A Time-of-Use Electricity Decomposition Method Based on Sensitivity Analysis

Abstract

The present invention proposes a time-sharing power consumption decomposition method based on sensitivity analysis, which relates to the technical field of power consumption decomposition. Historical and real-time time-sharing power consumption data and time-sharing meteorological data of a target area are obtained, and after grid processing, meteorological factors and a joint data set are obtained, and a set of key meteorological factors is screened out; a basic time-sharing power consumption decomposition model of quantile regression is constructed with historical time-sharing power consumption data, real-time data is input to obtain basic components and time-sharing power consumption residuals, and the residuals are subjected to secondary modal decomposition to obtain a set of meteorologically sensitive modal components; an MOEA / D multi-objective optimization model is constructed, with the set of meteorologically sensitive modal components as the input, solved by the Chebyshev decomposition method, and the weights are updated based on the key meteorological factors to obtain corrected meteorological power components; finally, the basic components and the corrected components are superimposed to obtain the final decomposition result. The present invention improves the accuracy of power consumption decomposition through the combination of multiple models, providing strong support for power dispatching and demand forecasting.

Description

Technical Field

[0001] The present invention relates to a method for decomposing time - of - use electricity consumption based on sensitivity analysis, belonging to the technical field of electricity consumption decomposition. Background Art

[0002] In the field of power load forecasting and management, the impact of meteorological factors on electricity demand has been attracting increasing attention. In recent years, with the progress of data analysis technology and meteorology, researchers have begun to explore incorporating meteorological factors into electricity consumption decomposition and forecasting models to improve the accuracy and reliability of forecasting.

[0003] For example, the Chinese invention patent with the publication number CN109299814A discloses a systematic method for predicting the electricity consumption affected by meteorology. This method systematically analyzes the impact of meteorological factors on power load through steps such as setting a base month, calculating meteorological correlation, and the growth rate of meteorology - affected electricity consumption. Specifically, this method divides a year into winter and summer, calculates the correlation respectively, and derives the importance of meteorological factors based on the obtained data. This process helps to significantly improve the accuracy of load forecasting and reduce errors.

[0004] However, the above - mentioned patent uses seasonal division (winter / summer) and linear growth rate calculation methods, which can only handle the linear relationship between a single meteorological factor and electricity consumption, and cannot effectively capture the non - linear coupling effects among multiple meteorological factors (temperature, humidity, wind speed, etc.). Its screening method based on the Pearson correlation coefficient is difficult to identify the combination of meteorological factors with significant interaction effects, resulting in limited prediction accuracy of the model under complex climate conditions. Summary of the Invention

[0005] In order to solve the problems existing in the above - mentioned prior art, the present invention proposes a method for decomposing time - of - use electricity consumption based on sensitivity analysis.

[0006] The technical solution of the present invention is as follows:

[0007] The present invention provides a method for decomposing time - of - use electricity consumption based on sensitivity analysis, and the method includes:

[0008] Obtain the time - of - use electricity consumption - related data of the target area, including historical time - of - use electricity consumption data, real - time time - of - use electricity consumption data, and time - of - use meteorological data, and perform grid processing on the time - of - use electricity consumption - related data to obtain meteorological factors and a time - of - use electricity consumption - meteorology joint dataset;

[0009] Calculate the sensitivity index of each meteorological factor to the time - of - use electricity consumption based on the time - of - use electricity consumption - meteorology joint dataset, and screen out the meteorological factors with sensitivity indexes greater than the sensitivity threshold to obtain a set of key meteorological factors;

[0010] Construct a basic time-sharing power consumption decomposition model based on historical time-sharing power consumption data, input the real-time time-sharing power consumption data into the basic time-sharing power consumption decomposition model to obtain basic time-sharing power consumption components; obtain time-sharing power consumption residuals based on the basic time-sharing power consumption components;

[0011] Use the improved complete ensemble empirical mode decomposition with adaptive noise algorithm ICEEMDAN to perform secondary mode decomposition on the time-sharing power consumption residuals to obtain a set of meteorological sensitive mode components;

[0012] Construct a MOEA / D multi-objective optimization model, use the set of meteorological sensitive mode components as input, and use the Chebyshev decomposition method to solve the MOEA / D multi-objective optimization model to obtain corrected meteorological power consumption components, where the MOEA / D multi-objective optimization model updates the weight vector based on a set of key meteorological factors;

[0013] Superimpose the basic time-sharing power consumption components and the corrected meteorological power consumption components to obtain the final time-sharing power consumption decomposition result.

