Rural power utilization reasonable interval determination, regulation and control method and device

Abstract

The invention discloses a method and a device for determining, regulating and controlling a rural power utilization reasonable interval, relates to the technical field of electric digital data processing, and is used for solving the technical problem of how to identify the rural power utilization reasonable interval corresponding to an artificial intelligence application and prevent a power grid from being overloaded by actively controlling the AI penetration degree. The method comprises the following steps: acquiring power consumption data and artificial intelligence application data; performing standardized preprocessing to obtain adaptive modeling data; constructing a reference regression model, and calculating an inverted-U-shaped inflection point parameter based on a first term coefficient and a second term coefficient; performing robustness and endogenous test by taking rural communication infrastructure data as a tool variable, and determining an effective inflection point value; and determining a reasonable power utilization interval by combining the rural power grid safe operation parameters, and generating a regulation and control instruction to be issued to the power grid terminal. According to the invention, precision and scenario of rural power utilization regulation and control are realized, stable operation of a power grid is effectively guaranteed, and load distribution is optimized.

Description

Technical Field

[0001] This invention relates to the field of digital data processing technology, and in particular to a method and apparatus for determining and regulating reasonable ranges of rural electricity consumption. Background Technology

[0002] The application of artificial intelligence (AI) technology in rural production and daily life continues to expand, with scenarios such as smart irrigation, precision farming, and automated processing becoming increasingly reliant on electricity supply. However, rural power grids have long been built primarily for "access to electricity," and their grid structure, load capacity, and level of intelligence are ill-suited to the surge and fluctuations in load brought about by AI applications. How to ensure the appropriate development of AI while preventing grid overload risks has become a key bottleneck in rural digital transformation.

[0003] Existing technologies mainly focus on improving power supply capacity through electrical engineering means such as grid expansion and equipment upgrades, or on passive predictive scheduling based on historical load data. They fail to fully identify the nonlinear correlation between the penetration rate of artificial intelligence applications and the level of rural electricity consumption, and lack quantitative judgment methods for effective inflection point values. This leads to frequent problems such as the blind deployment of high-load artificial intelligence terminals and conflicts between grid overload and residential electricity consumption, making it difficult to achieve risk prevention and dynamic balance under the existing hard constraints of the power grid.

[0004] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main objective of this invention is to provide a method, apparatus, equipment, and storage medium for determining and regulating reasonable rural electricity consumption ranges. This invention aims to solve the technical problem of how to identify reasonable rural electricity consumption ranges corresponding to artificial intelligence applications and prevent power grid overload by actively controlling the degree of AI penetration.

[0006] To achieve the above objectives, the present invention provides a method for determining and regulating a reasonable range of rural electricity consumption, the method comprising the following steps: Acquire electricity consumption data and artificial intelligence application data for the target area; The electricity consumption data and the artificial intelligence application data are preprocessed to obtain standardized data for adaptive modeling. A benchmark regression model is constructed based on the standardized data. The inverted U-shaped inflection point parameters related to rural electricity consumption are calculated based on the coefficients of the first and second terms of the benchmark regression model. Based on rural communication infrastructure data, robustness and endogeneity tests are performed on the associated inflection point parameters to determine effective inflection point values. Based on the effective inflection point value and the safe operation parameters of the rural power grid, a reasonable range of electricity consumption for the target area is determined, and a target electricity consumption control command is generated based on the reasonable range of electricity consumption and sent to the power grid terminal for control.

[0007] In one embodiment, the step of performing standardized preprocessing on the electricity consumption data and the artificial intelligence application data to obtain standardized data for adaptive modeling includes: The interquartile range method is used to identify outliers in the electricity consumption data of the target area and the artificial intelligence application data, and data with labeled outliers is obtained. Remove the outliers from the labeled outlier data to obtain the preliminary data after outlier removal; The missing items in the preliminary data after outlier removal are filled by interpolation to obtain complete data without missing values. Perform a logarithmic transformation on the continuous variables in the complete data without missing data to obtain logarithmic processed data; The logarithmically processed data is then standardized to obtain standardized intermediate data. Redundant variables in the standardized intermediate data are removed to obtain standardized data suitable for modeling.

[0008] In one embodiment, the step of constructing a benchmark regression model based on the standardized data, and calculating the inverted U-shaped inflection point parameters related to rural electricity consumption based on the coefficients of the first and second terms of the benchmark regression model, includes: An initial framework for a benchmark regression model, including regional and year-fixed effects, is constructed to obtain the initial framework for the benchmark regression model. The standardized data is divided into a training dataset and a validation dataset; The initial framework of the model is trained using the training dataset to obtain a preliminary draft of the benchmark regression model containing the optimized coefficients of the first and second terms. The initial draft of the benchmark regression model was validated using the validation dataset, resulting in a validated benchmark regression model. Based on the coefficients of the first and second terms in the validated benchmark regression model, the inverted U-shaped extreme points and the inverted U-shaped inflection point parameters related to rural electricity consumption are determined.

[0009] In one embodiment, the step of validating the initial draft of the benchmark regression model using the validation dataset to obtain a validated benchmark regression model includes: Based on the goodness-of-fit index and the mean squared error index as the judgment index, the validation dataset is input into the initial draft of the benchmark regression model to calculate the actual goodness-of-fit value and the actual mean squared error value. The actual goodness-of-fit value is compared with the preset goodness-of-fit threshold to obtain the first comparison result; By comparing the actual mean square error value with the preset mean square error threshold, a second comparison result is obtained; If both the first comparison result and the second comparison result meet the judgment criteria, the fitting accuracy is determined to be up to standard, and a validated benchmark regression model is obtained.

[0010] In one embodiment, the step of performing robustness and endogeneity tests on the correlation inflection point parameters based on rural communication infrastructure data to determine effective inflection point values ​​includes: The robustness of the correlation inflection point parameters is tested using the time lag method, and the first robustness test result is obtained. The robustness of the correlation inflection point parameters is tested twice using the sample tail reduction method to obtain the second robustness test result. Based on the core explanatory variables in the benchmark regression model, determine whether there is interference from omitted variables in the association between the core explanatory variables and the explained variable, and obtain the endogeneity determination result; Based on the endogeneity determination results, rural communication infrastructure data is used as an instrumental variable to correct interference, and the correlation inflection point parameters after eliminating endogeneity are obtained. By combining the results of the first robustness test, the second robustness test, and the correlation inflection point parameters after eliminating endogeneity, an effective inflection point value is determined.

[0011] In one embodiment, the step of correcting interference using rural communication infrastructure data as an instrumental variable based on the endogeneity determination result to obtain the correlation inflection point parameters after eliminating endogeneity includes: The first-phase and second-phase lagged terms of the artificial intelligence application-related data are extracted from the standardized data and determined as the first instrumental variable and the second instrumental variable, respectively. Rural communication infrastructure data is obtained from regional statistical data, and the rural communication infrastructure data is combined with the first-period lagged term to construct an interaction term, which is used as a third instrumental variable; Perform validity tests on the first instrumental variable, the second instrumental variable, and the third instrumental variable to obtain the instrumental variable validity test results; When the validity test results of the instrumental variables show that the first instrumental variable, the second instrumental variable, and the third instrumental variable all pass the validity test, the parameters of the benchmark regression model are re-estimated using the two-step optimal generalized moment estimation method to obtain the re-estimated correlation inflection point parameters. The re-estimated correlation inflection point parameters are compared with the original correlation inflection point parameters to confirm that the parameter consistency is within the preset error range, thus obtaining the correlation inflection point parameters after endogeneity processing.

[0012] In one embodiment, the step of determining a reasonable electricity consumption range for the target area based on the effective inflection point value and rural power grid safety operation parameters, and generating a target electricity consumption control command based on the reasonable electricity consumption range and issuing it to the power grid terminal for control, includes: Using the effective inflection point value as the core anchor point, and combining the distribution range of artificial intelligence application-related data in the target area sample, the upper and lower limits of the reasonable electricity consumption range of the target area are determined, and the reasonable electricity consumption range of the target area is obtained. Using the safe operation parameters of the rural power grid as constraints, the reasonable power consumption range is checked and corrected so that the active power of the transformer area and the node voltage corresponding to the boundary of the range are within the safe operation range, thus obtaining the corrected reasonable power consumption range; Based on the rural production and residential electricity consumption scenarios in the target area, specific electricity control measures are formulated to adapt to the revised reasonable electricity consumption range, resulting in scenario-adapted electricity control measures. The reasonable power consumption range of the target area and the power consumption control measures adapted to the scenario are converted into an instruction format that can be recognized by the power grid control terminal of the target area to obtain the initial instruction draft; The initial draft of the instruction is optimized into a format that conforms to the power grid dispatching protocol to obtain the target power consumption control instruction, which is then sent to the power grid terminal for control.

[0013] In one embodiment, the method further includes: Real-time monitoring of the operational status of the artificial intelligence application model in the target area; When an abnormality is detected in the operating status, it is determined that an artificial intelligence application model failure has occurred, and a preset emergency inflection point database is activated; Based on the current core production electricity consumption type in the target area, match the target historical valid inflection point sample value from the emergency inflection point database; Based on the emergency inflection point value, and combined with the rural power grid safety operation parameters and core production electricity demand thresholds of the target area, a reasonable range of electricity consumption under the current fault scenario is determined. Based on the reasonable power consumption range, the priority of ensuring core production power consumption and the temporary power restriction ratio of non-core power consumption are determined, and emergency power control instructions for fault scenarios are obtained. The emergency power control command for the fault scenario is pushed to the power grid control terminal for load allocation adjustment, and the emergency power control command for the fault scenario is pushed to the household power terminal for power outage or load reduction in the emergency power restriction section.

