Multi-region electricity consumption coupling prediction method

Abstract

The invention discloses a multi-region power consumption coupling prediction method, and relates to the technical field of power grids, and the method comprises the following steps: collecting power consumption coupling data of a to-be-detected region, the power consumption coupling data comprising power transmission influence data, power grid stability, historical power demand data and load fluctuation state data; the historical power demand data and the historical load fluctuation state data are predicted and analyzed through a RegARIMA model and a NeuralProphet model, and the power demand data and the load fluctuation state of the to-be-tested area in a period of time in the future are obtained; constructing a power consumption coupling model based on a machine learning algorithm according to the power demand, the power grid stability, the power transmission efficiency and the load fluctuation state data; and calculating according to the electricity consumption coupling model, the input predicted power demand, the power transmission efficiency and the load fluctuation state data to obtain the electricity consumption coupling coefficient between the to-be-measured areas. The electricity consumption coupling prediction model is constructed to solve the problem that when the load fluctuation and the change of the power demand are not considered, the fixed parameters are often used, and the power consumption of the to-be-measured areas is influenced. And the prediction is not accurate when the power demand change is large.

Description

Technical Field

[0001] The present invention relates to the technical field of power grids, and more specifically, to a method for predicting coupled power consumption in multiple regions. Background Art

[0002] As the proportion of renewable energy continues to increase, the operation of the power system faces new challenges. The randomness and volatility of renewable energy generation affect the prediction of electricity consumption. The multi-regional electricity consumption coupling prediction method better integrates the data of renewable energy and improves the overall prediction accuracy by considering the impact of electricity consumption between regions.

[0003] The power system is composed of multiple regions and subsystems, and there are complex power exchange and transmission relationships between regions. With the opening of the power market and the frequent power transactions, the coupling of power between regions has become increasingly important. Effective power forecasting not only improves the reliability of the power system, but also helps power market participants optimize scheduling and trading strategies.

[0004] The above disclosed technical solutions have at least the following technical problems: Traditional forecasting methods only focus on the electricity consumption of each region, and make forecasts through historical data and statistical models. They focus on the direct relationship between electricity consumption between regions, but do not consider the dynamic changes in regional electricity demand, resulting in an inability to accurately reflect the actual situation. At the same time, when load fluctuations and changes in electricity demand are not considered, fixed parameters are often used, resulting in inaccurate forecasts when electricity demand changes greatly.

[0005] In view of the above problems, the present invention proposes a solution. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a multi-region electricity consumption coupling prediction method, which constructs an electricity consumption coupling prediction model to solve the problem that fixed parameters are often used when load fluctuations and changes in electricity demand are not considered, resulting in inaccurate predictions when electricity demand changes greatly.

[0007] To achieve the above object, the present invention provides the following technical solutions: A multi-region power consumption coupling prediction method includes collecting power consumption coupling data of the area to be tested, wherein the power consumption coupling data includes power transmission impact data, power grid stability, historical power demand data and load fluctuation status data; predicting and analyzing the historical power demand data and historical load fluctuation status data through the RegARIMA model and the NeuralProphet model to obtain the power demand data and load fluctuation status of the area to be tested in the future; constructing a power consumption coupling model based on a machine learning algorithm according to the power demand, power grid stability, power transmission efficiency and load fluctuation status data; according to the power consumption coupling model, inputting the predicted power demand, power transmission efficiency and load fluctuation status data for calculation, and obtaining the power consumption coupling coefficient between the areas to be tested.

[0008] In a preferred embodiment, the power transmission influencing data includes transmission distance, load demand, resistance loss and resistivity. A power transmission influencing model is constructed based on the power transmission influencing data based on a machine learning algorithm to obtain the optimal power transmission efficiency, specifically: the specific steps for obtaining the power transmission efficiency are as follows: obtain the location information between the areas to be tested, and calculate the length of the power transmission line between the two areas by the Haversine formula; obtain the historical load data in the area to be tested, and clear the missing values ​​and abnormal values ​​in the collected historical load data by interpolation and interquartile range method, retain normal data, and calculate the average load in each time period using the cleaned load data; obtain the cable performance parameters for power transmission between the areas to be tested, and the cable performance parameters include resistivity, cross-sectional area and rated voltage; calculate the resistance loss based on the preset resistance loss formula according to the load demand, power transmission line length, resistivity, cross-sectional area and rated voltage; calculate the optimal power transmission efficiency between the areas to be tested based on the preset power transmission efficiency formula according to the resistance loss and load demand.