[0014] As a preferred embodiment of the present invention, the time-sharing meteorological data includes temperature, humidity, wind speed, sunshine intensity, and precipitation.

[0015] As a preferred embodiment of the present invention, the specific process of grid processing the time-sharing power consumption related data to obtain meteorological factors and a time-sharing power consumption - meteorology joint dataset is as follows:

[0016] Perform grid processing on the time-sharing meteorological data according to the city dimension to obtain grid meteorological indicators, that is, meteorological factors; perform grid processing on the historical time-sharing power consumption data and the real-time time-sharing power consumption data according to the city dimension and the industry dimension, and combine the meteorological factors to construct a time-sharing power consumption - meteorology joint dataset, where the grid resolution is 1km×1km;

[0017] Perform data preprocessing on the time-sharing power consumption - meteorology joint dataset, including data alignment, outlier removal, and feature standardization.

[0018] As a preferred embodiment of the present invention, the specific process of calculating the sensitivity index of each meteorological factor to the time-sharing power consumption based on the time-sharing power consumption - meteorology joint dataset is as follows:

[0019] Calculate the global sensitivity corresponding to the time-sharing power consumption - meteorology joint dataset through the Sobol index, which is expressed by the formula:

[0020] ;

[0021] In the formula, is the global sensitivity of the th meteorological factor; is the th meteorological factor; is other meteorological factors except ; is variance; is conditional expectation; is the variable of time-of-use electricity output; is when the th meteorological factor takes a value, the expectation of the time-of-use electricity output variable about other meteorological factors except ; is the expectation about the th meteorological factor to calculate variance; is the contrast of the time-of-use electricity output variable ; ;

[0022] Use kernel density estimation to calculate the mutual information between meteorological factors and time-of-use electricity, which is expressed by the formula:

[0023] ;

[0024] In the formula, is the and joint probability density; and are respectively the and marginal probability densities; and are respectively the and specific values;

[0025] Based on global sensitivity and mutual information to calculate the sensitivity index, which is expressed by the formula:

[0026] ;

[0027] In the formula, is the sensitivity index of the th meteorological factor; is the weight coefficient; is the information entropy of time-of-use electricity.

[0028] As a preferred embodiment of the present invention, specifically constructing a basic time-of-use electricity decomposition model based on historical time-of-use electricity data is:

[0029] The basic time-of-use electricity decomposition model is a quantile regression model, which is expressed by the formula:

[0030] ;

[0031] Wherein, is the basic time-of-use electricity quantity with the time-of-day percentile being During the training process, is the historical time-of-use electricity quantity with the time-of-day percentile being During real-time monitoring, is the real-time time-of-use electricity quantity with the time-of-day percentile being ; and are the regression coefficients with the time-of-day percentile being ; and respectively represent the th city and the th industry; and are the total number of cities and the total number of industries respectively; is the variable coefficient of the th city with the time-of-day percentile being ; is the variable coefficient of the th industry with the time-of-day percentile being ; is the error term with the time-of-day percentile being ; is the percentile;

[0032] Based on the historical time-of-use electricity quantity data, the basic time-of-use electricity quantity decomposition model is trained to obtain a trained basic time-of-use electricity quantity decomposition model, and the real-time time-of-use electricity quantity data is input into the basic time-of-use electricity quantity decomposition model to obtain the real-time basic time-of-use electricity quantity component.