[0014] In one embodiment, after the step of matching the target historical valid inflection point sample value from the emergency inflection point database according to the current core production power consumption type of the target area, the method further includes: The target historical valid inflection point sample values ​​are modified by power grid constraint adaptation to obtain the modified emergency inflection point values. The step of performing grid constraint adaptation correction on the target historical valid inflection point sample values ​​to obtain the corrected emergency inflection point values ​​includes: Obtain the operating status information of the power grid in the target area; Based on the rural power grid safety operation parameters, determine whether the electricity load level corresponding to the target historical effective inflection point sample value is suitable for the current power grid operation state, and obtain the suitability determination result. When the compatibility determination result is that the compatibility is not suitable, the target historical effective inflection point sample value is adjusted so that the power load level corresponding to the adjusted target historical effective inflection point sample value meets the rural power grid safe operation parameters. The adjusted target historical effective inflection point sample value is determined as the corrected emergency inflection point value.

[0015] Furthermore, to achieve the above objectives, the present invention also proposes a device for determining and regulating reasonable electricity consumption ranges in rural areas, the device comprising: The data acquisition module is used to acquire electricity consumption data and artificial intelligence application data for the target area; A standardized preprocessing module is used to perform standardized preprocessing on the electricity consumption data and the artificial intelligence application data to obtain standardized data adapted for modeling; The parameter determination module is used to construct a benchmark regression model based on the standardized data, and to calculate the inverted U-shaped inflection point parameters related to rural electricity consumption based on the coefficients of the first and second terms of the benchmark regression model. The parameter verification module is used to perform robustness and endogeneity tests on the correlation inflection point parameters based on rural communication infrastructure data, and to determine the effective inflection point values. The interval determination and instruction generation module is used to determine the reasonable power consumption interval of the target area based on the effective inflection point value and the rural power grid safety operation parameters, so as to generate a target power consumption control instruction based on the reasonable power consumption interval and send it to the power grid terminal for control.

[0016] Furthermore, to achieve the above objectives, the present invention also proposes a device for determining and regulating reasonable rural electricity consumption ranges. The device includes: a memory, a processor, and a program for determining and regulating reasonable rural electricity consumption ranges stored in the memory and executable on the processor. The program for determining and regulating reasonable rural electricity consumption ranges is configured to implement the steps of the method for determining and regulating reasonable rural electricity consumption ranges as described above.

[0017] Furthermore, to achieve the above objectives, the present invention also proposes a storage medium storing a program for determining and regulating reasonable rural electricity consumption ranges. When the program is executed by a processor, it implements the steps of the method for determining and regulating reasonable rural electricity consumption ranges as described above.

[0018] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the method for determining and regulating reasonable ranges of rural electricity consumption as described above.

[0019] One or more technical solutions proposed in this application have at least the following technical effects: By optimizing data quality through standardized preprocessing and deriving the inverted U-shaped correlation between artificial intelligence applications and rural electricity consumption through benchmark regression modeling, and improving parameter reliability through robustness testing and endogeneity interference correction, reasonable ranges for electricity consumption are scientifically defined, and control instructions adapted to the differentiated scenarios of rural production and life are generated. This effectively solves the problems of inaccurate range definition and poor scenario adaptability in traditional control, significantly improves the accuracy and effectiveness of rural electricity consumption control, ensures the stable operation of rural power grids, improves energy utilization efficiency, and provides strong support for the balance of rural electricity supply and demand in the current context. Attached Figure Description

[0020] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0021] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A flowchart illustrating the first embodiment of the method for determining and regulating reasonable ranges of rural electricity consumption in this application; Figure 2 This is a schematic diagram of the reasonable electricity consumption range provided in Embodiment 1 of the method for determining and regulating the reasonable range of rural electricity consumption in this application; Figure 3 This is a structural block diagram provided for Embodiment 2 of the method for determining and regulating reasonable ranges of rural electricity consumption in this application; Figure 4 This is a schematic diagram of the module structure of the rural electricity consumption reasonable range determination and control device according to an embodiment of this application; Figure 5This is a schematic diagram of the equipment structure of the hardware operating environment involved in the method for determining and regulating reasonable ranges of rural electricity consumption in the embodiments of this application.

[0023] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0024] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0025] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0026] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device capable of performing the above functions, such as a device for determining and controlling reasonable rural electricity consumption ranges. The following description uses a device for determining and controlling reasonable rural electricity consumption ranges as an example to illustrate this embodiment and the subsequent embodiments.

[0027] Based on this, embodiments of this application provide a method for determining and regulating reasonable ranges of rural electricity consumption, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the method for determining and regulating reasonable ranges of rural electricity consumption in this application.

[0028] In this embodiment, the method for determining and regulating reasonable rural electricity consumption ranges includes steps S10 to S50: Step S10: Obtain electricity consumption data and artificial intelligence application data for the target area; It should be noted that electricity consumption data refers to measurement information related to power consumption within the target area, namely the active power of the transformer area, node voltage, and load curve, which is used to characterize the level of rural electricity consumption.

[0029] It should be noted that the artificial intelligence application data refers to the penetration rate of registered and operating artificial intelligence terminals within the target area, which is used to characterize the deployment scale of artificial intelligence in rural electricity consumption scenarios.

[0030] Understandably, acquiring electricity consumption data and artificial intelligence application data for the target area involves pulling active power data from transformer substations and AI terminal registry entries through the existing measurement interface to form a raw dataset that can be repeatedly observed, providing a physical quantity basis for subsequent modeling.

[0031] The benefits of this step are that it clarifies the data source and obtains complete basic data, laying a reliable data foundation for subsequent standardized preprocessing and modeling analysis, and avoiding deviations in subsequent analysis results due to missing data or insufficient dimensions.

[0032] Step S20: Standardize and preprocess the electricity consumption data and artificial intelligence application data to obtain standardized data for adaptive modeling; It should be noted that standardization preprocessing refers to a series of operations on the original measurement data, including anomaly identification, missing data imputation, logarithmic transformation, and dimensionless processing, in order to eliminate dimensional differences and abnormal interference, and obtain a dataset with stable distribution and uniform dimensions.

[0033] The interquartile range method refers to using the difference between the upper and lower quartiles of the data distribution as a benchmark to identify and mark outliers that exceed a reasonable range, and is used to remove observation records that deviate significantly.

[0034] Interpolation refers to using observations from adjacent time points or adjacent monitoring stations as a benchmark, and filling in missing data points through linear or spline methods to restore continuous time series.

[0035] Logarithmic transformation refers to taking the natural logarithm of a continuous variable to compress the data scale and reduce the effects of heteroscedasticity.

[0036] Standardization refers to shifting and scaling data with the mean as the center and the standard deviation as the scale, in order to eliminate dimensional differences between different indicators.

[0037] Redundant variables are indicators that have extremely low correlation with the core explanatory or explained variables and do not make a significant contribution to the model's explanation. They are used to avoid multicollinearity interference.

[0038] Understandably, the electricity consumption data and the artificial intelligence application data are subjected to standardized preprocessing, namely, the interquartile range method for outlier removal, interpolation for missing data imputation, logarithmic transformation, standardization processing, and redundant variable removal are performed in sequence to obtain standardized data with uniform dimensions and stable distribution, which is used for subsequent benchmark regression modeling.

[0039] In one feasible implementation, step S20 includes steps A11 to A16: A11: Outliers in electricity consumption data and artificial intelligence application data in the target area are identified by the interquartile range method, and data with labeled outliers are obtained. It should be noted that the interquartile range (IVR) is a commonly used statistical identification method. It involves dividing the data into four equal parts after sorting them by size, taking the middle two intervals as the IVR, and then setting a reasonable range based on the IVR. Data exceeding this range are considered outliers. Outliers are values ​​that deviate significantly from the overall trend of the data. Data labeled with outliers refers to the dataset after identifying and marking the outliers.

[0040] It is understandable that outliers in electricity consumption data and artificial intelligence application data in the target area are identified by the interquartile range method, and the identified outliers are marked to obtain data with labeled outliers.

[0041] The benefit of this step is that it accurately identifies and labels outliers in the original data, providing a clear basis for subsequent outlier removal and preventing outliers from interfering with subsequent data processing and modeling results.

[0042] A12: Remove outliers from the data labeled with outliers to obtain preliminary data after outlier removal; It should be noted that the preliminary data after removing outliers refers to the relatively clean data set obtained after deleting the marked outliers from the data labeled with outliers, but which may still have other problems.

[0043] It is understandable that removing outliers from the data marked as outliers, i.e. deleting the marked outliers, yields the preliminary data after removing outliers.

[0044] The benefit of this step is that it eliminates the impact of outliers on data quality, yields relatively clean preliminary data, and improves the accuracy of subsequent data processing.

[0045] A13: By using interpolation to fill in the missing items in the preliminary data after removing outliers, complete data without missing values ​​is obtained; It should be noted that interpolation is a data completion method that fills in gaps by analyzing the valid data before and after the missing items and calculating reasonable values ​​based on the data change patterns. Missing items refer to blank data not recorded in the dataset, and complete dataset without missing items refers to a complete dataset where no blank data exists after filling in all missing items.

[0046] It is understandable that interpolation is used to fill in the missing items in the preliminary data after outliers have been removed, that is, to fill in the blank data by utilizing the changing patterns of the valid data, so as to obtain complete data without missing items.

[0047] The benefit of this step is that it fills in the missing items in the data, ensures the integrity of the data, and avoids incomplete modeling and analysis results due to missing data.

[0048] A14: Perform a logarithmic transformation on continuous variables in complete data without missing data to obtain logarithmically processed data; It should be noted that continuous variables refer to data types that can take any value and change continuously, such as electricity consumption and population size. Logarithmic transformation is a data transformation method that compresses the numerical range of data by taking the logarithm of each data value, reducing extreme fluctuations and making the data distribution more stable. Logarithmically processed data refers to the dataset obtained after logarithmic transformation.

[0049] It is understandable that performing a logarithmic transformation on continuous variables in complete data without missing data means performing a logarithmic transformation on data that meets the conditions to obtain logarithmically processed data.

[0050] The beneficial effect of this step is to reduce the impact of extreme data fluctuations on subsequent modeling, making the data distribution more stable and better meeting the analytical requirements of the subsequent benchmark regression model.