[0009] In a preferred embodiment, the specific steps for obtaining the grid stability are as follows: obtaining the generator performance parameters in the area to be tested, the generator performance parameters including the generator mechanical power, rated power, rated speed, voltage and current; collecting the power system frequency in the area to be tested within a period of time, and smoothing the acquired frequency data through a time series analysis method to obtain a characteristic value of the frequency; calculating the fundamental frequency of the system based on a preset system fundamental frequency formula according to the characteristic value of the frequency; collecting the voltage and current signals of the power system in the area to be tested, processing the signals through Fourier transform, extracting the frequency domain information, and obtaining the phase angle between the voltage and the current; calculating the electrical power based on the phase angle between the voltage and the current, the voltage and the current based on a preset electrical power formula; obtaining the generator-related constants through a moment of inertia testing method; calculating the moment of inertia based on the constants related to the generator, the rated power and the rated speed based on a preset moment of inertia formula; calculating the grid stability based on the moment of inertia, the fundamental frequency of the system, the electrical power and the mechanical power based on a preset grid stability model.

[0010] In a preferred embodiment, the specific steps for obtaining the power demand data are as follows: obtain the voltage and current change data of the power system in the test area over a period of time, and plot the voltage and current change data over a period of time into a curve graph; obtain the load change data over a period of time, obtain the average load and the maximum load, and calculate the load factor based on the average load and the maximum load based on a preset load factor formula; according to the load factor, voltage and current, the power demand data of the test area is calculated by an integral equation.

[0011] In a preferred embodiment, the specific steps for obtaining the load fluctuation state are as follows: obtaining the grid load data of the test area within a period of time, and processing the grid load data, removing abnormal data therein, and ensuring the reliability and accuracy of the data; calculating the average value of the grid load based on the grid load data, and recording the total number of observed grid load data; obtaining the load fluctuation state based on the total number of observed data, the grid load and the average value of the grid load, and calculating based on a preset load fluctuation state formula.

[0012] In a preferred embodiment, according to the power demand, grid stability, power transmission efficiency and load fluctuation state data, a power consumption coupling model is constructed based on a machine learning algorithm to obtain the power consumption coupling coefficient between the areas to be tested, specifically: the power demand data and load state data of the area to be tested are obtained, and the historical power demand data and the historical load fluctuation state data are predicted and analyzed by the RegARIMA model and the NeuralProphet model to obtain the power demand data and the load state data for a period of time in the future; the power demand data and the load state data for a period of time in the future are input into the power consumption coupling model to realize the prediction of the coupling of power consumption for a period of time in the future; if the value of the power consumption coupling is lower than the preset value, it is necessary to adjust the values ​​of other parameters in the power consumption coupling model; when the parameter value is adjusted to the optimal state, the value of the power consumption coupling is still lower than the preset value, the system will issue an alarm to remind the management personnel whether to interrupt the energy dispatch to other areas or increase the energy dispatch to other areas, thereby increasing the value of the power consumption coupling, so as to improve the emergency response capability of the system and reduce the risk of power supply interruption.

[0013] The technical effects and advantages of a multi-regional electricity consumption coupling prediction method of the present invention are as follows: 1. The present invention uses the RegARIMA model and the NeuralProphet model to make high-precision predictions of power demand data and load fluctuations in the future. This combination of methods takes advantage of time series analysis and deep learning to capture complex power consumption patterns and changing trends.

[0014] 2. The present invention constructs a power transmission impact model through power transmission impact data, optimizes power transmission efficiency, minimizes resistance loss, and ensures efficient flow of power between different regions. By analyzing power grid stability parameters, the stability of the power grid is accurately assessed, thereby identifying potential power supply risks and system failures in advance. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a structural schematic diagram of the multi-regional electricity consumption coupling prediction method of the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] Embodiment 1, Figure 1The invention provides a method for predicting coupled power consumption in multiple regions, comprising the following steps: S1, collecting power consumption coupling data of the area to be tested, wherein the power consumption coupling data includes power transmission data, grid stability, historical power demand data and load fluctuation status data.