[0033] As a preferred embodiment of the present invention, the time-of-use electricity quantity residual is obtained based on the basic time-of-use electricity quantity component and is expressed by the formula:

[0034] ;

[0035] Wherein, is the time-of-use electricity quantity residual at time is the real-time time-of-use electricity quantity at time

[0036] As a preferred embodiment of the present invention, the improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) algorithm is used to perform a secondary mode decomposition on the time-sharing electricity residual to obtain a set of meteorologically sensitive mode components, specifically as follows:

[0037] Perform the first ICEEMDAN decomposition on the time-sharing electricity residual, which is decomposed into intrinsic mode functions and a residual component, and is expressed by the formula:

[0038] ;

[0039] In the formula, is the th intrinsic mode function obtained from the first decomposition; is the residual component obtained from the first decomposition;

[0040] Select all the intrinsic mode functions obtained from the first decomposition for the second decomposition, and is expressed by the formula:

[0041] ;

[0042] In the formula, represents the th intrinsic mode function obtained by performing the second decomposition on , is the total number of intrinsic mode functions obtained from the second decomposition; is the residual component obtained from the second decomposition;

[0043] Merge all the intrinsic mode functions obtained from the second decomposition to obtain a set of meteorologically sensitive mode components , and is expressed by the formula:

[0044] .

[0045] As a preferred embodiment of the present invention, the constructed Multi-Objective Evolutionary Algorithm Based on Decomposition (MOEA / D) multi-objective optimization model is expressed by the formula:

[0046] ;

[0047] In the formula, is the objective of minimizing the residual fitting error; is the th non-linear mapping function of the th meteorological factor at time is the length of the real-time time-sharing electricity signal; is the th weight coefficient of the is the objective of maximizing the consistency between the weight and the sensitivity coefficient; is the total number of meteorological factors.

[0048] As a preferred embodiment of the present invention, the MOEA / D multi-objective optimization model is solved by using the Chebyshev decomposition method, specifically as follows:

[0049] Define the Chebyshev distance , which is expressed by the formula:

[0050] ;

[0051] In the formula, is the decision variable, used to represent the weight coefficient of the th meteorological factor; is the multi-objective trade-off weight set, ; is the objective index; is the ideal point set, ; is the objective function value corresponding to the decision variable ;

[0052] Generate a set of uniformly distributed initial weight vectors according to the key meteorological factor set, which is expressed by the formula:

[0053] ;

[0054] In the formula, is the set of initial weight vectors; is the bit index of the initial weight vector; is the total number of initial weight vectors, and each initial weight vector corresponds to a sub-problem;

[0055] Calculate the distance between the weight vector in the set of initial weight vectors and other weight vectors, and select the initial weight vectors that meet the distance threshold to form the corresponding neighborhood ;

[0056] Randomly generate the initial population set , which is expressed by the formula:

[0057] ;

[0058] In the formula, is the solution corresponding to the initial weight vector ;

[0059] Randomly select two solutions from the neighborhood , perform genetic operations on the two selected solutions to obtain new solutions, and calculate the objective function values of the new solutions;

[0060] Iterative solution is carried out, including calculating the Chebyshev distances of the current solution and the new solution with respect to the multi-objective trade-off weights and the ideal point for each sub-problem in the neighborhood in each iteration. If the Chebyshev distance corresponding to the new solution is less than or equal to the Chebyshev distance corresponding to the current solution, the current solution is replaced by the new solution;

[0061] If the objective function value is less than the ideal point, the ideal point is equal to the objective function value;

[0062] Repeat the above iterative solution process until the preset termination condition is met, then stop the iteration. At this time, the solutions in the population are the approximate optimal solutions of the MOEA / D multi-objective optimization model, and the corrected meteorological electricity components are obtained based on the approximate optimal solutions.

[0063] As a preferred embodiment of the present invention, the final time-of-use electricity decomposition result is expressed by the formula:

[0064] ;

[0065] In the formula, is the final time-of-use electricity decomposition result; is the corrected meteorological electricity component.

[0066] The present invention has the following beneficial effects:

[0067] 1. The present invention is a time-of-use electricity decomposition method based on sensitivity analysis. By calculating the sensitivity indexes of each meteorological factor to the time-of-use electricity and screening out the key meteorological factors with sensitivity indexes greater than the threshold, it focuses on the meteorological factors with greater influence on the time-of-use electricity, reduces the interference of irrelevant information, improves the calculation efficiency and accuracy of the subsequent model, and enables the model to decompose the electricity more pertinently;