[0051] A15: Perform standardization on the logarithmically processed data to obtain standardized intermediate data; It should be noted that standardization is a data normalization method. By calculating the difference between each data value and the average value, and then dividing by the standard deviation, data with different units and numerical ranges are converted into a unified standard. Standardized intermediate data refers to the intermediate data product obtained after standardization, which has not yet undergone the final removal of redundant variables.

[0052] It is understandable that standardization processing is performed on logarithmically processed data, that is, the data is transformed by a unified standard to eliminate the differences in data units and obtain standardized intermediate data.

[0053] The benefit of this step is that it eliminates the difference in units between different data, ensures that the weights of each variable are fair in subsequent modeling, and avoids deviations in analysis results due to different units.

[0054] A16: Remove redundant variables from the standardized intermediate data to obtain standardized data suitable for modeling.

[0055] It should be noted that redundant variables refer to variables that are not helpful to the rural electricity consumption analysis objective or that highly overlap with other variables. For example, area codes and area names used to identify the same area are redundant variables with overlapping information. Standardized data for modeling refers to a dataset whose data quality and format fully meet the requirements for subsequent modeling after removing redundant variables.

[0056] It is understandable that by removing redundant variables from the standardized intermediate data, i.e., deleting irrelevant or duplicate variables, we can obtain standardized data suitable for modeling.

[0057] The benefits of this step are that it simplifies the data structure, improves the efficiency of subsequent modeling, avoids redundant information from interfering with the capture of core correlation patterns, and ensures the accuracy of modeling and analysis.

[0058] Step S30: Construct a benchmark regression model based on standardized data, and calculate the inverted U-shaped inflection point parameters of rural electricity consumption correlation using the coefficients of the first and second terms of the benchmark regression model. It should be noted that the inverted U-shaped relationship criterion refers to the fact that the coefficient of the quadratic term of the AI ​​terminal penetration rate in the benchmark regression model is negative and passes the significance test, which is used to confirm the existence of a turning point of diminishing marginal benefits.

[0059] Understandably, a benchmark regression model is constructed based on the standardized data, with the active power of the transformer area as the dependent variable and the penetration rate of artificial intelligence terminals and its quadratic term as the core independent variables, while incorporating regional and year fixed effects. After estimating the coefficients of the first and second terms, the model checks whether the coefficient of the second term is significantly negative. If it is significantly negative, it is determined that there is an inverted U-shaped relationship, and then the inflection point parameter is calculated. If the test is not passed, the subsequent steps are terminated.

[0060] It should be noted that the benchmark regression model refers to a panel regression framework that uses the active power of the transformer area as the explained variable, the penetration rate of artificial intelligence terminals and its quadratic term as the core explanatory variables, and incorporates regional and year fixed effects to quantify the nonlinear relationship between artificial intelligence applications and rural electricity consumption.

[0061] It should be noted that the first-order coefficient refers to the estimated result of the first power of the penetration rate of artificial intelligence terminals in the model, which is used to characterize the initial change direction of the active power in the transformer area when the penetration rate increases.

[0062] It should be noted that the quadratic coefficient refers to the estimated result of the square of the penetration rate of artificial intelligence terminals in the model. It is used to depict the reversal of the change direction of the active power in the transformer area when the penetration rate continues to rise, that is, the curvature basis for forming the inverted U-shaped curve.

[0063] Understandably, a baseline regression model is constructed based on the standardized data, with the standardized active power of the transformer area as the explained variable, the standardized AI terminal penetration rate and its quadratic term as the core explanatory variables, and regional and year fixed effects are added to form a panel regression framework for estimating the coefficients of the linear and quadratic terms.

[0064] Understandably, the inverted U-shaped inflection point parameters related to rural electricity consumption are calculated using the first-order and second-order coefficients. This involves substituting the first-order and second-order coefficients into the inverted U-shaped extreme point formula to determine the AI ​​terminal penetration rate that enables the active power of the transformer area to reach its peak. This penetration rate and its corresponding active power value are then used together to determine the inverted U-shaped inflection point parameters related to rural electricity consumption, which are then used to set the boundaries of the reasonable electricity consumption range.

[0065] Step S40: Based on rural communication infrastructure data, perform robustness and endogeneity tests on the correlation inflection point parameters to determine the effective inflection point values; It should be noted that rural communication infrastructure data refers to data related to the construction of rural communication networks obtained from regional statistical data, including key information such as base station coverage and fiber-to-the-home rate. This data is the basic supporting data for key construction in the process and an important supporting condition for the application of artificial intelligence in rural areas.

[0066] In addition, robustness testing is a method to verify the reliability of correlation inflection point parameters or model results. The core logic is to observe whether the parameters or results remain stable by changing the analysis conditions, thereby judging whether the parameters are not affected by external factors and ensuring their reliability in different scenarios.

[0067] Furthermore, endogeneity testing is a method to check for the interference of omitted variables during the modeling process. Its purpose is to determine whether the association between the core explanatory variable and the explained variable is distorted due to the failure to consider certain key variables, thereby ensuring the authenticity of the association. Additionally, the effective inflection point value is an inflection point parameter confirmed as reliable after robustness and endogeneity testing. It is a value that can truly reflect the critical turning point in the association between rural electricity consumption and artificial intelligence applications, and can be directly used to define a reasonable range for electricity consumption.

[0068] Understandably, relying on rural communication infrastructure data, robustness and endogeneity tests are first conducted on the relevant inflection point parameters, and then the final effective inflection point value is determined by combining the results of the two tests. This dual testing eliminates issues of insufficient parameter stability and endogeneity interference, ensuring the reliability of the effective inflection point value. This provides a core basis for accurately defining reasonable electricity consumption ranges and improves the scientific validity and feasibility of rural electricity consumption control schemes.

[0069] In one feasible implementation, step S40 includes steps A21 to A25: A21: The robustness of the correlation inflection point parameters is tested using the time lag method, and the first robustness test result is obtained. It should be noted that the time lag method is a robustness test that recalculates the inflection point parameter by postponing the time dimension of the data and using data from different time periods. The robustness is then verified by comparing whether the two sets of parameters are close. The first robustness test result refers to the judgment on the stability of the parameter obtained after the time lag method test.

[0070] Understandably, the robustness test of the inflection point parameters is performed using the time lag method, which involves recalculating the parameters using data from the delayed time dimension and comparing them to obtain the first robustness test result.

[0071] The beneficial effect of this step is to verify the stability of the correlation inflection point parameters under time changes, ensure that the parameters do not change significantly with time fluctuations, and improve the reliability of the parameters.

[0072] A22: The robustness of the correlation inflection point parameters is tested twice using the sample tail reduction method to obtain the second robustness test result; It should be noted that the sample shrinking method is a robustness test. It verifies robustness by replacing the most extreme values ​​in the data with nearby non-extreme values, recalculating the inflection point parameter, and comparing the parameters before and after the replacement to see if they are similar. The second robustness test result refers to the judgment on parameter stability obtained after the sample shrinking method test.

[0073] Understandably, the second robustness test of the correlation inflection point parameters is performed by using the sample shrinking method, that is, by replacing the extreme data, recalculating the parameters and comparing them to obtain the second robustness test results.

[0074] The beneficial effect of this step is to further verify the stability of the correlation inflection point parameters, ensure that the parameters are not affected by extreme data, form a double robustness check guarantee, and improve the reliability of the parameters.

[0075] A23: Based on the core explanatory variables in the benchmark regression model, determine whether there is interference from omitted variables in the association between the core explanatory variables and the explained variable, and obtain the endogeneity determination result; It should be noted that the core explanatory variable refers to the key variable that has the greatest impact on the analysis objective, which in this embodiment is the data associated with the artificial intelligence application. The explained variable refers to the target object that the model is analyzing, which in this embodiment is rural electricity consumption data. The endogeneity determination result refers to the final conclusion that determines whether there is any interference from omitted variables in the association between the core explanatory variable and the explained variable.

[0076] Understandably, based on the core explanatory variables in the benchmark regression model, we analyze whether the association between the core explanatory variables and the explained variable is affected by omitted variables, and obtain the endogeneity determination result.

[0077] The beneficial effects of this step are to identify potential interference from omitted variables during the modeling process, clarify whether there is a risk of distortion in the correlation inflection point parameters, and provide a basis for whether subsequent correction of interference is necessary.

[0078] A24: Based on the endogeneity determination results, rural communication infrastructure data is used as an instrumental variable to correct for interference, and the correlation inflection point parameters after eliminating endogeneity are obtained; It should be noted that instrumental variables are variables that meet specific conditions: they are highly correlated with the core explanatory variables and are unaffected by endogeneity. They are used to remove interference and correct the correlation patterns. The inflection point parameter of the correlation after eliminating endogeneity refers to the inflection point parameter that can truly reflect the correlation pattern of the core variables after the interference has been corrected by instrumental variables.

[0079] Understandably, based on the endogeneity determination results, rural communication infrastructure data is used as an instrumental variable. Targeted methods are employed to remove endogeneity interference, ultimately yielding the correlation inflection point parameters after endogeneity elimination. This approach accurately removes implicit interference during the modeling process, allowing the correlation inflection point parameters to truly reflect the core relationship between rural electricity consumption and artificial intelligence applications. This lays a solid foundation for determining effective inflection point values ​​and further improves the accuracy of electricity consumption range definition.

[0080] Further, step A24 includes: The first and second lagged terms of the data related to artificial intelligence applications were extracted from the standardized data and determined as the first and second instrumental variables, respectively. Rural communication infrastructure data were obtained from regional statistical data, and the rural communication infrastructure data was combined with the first-period lagged terms to construct an interaction term, which was used as a third instrumental variable. Validity tests were performed on the first, second, and third instrumental variables to obtain the instrumental variable validity test results. When the instrumental variable validity test results show that the first, second, and third instrumental variables all pass the validity test, the parameters of the benchmark regression model are re-estimated using the two-step optimal generalized moment estimation method to obtain the re-estimated correlation inflection point parameters. The re-estimated correlation inflection point parameters are compared with the original correlation inflection point parameters to confirm that the parameter consistency is within the preset error range, thus obtaining the correlation inflection point parameters after endogeneity processing.