[0018] In this real-time example, the power consumption coupling data includes power transmission impact data, grid stability, historical power demand data and load fluctuation status data, specifically: Power transmission influencing data, the power transmission influencing data including transmission distance, load demand, resistance loss and resistivity, and constructing a power transmission influencing model based on the power transmission influencing data based on a machine learning algorithm to obtain the optimal power transmission efficiency, specifically: The specific steps for obtaining the power transmission efficiency are as follows: Obtain the location information between the areas to be tested, and use the Haversine formula to calculate the length of the power transmission line between the two areas; Obtain historical load data in the area to be tested, and remove missing values ​​and outliers in the collected historical load data through interpolation and interquartile range methods, retain normal data, and use the cleaned load data to calculate the average load in each time period; Acquiring performance parameters of cables for power transmission between the test areas, wherein the cable performance parameters include resistivity, cross-sectional area, and rated voltage; The resistance loss is calculated based on a preset resistance loss formula according to the load demand, the length of the power transmission line, the resistivity, the cross-sectional area and the rated voltage to obtain the resistance loss; The optimal power transmission efficiency between the areas to be tested is obtained by calculating based on a preset power transmission efficiency formula according to the resistance loss and load demand.

[0019] The specific calculation formula of the power transmission efficiency is as follows:

[0020] The specific calculation formula of the resistance loss is as follows:

[0021] In the formula, is the power transmission efficiency, is the load demand, Represents resistance loss, which is the energy loss caused by current passing through resistance. Indicates the voltage at which power is transmitted. represents the resistivity of the material, represents the cross-sectional area of ​​the transmission cable, Represents the length of the power transmission line between two areas.

[0022] Further, the advantages of considering resistance loss from the perspective of load demand, power transmission line length, resistivity, cross-sectional area and rated voltage are analyzed: Resistive loss plays a vital role in the power system, which directly affects the transmission efficiency and cost of electric energy. First, the increase in load demand will lead to an increase in current. According to Ohm's law, the resistance loss is proportional to the square of the current. Therefore, understanding the load demand helps to optimize the design of the power transmission system, select the appropriate voltage level and conductor size, effectively reduce resistance loss, and improve transmission efficiency.

[0023] Secondly, the length and resistivity of the power transmission line are important factors affecting resistance loss. As the line length increases, the resistance loss also increases. In long-distance transmission, the use of high-voltage transmission can effectively reduce the current, thereby reducing resistance loss. By selecting low-resistivity conductor materials, the transmission efficiency can be further improved and energy loss can be reduced.

[0024] Finally, there is a significant relationship between the cross-sectional area of ​​the wire and the resistance loss. Increasing the cross-sectional area of ​​the wire effectively reduces the resistance and thus reduces the loss. However, this will also increase the material cost, so a balance needs to be struck between cost and efficiency in the design. By comprehensively optimizing these factors, the overall performance of the power system can be significantly improved, ensuring safe and efficient transmission of electrical energy.

[0025] Grid stability refers to the ability of the power system to maintain its normal operating state and restore to a balanced state when it is disturbed or fails. Stability is crucial to the safe, reliable and economical operation of the power system.

[0026] The specific steps for obtaining the grid stability are as follows: Acquire performance parameters of the generator in the area to be tested, wherein the performance parameters of the generator include mechanical power, rated power, rated speed, voltage and current of the generator; Collect the power system frequency of the area to be tested within a period of time, and smooth the acquired frequency data through time series analysis method to obtain the characteristic value of the frequency; The fundamental frequency of the system is obtained by calculation based on a preset system fundamental frequency formula according to the characteristic value of the frequency; Collect the voltage and current signals in the power system in the area to be tested, process the signals through Fourier transform, extract the frequency domain information, and obtain the phase angle between the voltage and current; The electrical power is obtained by calculating the voltage and the current based on a phase angle between the voltage and the current and the voltage and the current based on a preset electrical power formula; The generator-related constants are obtained by the moment of inertia test method; The moment of inertia is calculated based on a preset moment of inertia formula according to the constants, rated power and rated speed related to the generator; The grid stability is obtained by calculating based on the rotational inertia, the fundamental frequency of the system, the electrical power and the mechanical power based on a preset grid stability model.

[0027] The specific calculation formula for the grid stability is as follows:

[0028] The specific calculation formula of the fundamental frequency of the system is as follows:

[0029] The specific calculation formula of the electrical power is as follows:

[0030] The specific calculation formula of the moment of inertia is as follows:

[0031] In the formula, is the grid stability factor, is the fundamental frequency of the system, is the mechanical power, is the electrical power, is the moment of inertia, is pi, is the system frequency, is the generator terminal voltage, is the generator output current, is the phase angle between voltage and current, is a constant related to the specific generator, is the rated power of the generator, is the rated speed of the generator.