[0068] 2. The present invention is a time-of-use electricity decomposition method based on sensitivity analysis. It uses historical time-of-use electricity data to construct a basic time-of-use electricity decomposition model (quantile regression model), and inputs real-time time-of-use electricity data into the model to obtain basic time-of-use electricity components. The quantile regression model can consider the electricity distribution under different quantiles, comprehensively reflect the change characteristics of electricity, and calculate the basic electricity components more accurately;

[0069] 3. The present invention is a time-of-use electricity decomposition method based on sensitivity analysis. The time-of-use electricity residuals are subjected to secondary modal decomposition using an improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) to obtain a set of meteorological sensitive modal components, further separating and refining the meteorological sensitive information in the residuals, excavating the meteorological influencing factors hidden in the residuals, providing more detailed information for subsequent correction of meteorological electricity components, and helping to improve the accuracy of electricity decomposition;

[0070] 4. The present invention relates to a time-sharing electricity decomposition method based on sensitivity analysis, which constructs a MOEA / D multi-objective optimization model, solves it using the Chebyshev decomposition method, updates the weight vector based on the key meteorological factor set, and thus obtains the corrected meteorological electricity component. The MOEA / D multi-objective optimization model comprehensively considers two objectives: minimizing the residual fitting error and maximizing the consistency between the weight and the sensitivity coefficient, and can balance and optimize among multiple objectives, making the corrected meteorological electricity component more in line with the actual situation and further improving the accuracy of electricity decomposition. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 It is a flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0072] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0073] It should be understood that the step numbers used in the text are only for convenient description and do not limit the execution order of the steps.

[0074] It should be understood that the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0075] The terms "comprising" and "including" indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their combinations.

[0076] The term " / and" refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0077] Embodiment 1:

[0078] Refer to Figure 1 , this embodiment provides a time-sharing electricity decomposition method based on sensitivity analysis, including the following steps:

[0079] S1. Obtain the time-of-use electricity consumption related data of the target area through the marketing business system and the data middle platform, including historical time-of-use electricity consumption data, real-time time-of-use electricity consumption data, and time-of-use meteorological data. Perform grid processing on the time-of-use electricity consumption related data to obtain meteorological factors and a time-of-use electricity consumption - meteorology joint dataset;

[0080] S11. The time-of-use meteorological data includes temperature, humidity, wind speed, sunshine intensity, and precipitation;

[0081] S12. Perform grid processing on the time-of-use meteorological data by city dimension to obtain grid meteorological indicators, that is, meteorological factors; perform grid processing on the historical time-of-use electricity consumption data and the real-time time-of-use electricity consumption data by city dimension and industry dimension, and combine the meteorological factors to construct a time-of-use electricity consumption - meteorology joint dataset, where the grid resolution is 1km×1km;

[0082] Furthermore, in this embodiment, the city dimension is divided according to administrative regions; the industry dimension includes industry, commerce, agriculture, animal husbandry, fishery, and urban and rural residents classification;

[0083] S13. Perform data preprocessing on the time-of-use electricity consumption - meteorology joint dataset, including data alignment, outlier removal, and feature standardization, and use the preprocessed time-of-use electricity consumption - meteorology joint dataset for subsequent decomposition calculations;

[0084] S2. Calculate the sensitivity index of each meteorological factor to the time-of-use electricity consumption based on the time-of-use electricity consumption - meteorology joint dataset, and screen the meteorological factors with sensitivity indexes greater than the sensitivity threshold to obtain a set of key meteorological factors;

[0085] S21. Calculate the global sensitivity corresponding to the time-of-use electricity consumption - meteorology joint dataset through the Sobol index, which is expressed by the formula:

[0086] ;

[0087] In the formula, is the global sensitivity of the th meteorological factor; is the th meteorological factor; is other meteorological factors except ; is the variance; is the conditional expectation; is the time-of-use electricity consumption output variable; is when the value of the th meteorological factor is given, the time-of-use electricity consumption output variable about other meteorological factors except ; is the expectation; To calculate the variance of the ith meteorological factor; To calculate the contrast of the output variable of time - of - use electricity ;

[0088] Calculate the mutual information between meteorological factors and time - of - use electricity using kernel density estimation , which is expressed by the formula:

[0089] ;

[0090] In the formula, is the joint probability density of and ; and are respectively the marginal probability densities of and ; and are respectively the specific values of