[0081] It should be noted that the first-period lag term refers to data from the previous time period in the AI ​​application-related data, and the second-period lag term refers to data from two time periods in the AI ​​application-related data. The first and second instrumental variables are determined by these two lag terms, respectively. Communication infrastructure data refers to data related to the construction of communication networks within the target area. The cross term refers to the new variable obtained by multiplying the communication infrastructure data by the first-period lag term; this new variable is the third instrumental variable. Validity testing is a method to verify whether the instrumental variables meet the two core conditions of being highly correlated with the core explanatory variables and unaffected by endogeneity interference. The result of the instrumental variable validity test is the conclusion determining whether the three instrumental variables meet the validity requirements. The two-step optimal generalized method of moments (GSM) is a parameter estimation method specifically designed to handle endogeneity problems. It completes parameter estimation in two steps to eliminate interference. The preset error range refers to a pre-set reasonable deviation threshold used to determine whether the re-estimated parameters are within an acceptable range of consistency with the original correlation inflection point parameters. The correlation inflection point parameters after endogeneity processing refer to the final inflection point parameters that have undergone complete processing, eliminating endogeneity interference and confirming reliability.

[0082] Understandably, the process involves first extracting the first and second phase lagged terms of AI application-related data from standardized data as the first and second instrumental variables. Then, communication infrastructure data is obtained from regional statistical data and interacted with the first phase lagged terms to form the third instrumental variable. Validity tests are then performed on the three instrumental variables. If all three pass the tests, the baseline regression model parameters are re-estimated using a two-step optimal generalized method of moments to obtain the re-estimated correlation inflection point parameters. Finally, the re-estimated parameters are compared with the original correlation inflection point parameters to confirm that their consistency is within the preset error range, ultimately yielding the endogeneity-processed correlation inflection point parameters.

[0083] The beneficial effect of this section is that by selecting multiple targeted instrumental variables and verifying their effectiveness, and then using professional methods to re-estimate the parameters and compare and confirm them, endogeneity interference can be accurately removed, ensuring that the final endogeneity-treated correlation inflection point parameters are true and reliable, and providing a high-quality parameter basis for the subsequent determination of effective inflection point values.

[0084] A25: Determine the effective inflection point value by combining the results of the first robustness test, the results of the second robustness test, and the correlation inflection point parameters after eliminating endogeneity.

[0085] It should be noted that the effective inflection point value refers to the final inflection point parameter that is confirmed to be stable and reliable by multiple tests and can be directly used to define the reasonable range of electricity consumption. It can truly and accurately reflect the critical value of the correlation between rural electricity consumption and artificial intelligence applications.

[0086] Understandably, by combining the results of the first robustness test, the second robustness test, and the correlation inflection point parameters after eliminating endogeneity, we can determine whether the parameters are stable and reliable, and thus determine the effective inflection point value.

[0087] The beneficial effect of this step is that by comprehensively considering the results of multiple tests, it ensures that the final determined effective inflection point value is stable, reliable, and free from interference, providing a core basis for the subsequent accurate definition of reasonable electricity consumption ranges.

[0088] Step S50: Determine the reasonable range of electricity consumption in the target area based on the effective inflection point value and the safe operation parameters of the rural power grid, and generate a target electricity consumption control command based on the reasonable range of electricity consumption and send it to the power grid terminal for control.

[0089] It should be noted that the reasonable electricity consumption range is a range determined by combining effective inflection point values ​​and rural power grid safety operation parameters. This range not only meets the actual electricity demand in rural areas but also ensures the safe operation of the power grid, guaranteeing stable grid operation and improving energy utilization efficiency. Target electricity consumption control instructions refer to instructions that meet power grid control requirements and can be directly used to guide rural electricity consumption control work in the target area.

[0090] Understandably, the reasonable range of electricity consumption in the target area is determined based on the effective inflection point value, and target electricity consumption control instructions are generated based on this reasonable range to guide electricity consumption control work.

[0091] like Figure 2 As shown, Figure 2 This diagram illustrates the inverted U-shaped relationship between rural electricity consumption and the penetration rate of AI terminals. It represents the first scenario where, under the constraints of rural power grid safety operation parameters, the allowed electricity consumption exceeds the inflection point power. The horizontal axis represents the AI ​​terminal penetration rate, and the vertical axis represents rural electricity consumption, primarily characterized by the active power of the distribution transformer area. The peak of the curve represents the effective inflection point, corresponding to the critical saturation state under the constraints of rural power grid safety operation parameters. The area enclosed by the dashed lines on both sides represents the reasonable electricity consumption range, the boundary of which is determined by the rural power grid safety operation parameters and used to directly guide the load regulation operation of power grid equipment. If the allowed electricity consumption is less than the inflection point power, the second scenario begins. In this case, the effective inflection point is located outside the right side of the reasonable range, and the right boundary of the reasonable range is limited by the power grid's carrying capacity. The control strategy then shifts to restricting AI penetration.

[0092] In one feasible implementation, step S50 includes steps A31 to A34: A31: Using the effective inflection point as the core anchor point, and combining the distribution range of AI application-related data in the target area sample, determine the upper and lower limits of the reasonable electricity consumption range of the target area, and obtain the reasonable electricity consumption range of the target area. It should be noted that the criteria for defining the upper and lower limits of the reasonable electricity consumption range refer to determining the boundary positions jointly by the effective inflection point value and the safe operation parameters of the rural power grid, including two scenarios. In the first scenario, if the allowable electricity consumption under the constraints of the rural power grid's safe operation parameters is greater than the inflection point power, then the effective inflection point value is used as a benchmark to extend in both directions. The upper limit is the maximum penetration rate jointly constrained by the rated capacity of the distribution area and the upper limit of the allowable deviation of the node voltage, while the lower limit is the minimum penetration rate constrained by the core production electricity demand threshold. In the second scenario, if the allowable electricity consumption is less than the inflection point power, then the upper limit is limited by the penetration rate corresponding to the allowable electricity consumption, while the lower limit is still the minimum penetration rate constrained by the core production electricity demand threshold. In this case, the effective inflection point value is located outside the right side of the reasonable range.

[0093] It should be noted that the core anchor point refers to the key parameter corresponding to the peak of the single inverted U-shaped curve relating rural electricity consumption and artificial intelligence applications, i.e., the effective inflection point. This parameter is the critical value at which the level of artificial intelligence application and rural electricity load reach a critical saturation state. It is the core reference for delineating a reasonable range of electricity consumption, rather than a fixed center of the range. In the first scenario, the range is delineated around this core anchor point. In the second scenario, the left boundary of the range is limited by the core production electricity demand threshold, and the right boundary is limited by the grid carrying capacity. The core anchor point is located outside the right side of the range.

[0094] Understandably, based on the relative magnitude of rural power grid safety operation parameters and inflection point power, the maximum and minimum values ​​of the reasonable power consumption range are dynamically determined, so that the range definition not only conforms to the actual data distribution of the region, but also adapts to different power grid carrying capacity scenarios.

[0095] The beneficial effect of this step is that, through the dynamic boundary determination mechanism, the determined reasonable power consumption range can adapt to both scenarios of sufficient and strained grid capacity, avoiding the inappropriate regulation caused by a single fixed range, and improving the practicality and adaptability of the solution.

[0096] It should be understood that the rationality of the scope of electricity use can be explained in depth from three levels: physical rigidity, economic flexibility, and systemic synergy. The rigidity of physics dictates that safety is paramount. Given the weak structure of rural power grids, any electricity usage exceeding transformer capacity or causing voltage overshoots can lead to tripped protection devices, equipment burnout, and even widespread power outages. The rated capacity of a distribution area represents the physical limit of a transformer; electricity usage within this range should not come at the expense of grid safety or the fundamental rights of other users.

[0097] The lower limit, the threshold for core production electricity demand, gives this range "economic necessity and rationality." In rural areas, AI is a tool, and production is the goal. If the integration of AI applications crowds out core production electricity that directly creates value, such as irrigation, temperature control, refrigeration, and feed processing, the lower limit ensures that AI electricity consumption is always in a subordinate position of "supplementing" rather than "replacing" core production, thus avoiding the "AI crowding-out effect."

[0098] The synergy of systems science has two aspects. First, the guiding role of the inflection point: the inflection point of the inverted U-shaped curve is the point where AI-enabled efficiency enhancement is most significant and the marginal electricity consumption benefit is highest. Using this point as a reference benchmark for a reasonable range means that the goal is not to suppress electricity consumption as much as possible, but to always anchor to the "optimal energy efficiency point." Second, the rationality of the buffer zone. Left buffer zone (from the inflection point to the lower limit): when the grid capacity is insufficient or production pressure is high, the level of AI application is allowed to temporarily regress, sacrificing some AI efficiency to ensure grid security and production needs. Right buffer zone (from the inflection point to the upper limit): when grid conditions are good and renewable energy output is sufficient, AI is allowed to moderately overclock, utilizing redundant power for in-depth data analysis or model iteration.

[0099] In summary, the reasonable range is actually a "three-dimensional intersection" among the constraints of electricity, production, and AI energy efficiency. The upper limit prevents ineffective energy consumption caused by technological frenzy, thus avoiding waste. The lower limit safeguards the lifeline of agricultural production, ensuring that the fundamentals are not sacrificed. The inflection point indicates that the ultimate goal of AI applications is high quality and high efficiency, rather than simply piling up computing power, thus avoiding losing direction.

[0100] A32: Using the safe operation parameters of the rural power grid as constraints, the reasonable power consumption range is checked and corrected so that the active power and node voltage of the transformer area corresponding to the boundary of the range are within the safe operation range, thus obtaining the corrected reasonable power consumption range; It should be noted that the active power of a distribution area refers to the total active power consumed by electrical equipment within a specific distribution area of ​​a rural power grid. It is a core indicator for measuring the electrical load of a distribution area and is directly related to the operating pressure and safety status of the equipment in that area. Additionally, node voltage refers to the voltage values ​​at various connection points in a rural power grid. Its stability is crucial for ensuring the normal operation of electrical equipment and the reliable operation of the power grid, and must comply with the allowable deviation range stipulated by the industry.