[0032] It should be noted that in the power system, the standard frequency of the system is in the range of 50Hz~60Hz. The higher the stability coefficient of the power grid, the higher the stability of the power grid and the efficiency of energy dispatch, the higher the coupling of power consumption between regions, and the smaller the safety risks; when the stability coefficient of the power grid is lower than the preset value, the efficiency of energy dispatch is lower, and the coupling of power consumption between regions is lower, indicating that the power grid system in the area has a fault and requires manual maintenance from the outside.

[0033] Electricity demand data,Electricity demand refers to the amount of electricity supply required by the power system or electricity users at a specific time. It is usually expressed in kilowatts or megawatts. The forecast of electricity demand is crucial to the planning and operation of the power system because it affects the decisions on power generation, transmission and distribution.

[0034] The specific steps for obtaining the power demand data are as follows: Acquire the voltage and current variation data of the power system in the test area within a period of time, and plot the voltage and current variation data within a period of time into a curve graph; Obtain load change data within a period of time to obtain the average load and the maximum load, and calculate the load factor based on the average load and the maximum load based on a preset load factor formula; According to the load factor, voltage and current, the power demand data of the area to be measured is calculated by an integral equation.

[0035] The specific calculation formula of the power demand data is as follows:

[0036] In the formula, is the power demand, is the voltage, represents the instantaneous voltage at time t, is the current, represents the instantaneous current at time t, is the length of the test period, is the load factor.

[0037] It should be noted that the load factor is the ratio of the actual load to the maximum load. The normal value is between 0.5 and 0.9. The power demand is directly proportional to the load factor. The power demand increases with the increase of the load factor.

[0038] Load fluctuation state, load fluctuation refers to the degree of change of power load within a certain period of time. It is an important indicator used to describe the degree and nature of load changes in the power system. It reflects the fluctuation of power load within a certain period of time. The greater the volatility, the more drastic the change of power load, and the greater the pressure faced by the power system in meeting load demand.

[0039] The specific steps for obtaining the load fluctuation state are as follows: Obtain the grid load data of the area to be tested within a period of time, process the grid load data, remove abnormal data, and ensure the reliability and accuracy of the data; According to the power grid load data, an average value of the power grid load is calculated, and a total amount of observed power grid load data is recorded; The load fluctuation state is obtained according to the total number of the observed data, the grid load and the average value of the grid load and based on a preset load fluctuation state formula.

[0040] The specific calculation formula of the load fluctuation state is as follows:

[0041] In the formula, It is a load fluctuation state. is the total number of observations, is the grid load, represents the grid load value at the i-th time point, Indicates the average value of the grid load.

[0042] S2, through the RegARIMA model and NeuralProphet model, the historical power demand data and historical load fluctuation status data are predicted and analyzed to obtain the power demand data and load fluctuation status of the tested area in the future.

[0043] In this embodiment, the historical power demand data and the historical load fluctuation state data are predicted and analyzed by the RegARIMA model and the NeuralProphet model to obtain the power demand data and the load fluctuation state of the tested area in the future period of time, specifically: The specific calculation formula of the RegARIMA model is as follows:

[0044] The specific calculation formula of the trend function is as follows:

[0045] The specific calculation formula of the periodic change is as follows:

[0046] The specific calculation formula for the holiday item is as follows:

[0047] The specific calculation formula of the NeuralProphet model is as follows:

[0048] In the formula, is the trend function, Represents periodic changes, It is a holiday item. It is used to reflect abnormal changes not reflected in the model, that is, the error term. is the carrying capacity, is the growth rate, is the offset parameter, is the regression parameter, is with The regression variables observed in the same period, is the error time series.

[0049] S3, constructing a power consumption coupling model based on a machine learning algorithm according to the power demand, power transmission efficiency and load fluctuation status data.

[0050] In this embodiment, the power consumption coupling model is constructed based on the machine learning algorithm according to the power demand, power transmission efficiency and load fluctuation state data, specifically: The specific calculation formula of the power consumption coupling model is as follows:

[0051] In the formula, It is the coupling of power consumption. represents the grid stability in the region, represents the electricity demand data of the region, represents the load fluctuation state of the th area, and i represents the number of areas.

