0134] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0135] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0136] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for predicting electricity consumption, characterized in that, The method includes: Based on historical electricity consumption data, an electricity consumption sequence is determined, and based on the electricity consumption sequence, initial predicted electricity consumption information for each industry in each region within a future preset time period is determined. The electricity consumption sequence is used to indicate the electricity consumption information of each industry in each region within a historical preset time period. Obtain weather information and historical prediction error information within the preset future time period, and determine initial correction information based on the weather information and the historical prediction error information; The initial correction information is adjusted to obtain target correction information, and the initial predicted electricity consumption information is corrected based on the target correction information to obtain the target predicted electricity consumption information for each industry in each region within the future preset time period.

2. The method according to claim 1, characterized in that, The determination of initial correction information based on the weather information and the historical forecast error information includes: Based on the weather information, a weather correction component is determined; based on the historical forecast error information, a residual correction component is determined; and based on historical electricity consumption data, a trend correction component is determined. The initial correction information is determined based on the weather correction component, the residual correction component, and the trend correction component.

3. The method according to claim 2, characterized in that, The step of determining the weather correction component based on the weather information includes: Based on the weather information, the cooling load pressure, heating load pressure, and target temperature days are determined, and a temperature correction component is determined based on the cooling load pressure, heating load pressure, and target temperature days. Extreme weather events are identified in the weather information, and extreme weather correction components are determined based on the identification results; The weather correction component is determined based on the temperature correction component and the extreme climate correction component.

4. The method according to claim 1, characterized in that, The process of adjusting the initial correction information to obtain the target correction information includes: Based on the weather information, determine the climate scenario type corresponding to the future preset time period, and determine the adjustment strategy based on the climate scenario type; The initial correction information is adjusted based on the adjustment strategy to obtain the target correction information.

5. The method according to any one of claims 1 to 4, characterized in that, The step of correcting the initial predicted electricity consumption information based on the target correction information to obtain the target predicted electricity consumption information for each industry in each region within the preset future time period includes: The weight values ​​of each industry in each region are determined based on the initial predicted electricity consumption information and industry sensitivity. Using the weight values, the initial predicted electricity consumption information is corrected based on the target correction information to obtain the target predicted electricity consumption information.

6. The method according to claim 1, characterized in that, The step of determining the initial predicted electricity consumption information for each industry in each region within a preset time period based on the electricity consumption sequence includes: The electricity consumption sequence is input into a pre-trained prediction model to obtain the initial predicted electricity consumption information output by the prediction model.

7. An electricity consumption prediction device, characterized in that, The device includes: The determination module is used to determine the electricity consumption sequence based on historical electricity consumption data, and to determine the initial predicted electricity consumption information of each industry in each region within a future preset time period based on the electricity consumption sequence. The electricity consumption sequence is used to indicate the electricity consumption information of each industry in each region within the historical preset time period. The acquisition module is used to acquire weather information and historical prediction error information within the future preset time period, and determine initial correction information based on the weather information and the historical prediction error information; An execution module is used to adjust the initial correction information to obtain target correction information, and to correct the initial predicted electricity consumption information based on the target correction information to obtain target predicted electricity consumption information for each industry in each region within the future preset time period.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images