[0101] Furthermore, the safe operating range refers to the numerical range within which the active power and node voltage of the distribution area can ensure the fault-free operation of the power grid equipment. This range is determined based on the equipment performance of the rural power grid and industry safety standards. Additionally, the revised reasonable electricity consumption range is obtained after verification using the safety operation parameters of the rural power grid, retaining the scientific basis of effective inflection point values ​​while adapting to the actual safety carrying capacity of the power grid.

[0102] Understandably, using rural power grid safety operation parameters as constraints allows for the checking and adjustment of the initially determined reasonable electricity consumption range. This ensures that the active power of the transformer substations and the node voltage at the range boundaries fall within the safe operating range, ultimately yielding a revised reasonable electricity consumption range. This verification and correction process prevents the reasonable electricity consumption range from exceeding the power grid's safe carrying capacity, ensuring that the range meets both rural electricity demand and power grid stability. This improves the safety and operability of defining the electricity consumption range, providing a more reliable basis for subsequent precise regulation.

[0103] A33: Combining the rural production electricity consumption scenarios and rural living electricity consumption scenarios in the target area, formulate specific electricity consumption control measures for the adapted and corrected reasonable electricity consumption range, and obtain scenario-adapted electricity consumption control measures; It should be noted that rural production electricity use scenarios refer to electricity use scenarios related to agricultural production in rural areas, such as agricultural irrigation and agricultural product processing. Rural residential electricity use scenarios refer to electricity use scenarios related to the daily lives of rural residents, such as household lighting and the use of household appliances. Scenario-adaptive electricity control measures refer to specific control schemes formulated for different electricity use scenarios and adapted to reasonable electricity consumption ranges.

[0104] It is understandable that specific electricity control measures should be formulated to match the reasonable range of electricity consumption, taking into account the electricity consumption scenarios for rural production and rural life in the target area, so as to obtain scenario-appropriate electricity control measures.

[0105] For example, during the peak irrigation season (June-August), electricity consumption for production will be reduced by 100,000 kWh daily from 9 to 11 a.m.; during the Spring Festival, residents will be guided to use high-power appliances during off-peak hours.

[0106] The beneficial effects of this step are that it enables scenario-based differentiation of control measures, avoids one-size-fits-all control instructions, and improves the adaptability and execution effectiveness of control measures for different electricity consumption scenarios.

[0107] A34: Convert the reasonable power consumption range and scenario-adaptive power consumption control measures of the target area into an instruction format that can be recognized by the power grid control terminal of the target area, and obtain the initial draft of the instruction; It should be noted that the command format recognizable by the power grid control terminal refers to the command format that the relevant power grid control equipment can directly interpret and execute, and this format meets the equipment's information reception and processing requirements. The initial command draft refers to the preliminary command after format conversion, but before it has undergone industry standard optimization.

[0108] Understandably, the reasonable range of electricity consumption in the target area and the electricity control measures adapted to the scenario are converted into an instruction format that the power grid control terminal in the target area can recognize, thus obtaining the initial draft of the instruction.

[0109] The beneficial effect of this step is that it enables the conversion of reasonable electricity consumption ranges and control measures from textual descriptions to a machine-readable format, laying the foundation for the execution of subsequent instructions.

[0110] A35: Optimize the initial draft of the instruction into a format that conforms to the power grid dispatching protocol, obtain the target power consumption control instruction, and send it to the power grid terminal for control.

[0111] It should be noted that the power grid dispatching protocol refers to the unified technical standards stipulated by the power grid industry, used to standardize the format, transmission method, and execution logic of instructions, ensuring compatibility between different power grid equipment. The target power consumption control instruction refers to the final control instruction that has been optimized in format, fully complies with industry standards, and can be directly issued and executed.

[0112] Understandably, the initial draft of the instruction is optimized to conform to the format of the power grid dispatching protocol, so that the instruction meets the requirements of industry standards and the target power consumption control instruction is obtained.

[0113] The beneficial effect of this step is to ensure that the final generated target power consumption control instructions comply with industry standards, can be smoothly executed in the power grid system, and guarantee the smooth progress of power consumption control work.

[0114] Furthermore, this method also includes: Real-time monitoring of the operational status of the artificial intelligence application model in the target area; When an abnormality is detected in the operating status, it is determined that an artificial intelligence application model failure has occurred, and a preset emergency inflection point database is activated; Based on the current core production electricity consumption type in the target area, match the target historical valid inflection point sample value from the emergency inflection point database; Based on the emergency inflection point value, and combined with the rural power grid safety operation parameters and core production electricity demand thresholds of the target area, a reasonable range of electricity consumption under the current fault scenario is determined. Based on the reasonable power consumption range, the priority of ensuring core production power consumption and the temporary power restriction ratio of non-core power consumption are determined, and emergency power control instructions for fault scenarios are obtained. The emergency power control command for the fault scenario is pushed to the power grid control terminal for load allocation adjustment, and the emergency power control command for the fault scenario is pushed to the household power terminal for power outage or load reduction in the emergency power restriction section.

[0115] It should be noted that this section provides emergency control solutions for extreme scenarios where abnormal operation of artificial intelligence application models makes it impossible to define the reasonable range of normal electricity consumption.

[0116] The emergency inflection point database refers to a pre-built database storing valid inflection point sample values ​​under historical normal operating scenarios in the target area, after robustness testing and endogeneity correction. The core production electricity demand threshold refers to the minimum electricity load value required to ensure uninterrupted core rural production activities (such as agricultural irrigation and agricultural product preservation). The emergency power control instructions for fault scenarios refer to dedicated control instructions that include reasonable power consumption ranges under fault scenarios, priority for core production electricity supply, and temporary power rationing ratios for non-core electricity consumption. By monitoring the model's operating status in real time, the emergency inflection point database is quickly activated upon identifying anomalies. It matches historical valid inflection point sample values ​​that are compatible with the current core production electricity consumption type. Combined with the grid's preset safe load threshold and core production electricity demand threshold, a reasonable power consumption range is defined, power consumption priorities and rationing ratios are clarified, and instructions are pushed to the grid control terminal and farmers' electricity terminals, achieving closed-loop management of power consumption control during fault periods.

[0117] It solves the problem of failure in defining reasonable power consumption ranges when artificial intelligence application models malfunction, ensuring that reliable power control schemes can still be output in extreme scenarios; it guarantees the core production power needs in rural areas during fault periods, reducing agricultural production losses; and it achieves coordinated response between the power grid side and the farmers' side, improving the timeliness and effectiveness of emergency control.

[0118] Furthermore, this method also includes the step of performing grid constraint adaptation correction on the target historical valid inflection point sample values ​​to obtain the corrected emergency inflection point values, including: Obtain the operating status information of the power grid in the target area; Based on the safe operation parameters of the rural power grid, it is determined whether the electricity load level corresponding to the target historical effective inflection point sample value is suitable for the current power grid operation state, and the suitability determination result is obtained. When the compatibility determination result is that the compatibility is not suitable, the target historical effective inflection point sample value is adjusted so that the power load level corresponding to the adjusted target historical effective inflection point sample value meets the rural power grid safe operation parameters. The adjusted target historical effective inflection point sample value is determined as the corrected emergency inflection point value.

[0119] It should be noted that this section is used to correct the mismatch between the target's historical effective inflection point sample values ​​and the current actual operating status of the power grid.

[0120] Grid constraint adaptation correction refers to the process of adjusting and optimizing historically valid inflection point sample values ​​based on the current grid operating status and safety constraints. The adaptation judgment result refers to the conclusion that the electricity load level corresponding to the historically valid inflection point sample value meets the current grid safety operation constraints. By obtaining the current operating status information of the target area grid and combining it with preset grid safety operation constraints, it is determined whether the electricity load level corresponding to the target historically valid inflection point sample value is adapted to the current grid status. If it is determined to be unsuitable, the sample value is adjusted so that its corresponding electricity load level meets the grid safety constraints, and finally, the corrected emergency inflection point value is determined.

[0121] This addresses the issue that while historical effective inflection point sample values ​​are derived from historical grid conditions and do not match the current grid topology or load distribution, resulting in theoretically feasible but practically unenforceable emergency solutions. It ensures that the corrected emergency inflection point values ​​not only meet the core rural production power demand but also satisfy grid safety operation constraints, thereby improving the practical operability of emergency control commands. It implicitly guarantees the priority of core production power during the adjustment process, avoiding power outages in core production due to grid constraint adjustments.

[0122] This embodiment provides a method for determining and regulating reasonable ranges of rural electricity consumption. By improving data quality through standardized preprocessing throughout the entire process, and determining accurate and effective inflection point values ​​through modeling, dual robustness checks, and endogeneity correction, reasonable ranges of electricity consumption are defined and scenario-based regulation instructions are formulated. This effectively solves the pain points of poor data, biased range definition, and poor adaptability of measures in traditional rural electricity consumption regulation, improves regulation effectiveness, ensures grid stability and efficient energy utilization, and supports the balance of rural electricity supply and demand.

[0123] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 3 Step S30 includes steps S301 to S305: Step S301: Construct the initial framework of the benchmark regression model, which includes regional fixed effects and year fixed effects, to obtain the initial framework of the benchmark regression model. It should be noted that the regional fixed effect refers to fixing the inherent differences between different rural areas. These inherent differences may affect electricity consumption data. Fixing this effect can eliminate the interference of such regional differences on the core association analysis. The year fixed effect refers to fixing the common external factors that vary across different years. These changes may affect electricity consumption. Fixing this effect can prevent common factors at the time level from interfering with the capture of core association patterns. The initial framework of the baseline regression model refers to the basic mathematical calculation structure that includes the regional fixed effect, the year fixed effect, and the various variables defined above. It serves as the foundational framework for subsequent model training.

[0124] Understandably, constructing an initial framework for a benchmark regression model that includes regional and year-specific fixed effects forms the basic computational structure required for subsequent model training, thus obtaining the initial framework for the benchmark regression model.