[0052] It should be noted that the value of power consumption coupling is between [0,1]. The higher the value of power consumption coupling, the faster the dynamic response of the system, and other power generation equipment will be adjusted quickly to maintain stability. If the coupling value is lower, the dynamic response of the system will become slower, resulting in instability caused by load fluctuations, and then causing abnormal changes in voltage and frequency.

[0053] S4, according to the power consumption coupling model, the predicted power demand, power transmission efficiency and load fluctuation state data are input for calculation to obtain the power consumption coupling coefficient between the areas to be measured.

[0054] In this embodiment, according to the power consumption coupling model, the predicted power demand, power transmission efficiency and load fluctuation state data are input for calculation to obtain the power consumption coupling coefficient between the areas to be measured, which is specifically: Obtain the power demand data and load status data of the area to be tested, and use the RegARIMA model and NeuralProphet model to predict and analyze the historical power demand data and historical load fluctuation status data to obtain the power demand data and load status data for a period of time in the future; Inputting the power demand data and load status data in the future period into the power consumption coupling model to achieve the prediction of the coupling of power consumption in the future period; If the value of the power consumption coupling is lower than the preset value, it is necessary to adjust the values ​​of other parameters in the power consumption coupling model; When the parameter value is adjusted to the optimal state, and the power consumption coupling value is still lower than the preset value, the system will issue an alarm to remind managers whether to interrupt energy scheduling to other areas or increase energy scheduling to other areas, thereby increasing the power consumption coupling value to improve the system's emergency response capabilities and reduce the risk of power outages.

[0055] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0056] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0057] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0058] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0059] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0060] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A multi-regional electricity consumption coupling prediction method, characterized in that: include: Collecting power consumption coupling data of the area to be tested, wherein the power consumption coupling data includes power transmission impact data, power grid stability, historical power demand data and load fluctuation status data; The historical power demand data and historical load fluctuation status data are predicted and analyzed by the RegARIMA model and the NeuralProphet model to obtain the power demand data and load fluctuation status of the tested area in the future. According to the power demand, grid stability, power transmission efficiency and load fluctuation state data, a power consumption coupling model is constructed based on a machine learning algorithm; According to the power consumption coupling model, the predicted power demand, power transmission efficiency and load fluctuation state data are input for calculation to obtain the power consumption coupling coefficient between the areas to be measured.

2. The multi-regional electricity consumption coupling prediction method according to claim 1 is characterized in that: The power transmission influencing data includes transmission distance, load demand, resistance loss and resistivity. A power transmission influencing model is constructed based on the power transmission influencing data and a machine learning algorithm to obtain the optimal power transmission efficiency, specifically: The specific steps for obtaining the power transmission efficiency are as follows: Obtain the location information between the areas to be tested, and use the Haversine formula to calculate the length of the power transmission line between the two areas; Obtain historical load data in the area to be tested, and remove missing values ​​and outliers in the collected historical load data through interpolation and interquartile range methods, retain normal data, and use the cleaned load data to calculate the average load in each time period; Acquiring performance parameters of cables for power transmission between the test areas, wherein the cable performance parameters include resistivity, cross-sectional area, and rated voltage; The resistance loss is calculated based on a preset resistance loss formula according to the load demand, the length of the power transmission line, the resistivity, the cross-sectional area and the rated voltage to obtain the resistance loss; The optimal power transmission efficiency between the areas to be tested is obtained by calculating based on a preset power transmission efficiency formula according to the resistance loss and load demand.

3. The multi-regional electricity consumption coupling prediction method according to claim 2 is characterized in that: The specific steps for obtaining the grid stability are as follows: Acquire performance parameters of the generator in the area to be tested, wherein the performance parameters of the generator include mechanical power, rated power, rated speed, voltage and current of the generator; Collect the power system frequency of the area to be tested within a period of time, and smooth the acquired frequency data through time series analysis method to obtain the characteristic value of the frequency; The fundamental frequency of the system is obtained by calculation based on a preset system fundamental frequency formula according to the characteristic value of the frequency; Collect the voltage and current signals in the power system in the area to be tested, process the signals through Fourier transform, extract the frequency domain information, and obtain the phase angle between the voltage and current; The electrical power is obtained by calculating the voltage and the current based on a phase angle between the voltage and the current and the voltage and the current based on a preset electrical power formula; The generator-related constants are obtained by the moment of inertia test method; The moment of inertia is calculated based on a preset moment of inertia formula according to the constants, rated power and rated speed related to the generator; The grid stability is obtained by calculating based on the rotational inertia, the fundamental frequency of the system, the electrical power and the mechanical power based on a preset grid stability model.