[0125] The beneficial effect of this step is that by introducing two fixed effects, the interference of regional differences and common temporal changes can be eliminated in advance, and a model framework that is more in line with actual analysis needs can be built, laying the foundation for accurately capturing core correlation patterns in the future.

[0126] Step S302: Divide the standardized data into a training dataset and a validation dataset to obtain the training dataset and the validation dataset; It should be noted that the training dataset refers to the partially standardized data used to train the initial framework of the benchmark regression model and optimize the model parameters; it is the core data source for the model to learn data patterns. The validation dataset refers to the partially standardized data used to test the fitting accuracy of the model after training; its data patterns have not been learned by the model in advance and can objectively reflect the actual performance of the model.

[0127] It is understandable that standardized data is divided into two parts: a training dataset and a validation dataset, resulting in the training dataset and the validation dataset, respectively.

[0128] The benefits of this step are that it enables the division of data usage, the training dataset provides support for model learning, and the validation dataset provides a basis for evaluating model performance, thus avoiding misjudgments of model performance due to data reuse.

[0129] Step S303: Train the initial framework of the model using the training dataset to obtain a preliminary draft of the benchmark regression model containing the optimized coefficients of the first and second terms. It should be noted that the linear coefficients refer to the weighted coefficients of the linear components of the core explanatory variables in the benchmark regression model, used to quantify the degree of linear association between the core explanatory variables and the explained variable. The quadratic coefficients refer to the weighted coefficients of the quadratic components of the core explanatory variables, used to quantify the degree of non-linear association between the core explanatory variables and the explained variable, and are key coefficients for capturing inverted U-shaped associations. The initial draft of the benchmark regression model refers to the preliminary model obtained after training on the training dataset, containing optimized linear and quadratic coefficients, and has not yet undergone accuracy validation.

[0130] Understandably, the initial framework of the benchmark regression model is trained using the training dataset, and the coefficients of the first and second terms of the model are optimized to obtain the initial draft of the benchmark regression model containing the optimized coefficients.

[0131] The beneficial effect of this step is that, through training, the model learns the core correlation patterns in the data, and the optimized coefficients enable the model to initially possess the ability to quantify the correlation between rural electricity consumption and artificial intelligence applications.

[0132] Step S304: Validate the initial draft of the benchmark regression model using the validation dataset to obtain the validated benchmark regression model; It should be noted that the validated benchmark regression model refers to the model whose fitting accuracy meets the preset requirements after being tested with the validation dataset. This model can stably and accurately capture the correlation between rural electricity consumption and artificial intelligence applications.

[0133] Understandably, the validation dataset is used to test the fitting accuracy of the initial draft of the benchmark regression model, and models that meet the fitting accuracy standard are selected to obtain the validated benchmark regression model.

[0134] In one feasible implementation, step S304 includes steps A51 to A55: A51: Based on the goodness-of-fit index and mean squared error index as the judgment index, the validation dataset is input into the initial draft of the benchmark regression model to calculate the actual goodness-of-fit value and the actual mean squared error value. It should be noted that the judgment metric measures the model's ability to explain data patterns; the closer its value is to 1, the stronger the model's explanatory power. The mean squared error metric measures the deviation between the model's predicted values ​​and the actual data values; the smaller its value, the higher the model's prediction accuracy. The criteria for judging the fit accuracy refer to the standards used to determine whether the model's fit is satisfactory; the judgment metrics are the selected goodness-of-fit metric and the mean squared error metric.

[0135] It is understandable that the goodness-of-fit index and the mean squared error index are selected as the basis for judging the model fitting accuracy, and these two indices are determined as the fitting accuracy judgment indices.

[0136] It should be noted that the actual goodness-of-fit value refers to the specific numerical value of the goodness-of-fit index calculated after inputting the validation dataset into the initial draft of the benchmark regression model, reflecting the model's actual interpretability of the validation data. The actual mean squared error value refers to the specific numerical value of the mean squared error index calculated after inputting the validation dataset into the initial draft, reflecting the model's actual prediction deviation from the validation data.

[0137] Understandably, the validation dataset is input into the initial draft of the baseline regression model, and the actual goodness-of-fit value and the actual mean squared error value are obtained through model calculation.

[0138] The benefit of this step is that it obtains actual performance data of the model on unlearned validation data, providing a quantitative basis for subsequent judgment on whether the model is qualified.

[0139] A52: Compare the actual goodness-of-fit value with the preset goodness-of-fit threshold to obtain the first comparison result; It should be noted that the first comparison result refers to the conclusion after comparing the actual goodness-of-fit value with the preset goodness-of-fit threshold.

[0140] A53: Compare the actual mean square error value with the preset mean square error threshold to obtain the second comparison result; It should be noted that the preset goodness-of-fit threshold refers to the pre-set passing standard for the goodness-of-fit index. When the actual goodness-of-fit value reaches or exceeds this threshold, it indicates that the model's interpretability is acceptable. The preset mean squared error threshold refers to the pre-set passing standard for the mean squared error index. When the actual mean squared error value is lower than or equal to this threshold, it indicates that the model's prediction bias is acceptable. The second comparison result refers to the conclusion drawn after comparing the actual mean squared error value with the preset mean squared error threshold.

[0141] Understandably, the first comparison result and the second comparison result are obtained by comparing the actual goodness-of-fit value with the preset goodness-of-fit threshold and the actual mean square error value with the preset mean square error threshold, respectively.

[0142] The beneficial effect of this step is that by comparing with the preset standard, the model's qualification status in the two core accuracy indicators is clarified, providing a direct basis for subsequent model qualification determination.

[0143] A54: If both the first and second comparison results meet the judgment criteria, the fitting accuracy is determined to be up to standard, and the validated benchmark regression model is obtained. Understandably, when both the first and second comparison results meet the preset qualification requirements, the fitting accuracy of the initial draft of the benchmark regression model is deemed to have met the standard, and the initial draft is determined to be the validated benchmark regression model.

[0144] The benefit of this step is that it selects models with satisfactory fitting accuracy, ensuring that the final model used to calculate the inflection point parameters has reliable interpretability and prediction accuracy.

[0145] Furthermore, if the fitting accuracy is not up to standard, the model parameters of the initial draft of the benchmark regression model are adjusted and retrained until a validated benchmark regression model is obtained.

[0146] It should be noted that model parameters refer to the various adjustable coefficients in the benchmark regression model, including the coefficients of the linear term, the coefficients of the quadratic term, and the coefficients corresponding to the control variables.

[0147] Understandably, when the fitting accuracy is not up to standard, the model parameters of the initial draft of the baseline regression model are adjusted, and the model is retrained using the training dataset. The above verification process is repeated until a valid baseline regression model with satisfactory fitting accuracy is obtained.

[0148] The beneficial effect of this step is that, through the iterative process of parameter adjustment and retraining, the model accuracy is continuously optimized, avoiding poor fitting results caused by unreasonable model parameters, and ensuring that the final model meets the analysis requirements.

[0149] Step S305: Determine the inverted U-shaped extreme points and the inverted U-shaped inflection point parameters related to rural electricity consumption based on the coefficients of the first and second terms in the validated benchmark regression model.

[0150] It should be noted that the coefficient of the linear term is the weighting coefficient of the linear part of the core explanatory variables in the model, quantifying the strength and direction of the linear association. Furthermore, the coefficient of the quadratic term is the weighting coefficient of the quadratic term of the core explanatory variables, and is key to capturing inverted U-shaped associations. Further, the inverted U-shaped extreme points are the peak points of the inverted U-shaped curve, corresponding to the optimal state of the association. Also, the inverted U-shaped inflection point parameter is a critical value based on the extreme points, reflecting the boundary of the association's inflection point.

[0151] Understandably, by extracting the coefficients of the first and second terms of the validated benchmark regression model, the inverted U-shaped extreme points are determined, thereby obtaining the inverted U-shaped inflection point parameters related to rural electricity consumption. Accurately identifying key parameters through model coefficients ensures parameter accuracy, provides scientific support for subsequent interval definition, and improves the accuracy of the solution.

[0152] The beneficial effect of this step is that by using model coefficients with reliable fitting accuracy to calculate inflection point parameters, the correlation inflection point parameters can accurately reflect the critical value of the turning point of the core correlation, providing high-quality basic parameters for the subsequent determination of effective inflection point values.

[0153] This embodiment provides a method for determining and regulating reasonable electricity consumption ranges in rural areas. By constructing a benchmark regression model framework that includes regional and yearly fixed effects, the interference of regional differences and temporal commonalities is effectively eliminated. The division of labor between training and validation datasets ensures the objectivity of model training and testing. The model's linear and quadratic coefficients are optimized through training, and the model accuracy is verified using both goodness of fit and mean squared error. If the verification fails, parameters are iteratively adjusted and retrained to ensure the model possesses reliable interpretability and predictive accuracy. Based on the validated model coefficients, correlation inflection point parameters are calculated, significantly improving the accuracy and reliability of these parameters. This lays a solid model foundation for determining effective inflection point values ​​and defining reasonable electricity consumption ranges, enhancing the scientific rigor and feasibility of the entire rural electricity consumption regulation scheme.

[0154] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the method for determining and regulating the reasonable range of rural electricity use in this application. Any simple modifications based on this technical concept are within the scope of protection of this application.

[0155] This application also provides a device for determining and regulating reasonable electricity consumption ranges in rural areas. Please refer to [reference needed]. Figure 4 The device for determining and regulating reasonable electricity consumption ranges in rural areas includes: Data acquisition module 10 is used to acquire electricity consumption data and artificial intelligence application data of the target area; The standardization preprocessing module 20 is used to perform standardization preprocessing on electricity consumption data and artificial intelligence application data to obtain standardized data suitable for modeling. The parameter determination module 30 is used to construct a benchmark regression model based on standardized data, and to calculate the inverted U-shaped inflection point parameters of rural electricity consumption based on the coefficients of the first and second terms of the benchmark regression model. The parameter verification module 40 is used to perform robustness and endogeneity tests on the associated inflection point parameters based on rural communication infrastructure data, and to determine the effective inflection point values. The interval determination and instruction generation module 50 is used to determine the reasonable power consumption interval of the target area based on the effective inflection point value and the safe operation parameters of the rural power grid, so as to generate the target power consumption control instruction based on the reasonable power consumption interval and send it to the power grid terminal for control.

[0156] The rural electricity consumption reasonable range determination and control device provided in this application, employing the rural electricity consumption reasonable range determination and control method in the above embodiments, can solve the technical problem of how to identify the rural electricity consumption reasonable range corresponding to artificial intelligence applications and prevent grid overload by actively controlling the degree of AI penetration. Compared with the prior art, the beneficial effects of the rural electricity consumption reasonable range determination and control device provided in this application are the same as those of the rural electricity consumption reasonable range determination and control method provided in the above embodiments, and other technical features in the rural electricity consumption reasonable range determination and control device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0157] In one embodiment, the standardization preprocessing module 20 is further used to identify outliers in the target area's electricity consumption data and artificial intelligence application data using the interquartile range method, and to obtain data labeled with outliers; Remove outliers from the data labeled with outliers to obtain preliminary data after outlier removal; By using interpolation to fill in the missing items in the preliminary data after removing outliers, complete data without missing values ​​is obtained. Perform a logarithmic transformation on continuous variables in complete, unmissing data to obtain logarithmically processed data; Perform standardization on the logarithmically processed data to obtain standardized intermediate data; Redundant variables in the standardized intermediate data are removed to obtain standardized data suitable for modeling.

[0158] In one embodiment, the parameter determination module 30 is further used to construct an initial framework for a benchmark regression model that includes regional fixed effects and year fixed effects, thereby obtaining the initial framework for the benchmark regression model. The standardized data is divided into training and validation datasets. The initial framework of the model was trained using the training dataset, resulting in a preliminary draft of the benchmark regression model containing the optimized coefficients of the first and second terms. The initial draft of the benchmark regression model was validated using a validation dataset, resulting in a validated benchmark regression model. Based on the coefficients of the first and second terms in the validated benchmark regression model, the inverted U-shaped extreme points and the inverted U-shaped inflection point parameters related to rural electricity consumption are determined.

[0159] In one embodiment, the parameter determination module 30 is further configured to input the verification dataset into the initial draft of the benchmark regression model based on the goodness-of-fit index and the mean squared error index as judgment indicators, and calculate the actual goodness-of-fit value and the actual mean squared error value. The first comparison result is obtained by comparing the actual goodness-of-fit value with the preset goodness-of-fit threshold. The second comparison result is obtained by comparing the actual mean square error value with the preset mean square error threshold. If both the first and second comparison results meet the criteria, the fitting accuracy is deemed satisfactory, and the validated benchmark regression model is obtained.

[0160] In one embodiment, the parameter verification module 40 is further configured to perform a robustness test on the associated inflection point parameters using the time lag method to obtain the first robustness test result. The robustness of the correlation inflection point parameters is tested twice using the sample tail reduction method, and the results of the second robustness test are obtained. Based on the core explanatory variables in the benchmark regression model, we determine whether there is interference from omitted variables in the association between the core explanatory variables and the explained variable, and obtain the endogeneity determination result. Based on the endogeneity determination results, rural communication infrastructure data was used as an instrumental variable to correct for interference, and the correlation inflection point parameters after eliminating endogeneity were obtained. The effective inflection point value is determined by combining the results of the first robustness test, the results of the second robustness test, and the correlation inflection point parameters after eliminating endogeneity.

[0161] In one embodiment, the parameter verification module 40 is further used to extract the first-period lag term and the second-period lag term of the artificial intelligence application-related data from the standardized data, and determine them as the first instrumental variable and the second instrumental variable, respectively. Rural communication infrastructure data were obtained from regional statistical data, and the rural communication infrastructure data was combined with the first-period lagged terms to construct an interaction term, which was used as a third instrumental variable. Validity tests were performed on the first, second, and third instrumental variables to obtain the instrumental variable validity test results. When the instrumental variable validity test results show that the first, second, and third instrumental variables all pass the validity test, the parameters of the benchmark regression model are re-estimated using the two-step optimal generalized moment estimation method to obtain the re-estimated correlation inflection point parameters. The re-estimated correlation inflection point parameters are compared with the original correlation inflection point parameters to confirm that the parameter consistency is within the preset error range, thus obtaining the correlation inflection point parameters after endogeneity processing.

[0162] In one embodiment, the interval determination and instruction generation module 50 is further configured to determine the upper and lower limits of the reasonable power consumption range of the target area by taking the effective inflection point value as the core anchor point and combining the distribution range of artificial intelligence application-related data in the target area sample, so as to obtain the reasonable power consumption range of the target area. Using the safe operation parameters of the rural power grid as constraints, the reasonable range of electricity consumption is checked and corrected so that the active power of the transformer area and the node voltage corresponding to the boundary of the range are within the safe operation range, thus obtaining the corrected reasonable range of electricity consumption. Based on the rural production and rural living electricity consumption scenarios in the target area, specific electricity consumption control measures are formulated to adapt to the adjusted reasonable electricity consumption range, resulting in scenario-adapted electricity consumption control measures. The reasonable range of electricity consumption in the target area and the electricity consumption control measures adapted to the scenario are converted into an instruction format that can be recognized by the power grid control terminal in the target area, and the initial draft of the instruction is obtained. The initial draft of the instruction is optimized into a format that conforms to the power grid dispatching protocol, resulting in the target power consumption control instruction, which is then sent to the power grid terminal for control.

[0163] In one embodiment, the interval determination and instruction generation module 50 is also used to monitor the running status of the artificial intelligence application model in the target area in real time; When an abnormality is detected in the operation status, it is determined that there is a failure in the artificial intelligence application model, and the preset emergency inflection point database is activated; Based on the current core production electricity consumption type in the target area, match the target's historical valid inflection point sample values ​​from the emergency inflection point database; Based on the emergency inflection point value, combined with the rural power grid safety operation parameters and core production electricity demand thresholds in the target area, a reasonable range of electricity consumption under the current fault scenario is determined. Based on the reasonable range of electricity consumption, the priority of ensuring core production electricity consumption and the proportion of temporary power restriction for non-core electricity consumption are determined, and emergency power control instructions for fault scenarios are obtained. Emergency power control commands for fault scenarios are pushed to the power grid control terminal for load allocation adjustment, and emergency power control commands for fault scenarios are pushed to the electricity terminals of farmers to cut off power or reduce load in the emergency power restriction section.

[0164] In one embodiment, the interval determination and instruction generation module 50 is further used to perform power grid constraint adaptation correction on the target historical valid inflection point sample value to obtain the corrected emergency inflection point value; The steps for performing grid constraint adaptation correction on the target historical effective inflection point sample values ​​to obtain the corrected emergency inflection point values ​​include: Obtain operational status information of the power grid in the target area; Based on the safe operation parameters of rural power grid, it is determined whether the electricity load level corresponding to the target historical effective inflection point sample value is suitable for the current power grid operation status, and the suitability judgment result is obtained. When the compatibility determination result is that the compatibility is not suitable, the target historical effective inflection point sample value is adjusted so that the power load level corresponding to the adjusted target historical effective inflection point sample value meets the safe operation parameters of the rural power grid. The adjusted historical effective inflection point sample value is determined as the corrected emergency inflection point value.

[0165] This application provides a device for determining and regulating reasonable ranges of rural electricity consumption. The device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method for determining and regulating reasonable ranges of rural electricity consumption in the above embodiment 1.

[0166] The following is for reference. Figure 5 The diagram illustrates a structural schematic of a device suitable for determining and regulating reasonable rural electricity consumption ranges in the embodiments of this application. The device for determining and regulating reasonable rural electricity consumption ranges in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), vehicle terminals (e.g., vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 5 The illustrated equipment for determining and regulating reasonable rural electricity consumption ranges is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments in this application.

[0167] like Figure 5As shown, the rural electricity consumption rational range determination and control equipment may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to the program stored in ROM (Read Only Memory) 1002 or the program loaded from storage device 1003 into RAM (Random Access Memory) 1004. RAM 1004 also stores various programs and data required for the operation of the rural electricity consumption rational range determination and control equipment. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. Input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the rural electricity consumption rational range determination and control equipment to exchange data wirelessly or via wired communication with other devices. Although the rural electricity consumption rational range determination and control equipment with various systems is shown in the figure, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems can be implemented alternatively.

[0168] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0169] The rural electricity consumption reasonable range determination and control device provided in this application, employing the rural electricity consumption reasonable range determination and control method in the above embodiments, can solve the technical problem of how to identify the rural electricity consumption reasonable range corresponding to artificial intelligence applications and prevent grid overload by actively controlling the degree of AI penetration. Compared with the prior art, the beneficial effects of the rural electricity consumption reasonable range determination and control device provided in this application are the same as those of the rural electricity consumption reasonable range determination and control method provided in the above embodiments, and other technical features in this rural electricity consumption reasonable range determination and control device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0170] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0171] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0172] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the method for determining and controlling reasonable ranges of rural electricity consumption in the above embodiments.

[0173] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable Read Only Memory or Flash Memory), optical fibers, CD-ROM (CD-Read Only Memory), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0174] The aforementioned computer-readable storage medium may be included in the rural electricity consumption reasonable range determination and control equipment; or it may exist independently and not be assembled into the rural electricity consumption reasonable range determination and control equipment.

[0175] The aforementioned computer-readable storage medium carries one or more programs. When these programs are executed by the rural electricity consumption reasonable range determination and control equipment, the rural electricity consumption reasonable range determination and control equipment: acquires electricity consumption data and artificial intelligence application data of the target area; performs standardized preprocessing on the electricity consumption data and artificial intelligence application data to obtain standardized data for adaptive modeling; constructs a benchmark regression model based on the standardized data, and calculates the inverted U-shaped inflection point parameters of rural electricity consumption correlation using the coefficients of the first and second terms of the benchmark regression model; performs robustness and endogeneity tests on the correlation inflection point parameters based on rural communication infrastructure data to determine the effective inflection point value; and determines the reasonable range of electricity consumption in the target area based on the effective inflection point value and rural power grid safety operation parameters, so as to generate target electricity consumption control instructions and send them to the power grid terminal for control.