4. The multi-regional electricity consumption coupling prediction method according to claim 3 is characterized in that: The specific steps for obtaining the power demand data are as follows: Acquire the voltage and current variation data of the power system in the test area within a period of time, and plot the voltage and current variation data within a period of time into a curve graph; Obtain load change data within a period of time to obtain the average load and the maximum load, and calculate the load factor based on the average load and the maximum load based on a preset load factor formula; According to the load factor, voltage and current, the power demand data of the area to be measured is calculated by an integral equation.

5. The multi-regional electricity consumption coupling prediction method according to claim 4 is characterized in that: The specific steps for obtaining the load fluctuation state are as follows: Obtain the grid load data of the area to be tested within a period of time, process the grid load data, remove abnormal data, and ensure the reliability and accuracy of the data; According to the power grid load data, an average value of the power grid load is calculated, and a total amount of observed power grid load data is recorded; The load fluctuation state is obtained according to the total number of the observed data, the grid load and the average value of the grid load and based on a preset load fluctuation state formula.

6. The multi-regional electricity consumption coupling prediction method according to claim 5 is characterized in that: According to the power demand, grid stability, power transmission efficiency and load fluctuation state data, a power consumption coupling model is constructed based on a machine learning algorithm to obtain the power consumption coupling coefficient between the areas to be tested, specifically: Obtain the power demand data and load status data of the area to be tested, and use the RegARIMA model and NeuralProphet model to predict and analyze the historical power demand data and historical load fluctuation status data to obtain the power demand data and load status data for a period of time in the future; Inputting the power demand data and load status data in the future period into the power consumption coupling model to achieve the prediction of the coupling of power consumption in the future period; If the value of the power consumption coupling is lower than the preset value, it is necessary to adjust the values ​​of other parameters in the power consumption coupling model; When the parameter value is adjusted to the optimal state, and the power consumption coupling value is still lower than the preset value, the system will issue an alarm to remind managers whether to interrupt energy scheduling to other areas or increase energy scheduling to other areas, thereby increasing the power consumption coupling value to improve the system's emergency response capabilities and reduce the risk of power outages.

7. The multi-regional electricity consumption coupling prediction method according to claim 6 is characterized in that: The specific calculation formula of the RegARIMA model is as follows: The specific calculation formula of the trend function is as follows: The specific calculation formula of the periodic change is as follows: The specific calculation formula for the holiday item is as follows: The specific calculation formula of the NeuralProphet model is as follows: In the formula, is the trend function, Represents periodic changes, It is a holiday item. It is used to reflect abnormal changes not reflected in the model, that is, the error term. is the carrying capacity, is the growth rate, is the offset parameter, is the regression parameter, is with The regression variables observed in the same period, is the error time series.

8. The multi-regional electricity consumption coupling prediction method according to claim 7 is characterized in that: The specific calculation formula for the grid stability is as follows: The specific calculation formula of the fundamental frequency of the system is as follows: The specific calculation formula of the electrical power is as follows: The specific calculation formula of the moment of inertia is as follows: In the formula, is the grid stability factor, is the fundamental frequency of the system, is the mechanical power, is the electrical power, is the moment of inertia, is pi, is the system frequency, is the generator terminal voltage, is the generator output current, is the phase angle between voltage and current, is a constant related to the specific generator, is the rated power of the generator, is the rated speed of the generator.

9. The multi-regional electricity consumption coupling prediction method according to claim 8 is characterized in that: The specific calculation formula of the load fluctuation state is as follows: In the formula, It is a load fluctuation state. is the total number of observations, is the grid load, represents the grid load value at the i-th time point, Indicates the average value of the grid load.

10. The multi-regional electricity consumption coupling prediction method according to claim 9 is characterized in that: The power consumption coupling model is constructed based on the machine learning algorithm according to the power demand, grid stability, power transmission efficiency and load fluctuation state data, specifically: The specific calculation formula of the power consumption coupling model is as follows: In the formula, It is the coupling of power consumption. represents the grid stability in the region, represents the electricity demand data of the region, represents the load fluctuation state of the th area, and i represents the number of areas.

Images