[0176] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including LAN (Local Area Network) or WAN (Wide Area Network)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0177] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0178] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0179] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described method for determining and controlling reasonable rural electricity consumption ranges. This solves the technical problem of how to identify reasonable rural electricity consumption ranges corresponding to artificial intelligence applications and prevent grid overload by actively controlling the degree of AI penetration. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the method for determining and controlling reasonable rural electricity consumption ranges provided in the above embodiments, and will not be repeated here.

[0180] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method for determining and regulating reasonable ranges of rural electricity consumption.

[0181] The computer program product provided in this application can solve the technical problem of how to identify reasonable rural electricity consumption ranges corresponding to artificial intelligence applications and prevent power grid overload by actively controlling the degree of AI penetration. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the rural electricity consumption reasonable range determination and control method provided in the above embodiments, and will not be repeated here.

[0182] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for determining and regulating a reasonable range of rural electricity consumption, characterized in that, The method includes: Acquire electricity consumption data and artificial intelligence application data for the target area; The electricity consumption data and the artificial intelligence application data are preprocessed to obtain standardized data for adaptive modeling. A benchmark regression model is constructed based on the standardized data. The inverted U-shaped inflection point parameters related to rural electricity consumption are calculated based on the coefficients of the first and second terms of the benchmark regression model. Based on rural communication infrastructure data, robustness and endogeneity tests are performed on the associated inflection point parameters to determine effective inflection point values. Based on the effective inflection point value and the safe operation parameters of the rural power grid, a reasonable range of electricity consumption for the target area is determined, and a target electricity consumption control command is generated based on the reasonable range of electricity consumption and sent to the power grid terminal for control.

2. The method as described in claim 1, characterized in that, The step of performing standardized preprocessing on the electricity consumption data and the artificial intelligence application data to obtain standardized data for adaptive modeling includes: The interquartile range method is used to identify outliers in the electricity consumption data of the target area and the artificial intelligence application data, and data with labeled outliers is obtained. Remove the outliers from the labeled outlier data to obtain the preliminary data after outlier removal; The missing items in the preliminary data after outlier removal are filled by interpolation to obtain complete data without missing values. Perform a logarithmic transformation on the continuous variables in the complete data without missing data to obtain logarithmic processed data; The logarithmically processed data is then standardized to obtain standardized intermediate data. Redundant variables in the standardized intermediate data are removed to obtain standardized data suitable for modeling.

3. The method as described in claim 1, characterized in that, The step of constructing a benchmark regression model based on the standardized data, and calculating the inverted U-shaped inflection point parameters related to rural electricity consumption based on the coefficients of the first and second terms of the benchmark regression model, includes: An initial framework for a benchmark regression model, including regional and year-fixed effects, is constructed to obtain the initial framework for the benchmark regression model. The standardized data is divided into a training dataset and a validation dataset; The initial framework of the model is trained using the training dataset to obtain a preliminary draft of the benchmark regression model containing the optimized coefficients of the first and second terms. The initial draft of the benchmark regression model was validated using the validation dataset, resulting in a validated benchmark regression model. Based on the coefficients of the first and second terms in the validated benchmark regression model, the inverted U-shaped extreme points and the inverted U-shaped inflection point parameters related to rural electricity consumption are determined.

4. The method as described in claim 3, characterized in that, The step of validating the initial draft of the benchmark regression model using the validation dataset to obtain a validated benchmark regression model includes: Based on the goodness-of-fit index and the mean squared error index as the judgment index, the validation dataset is input into the initial draft of the benchmark regression model to calculate the actual goodness-of-fit value and the actual mean squared error value. The actual goodness-of-fit value is compared with the preset goodness-of-fit threshold to obtain the first comparison result; By comparing the actual mean square error value with the preset mean square error threshold, a second comparison result is obtained; If both the first comparison result and the second comparison result meet the judgment criteria, the fitting accuracy is determined to be up to standard, and a validated benchmark regression model is obtained.

5. The method as described in claim 1, characterized in that, The step of performing robustness and endogeneity tests on the correlation inflection point parameters based on rural communication infrastructure data to determine effective inflection point values ​​includes: The robustness of the correlation inflection point parameters is tested using the time lag method, and the first robustness test result is obtained. The robustness of the correlation inflection point parameters is tested twice using the sample tail reduction method to obtain the second robustness test result. Based on the core explanatory variables in the benchmark regression model, determine whether there is interference from omitted variables in the association between the core explanatory variables and the explained variable, and obtain the endogeneity determination result; Based on the endogeneity determination results, rural communication infrastructure data is used as an instrumental variable to correct for interference, and the correlation inflection point parameters after eliminating endogeneity are obtained. By combining the results of the first robustness test, the second robustness test, and the correlation inflection point parameters after eliminating endogeneity, an effective inflection point value is determined.

6. The method as described in claim 5, characterized in that, The step of correcting interference using rural communication infrastructure data as an instrumental variable based on the endogeneity determination result to obtain the correlation inflection point parameters after eliminating endogeneity includes: The first-phase and second-phase lagged terms of the artificial intelligence application-related data are extracted from the standardized data and determined as the first instrumental variable and the second instrumental variable, respectively. Rural communication infrastructure data is obtained from regional statistical data, and the rural communication infrastructure data is combined with the first-period lagged term to construct an interaction term, which is used as a third instrumental variable; Perform validity tests on the first instrumental variable, the second instrumental variable, and the third instrumental variable to obtain the instrumental variable validity test results; When the validity test results of the instrumental variables show that the first instrumental variable, the second instrumental variable, and the third instrumental variable all pass the validity test, the parameters of the benchmark regression model are re-estimated using the two-step optimal generalized moment estimation method to obtain the re-estimated correlation inflection point parameters. The re-estimated correlation inflection point parameters are compared with the original correlation inflection point parameters to confirm that the parameter consistency is within the preset error range, thus obtaining the correlation inflection point parameters after endogeneity processing.

7. The method as described in claim 1, characterized in that, The step of determining the reasonable electricity consumption range of the target area based on the effective inflection point value and rural power grid safety operation parameters, and generating a target electricity consumption control command based on the reasonable electricity consumption range and issuing it to the power grid terminal for control, includes: Using the effective inflection point value as the core anchor point, and combining the distribution range of artificial intelligence application-related data in the target area sample, the upper and lower limits of the reasonable electricity consumption range of the target area are determined, and the reasonable electricity consumption range of the target area is obtained. Using the safe operation parameters of the rural power grid as constraints, the reasonable power consumption range is checked and corrected so that the active power of the transformer area and the node voltage corresponding to the boundary of the range are within the safe operation range, thus obtaining the corrected reasonable power consumption range; Based on the rural production and residential electricity consumption scenarios in the target area, specific electricity control measures adapted to the revised reasonable electricity consumption range are formulated to obtain scenario-adapted electricity control measures. The reasonable power consumption range of the target area and the power consumption control measures adapted to the scenario are converted into an instruction format that can be recognized by the power grid control terminal of the target area to obtain the initial instruction draft; The initial draft of the instruction is optimized into a format that conforms to the power grid dispatching protocol to obtain the target power consumption control instruction, which is then sent to the power grid terminal for control.

8. The method as described in claim 1, characterized in that, The method further includes: Real-time monitoring of the operational status of the artificial intelligence application model in the target area; When an abnormality is detected in the operating status, it is determined that an artificial intelligence application model failure has occurred, and a preset emergency inflection point database is activated; Based on the current core production electricity consumption type in the target area, match the target historical valid inflection point sample value from the emergency inflection point database; Based on the emergency inflection point value, and combined with the rural power grid safety operation parameters and core production electricity demand thresholds of the target area, a reasonable range of electricity consumption under the current fault scenario is determined. Based on the reasonable power consumption range, the priority of ensuring core production power consumption and the temporary power restriction ratio of non-core power consumption are determined, and emergency power control instructions for fault scenarios are obtained. The emergency power control command for the fault scenario is pushed to the power grid control terminal for load allocation adjustment, and the emergency power control command for the fault scenario is pushed to the household power terminal for power outage or load reduction in the emergency power restriction section.

9. The method as described in claim 8, characterized in that, After the step of matching the target historical valid inflection point sample value from the emergency inflection point database according to the current core production power consumption type of the target area, the method further includes: The target historical valid inflection point sample values ​​are modified by power grid constraint adaptation to obtain the modified emergency inflection point values. The step of performing grid constraint adaptation correction on the target historical valid inflection point sample values ​​to obtain the corrected emergency inflection point values ​​includes: Obtain the operating status information of the power grid in the target area; Based on the rural power grid safety operation parameters, determine whether the electricity load level corresponding to the target historical effective inflection point sample value is suitable for the current power grid operation state, and obtain the suitability determination result. When the compatibility determination result is that the compatibility is not suitable, the target historical effective inflection point sample value is adjusted so that the power load level corresponding to the adjusted target historical effective inflection point sample value meets the rural power grid safe operation parameters. The adjusted target historical effective inflection point sample value is determined as the corrected emergency inflection point value.

10. A device for determining and regulating a reasonable range of rural electricity consumption, characterized in that, The device includes: The data acquisition module is used to acquire electricity consumption data and artificial intelligence application data for the target area; A standardized preprocessing module is used to perform standardized preprocessing on the electricity consumption data and the artificial intelligence application data to obtain standardized data adapted for modeling; The parameter determination module is used to construct a benchmark regression model based on the standardized data, and to calculate the inverted U-shaped inflection point parameters related to rural electricity consumption based on the coefficients of the first and second terms of the benchmark regression model. The parameter verification module is used to perform robustness and endogeneity tests on the correlation inflection point parameters based on rural communication infrastructure data, and to determine the effective inflection point values. The interval determination and instruction generation module is used to determine the reasonable power consumption interval of the target area based on the effective inflection point value and the rural power grid safety operation parameters, so as to generate a target power consumption control instruction based on the reasonable power consumption interval and send it to the power grid terminal for control.

Images