[0091] Based on global sensitivity and mutual information calculate the sensitivity index, which is expressed by the formula:

[0092] ;

[0093] In the formula, is the sensitivity index of the ith meteorological factor; is the weight coefficient; is the information entropy of time - of - use electricity;

[0094] S22. The sensitivity threshold is dynamically determined through Monte Carlo cross - validation, which is a conventional technical means in this field and will not be elaborated here;

[0095] S3. Based on historical time - of - use electricity data, construct a basic time - of - use electricity decomposition model. Input real - time time - of - use electricity data into the basic time - of - use electricity decomposition model to obtain basic time - of - use electricity components; Obtain time - of - use electricity residuals based on the basic time - of - use electricity components;

[0096] S31. In this embodiment, historical time - of - use electricity data for winter (December - February) and summer (June - August) can be further selected to establish quantile regression models for each of the 24 time - of - use points in a day, which is the basic time - of - use electricity decomposition model, and is expressed by the formula:

[0097] ; ​

[0098] In the formula, is the basic time-of-use electricity at the time-of-use quantile; during the training process, is the historical time-of-use electricity at the time-of-use quantile; during real-time monitoring, is the real-time time-of-use electricity at the time-of-use quantile; and are the regression coefficients at the time-of-use quantile; and respectively represent the th city and the th industry; and are respectively the total number of cities and the total number of industries; is the variable coefficient of the th city at the time-of-use quantile; is the variable coefficient of the th industry at the time-of-use quantile; is the error term at the time-of-use quantile; is the quantile;

[0099] The reasons for choosing winter and summer include: 1) These two seasons are the seasons with the most obvious temperature fluctuations, and the impact of temperature and humidity on electricity consumption is particularly prominent; 2) Temperature and electricity consumption show negative and positive correlations respectively in these two seasons, which are representative;

[0100] The selection of the quantile is also crucial. Different quantiles result in very different regression results. Specifically:

[0101] 1) When = 0.1 and = 0.9, it is used to capture extreme electricity consumption scenarios:

[0102] = 0.1: Reflects the minimum electricity demand under extreme low temperature or low humidity conditions (such as winter cold snaps, summer rainstorms), corresponding to the "guaranteed" load of the power system;

[0103] = 0.9: Reflects the peak electricity demand under extreme high temperature or high humidity conditions (such as in the sweltering heat of summer or the peak heating period in winter), corresponding to the "overload" risk point of the power grid;

[0104] When = 0.25 and = 0.75, analyze the conventional fluctuation range:

[0105] = 0.25: Characterizes the lower interval of the electricity quantity distribution and can be used to evaluate the base load under mild weather conditions;

[0106] = 0.75: Characterizes the higher interval of the electricity quantity distribution, corresponding to the electricity consumption growth during common high temperatures or rising humidity;

[0107] When = 0.5 (median), it is used for robust base electricity estimation:

[0108] = 0.5, as the median of the electricity quantity distribution, is insensitive to outliers (such as holidays and equipment failures), and is used to calculate the base electricity quantity and separate the baseline affected by meteorological factors;

[0109] In this embodiment, because the median of the electricity quantity distribution is selected, a robust baseline load can be provided to avoid the interference of extreme events on the model results. Therefore, the quantile is selected as ;

[0110] S32. Train the base time-of-use electricity decomposition model based on historical time-of-use electricity data to obtain a trained base time-of-use electricity decomposition model, and input the real-time time-of-use electricity data into the base time-of-use electricity decomposition model to obtain the real-time base time-of-use electricity components;

[0111] S33. Obtain the time-of-use electricity residuals based on the base time-of-use electricity components, which is expressed by the formula:

[0112] ;

[0113] In the formula, is the time-of-use electricity residual at time is the real-time time-of-use electricity at time

[0114] S4. Use the improved complete ensemble empirical mode decomposition with adaptive noise algorithm ICEEMDAN to perform a secondary mode decomposition on the time-of-use electricity residuals to obtain a set of meteorological sensitive mode components. Specifically:

[0115] S41. Perform the first ICEEMDAN decomposition on the time-of-use electricity residuals, and decompose them into An intrinsic mode function and a residual component, expressed by the formula:

[0116] ;

[0117] In the formula, is the th intrinsic mode function obtained from the first decomposition; is the residual component obtained from the first decomposition;

[0118] S42. Select all the intrinsic mode functions obtained from the first decomposition for the second decomposition, expressed by the formula:

[0119] ;

[0120] In the formula, represents the th intrinsic mode function obtained by performing the second decomposition on , is the total number of intrinsic mode functions obtained from the second decomposition; is the residual component obtained from the second decomposition;

[0121] S43. Combine all the intrinsic mode functions obtained from the second decomposition to obtain the meteorological sensitive mode component set , expressed by the formula:

[0122] ;

[0123] S5. Construct a multi-objective evolutionary algorithm MOEA / D multi-objective optimization model based on the decomposition, use the meteorological sensitive mode component set as the input, and solve the MOEA / D multi-objective optimization model using the Chebyshev decomposition method to obtain the corrected meteorological power component, where the MOEA / D multi-objective optimization model updates the weight vector based on the key meteorological factor set;

[0124] S51. In this embodiment, the two set goals include minimizing the residual fitting error and maximizing the consistency between the weight and the sensitivity coefficient. The constructed MOEA / D multi-objective optimization model is expressed by the formula:

[0125] ;

[0126] In the formula, is the goal of minimizing the residual fitting error; is the th non-linear mapping function of the th meteorological factor at the time; is the length of the real-time time-sharing power signal; is the weight coefficient of the For the goal of maximizing the consistency of weights and sensitivity coefficients; Is the total number of meteorological factors;

[0127] Among them, minimizing the residual fitting error is to make the combination of meteorological factors best fit the meteorological sensitive signals in the residuals; maximizing the consistency of weights and sensitivity coefficients is to prevent the optimized weights from completely deviating from the physical meaning (for example, a meteorological factor with little actual impact is given too high a weight). Through the above two goals, both the residual signals in the data are utilized and the physical impact mechanism of meteorological factors is respected, avoiding the overfitting risk of pure data-driven models;

[0128] S52. Solve the MOEA / D multi-objective optimization model using the Chebyshev decomposition method, specifically:

[0129] Define the Chebyshev distance , expressed by the formula:

[0130] ;

[0131] In the formula, Is the decision variable, used to represent the weight coefficient of the th meteorological factor; Is the multi-objective trade-off weight set. Specifically in this embodiment, since the goal of this embodiment is two, the corresponding ; Is the objective index; Is the ideal point set, ; Is the objective function value corresponding to the decision variable ;

[0132] Generate a set of uniformly distributed initial weight vector sets according to the key meteorological factor set, expressed by the formula:

[0133] ;

[0134] In the formula, Is the initial weight vector set; Is the bit index of the initial weight vector; Is the total number of initial weight vectors, and each initial weight vector corresponds to a sub-problem;

[0135] Calculate the distance between the weight vector in the initial weight vector set and other weight vectors, and select the initial weight vectors whose distances meet the distance threshold to form the corresponding neighborhood ;

[0136] Randomly generate the initial population set , expressed by the formula:

[0137] ;

[0138] In the formula, is the solution corresponding to the initial weight vector .

[0139] Randomly select two solutions from the neighborhood , perform genetic operations on the two selected solutions to obtain new solutions, and calculate the objective function values of the new solutions.

[0140] Perform iterative solution, including for each sub-problem in the neighborhood in each iteration, calculate the Chebyshev distances of the current solution and the new solution with respect to the multi-objective trade-off weights and the ideal point respectively. If the Chebyshev distance corresponding to the new solution is less than or equal to the Chebyshev distance corresponding to the current solution, then replace the current solution with the new solution.

[0141] If the objective function value is less than the ideal point, then the ideal point is equal to the objective function value.

[0142] Repeat the above iterative solution process until the preset termination condition is satisfied, then stop the iteration. At this time, the solutions in the population are the approximate optimal solutions of the MOEA / D multi-objective optimization model, and the corrected meteorological power components are obtained based on the approximate optimal solutions.

[0143] S6. Superimpose the basic time-of-use power components and the corrected meteorological power components to obtain the final time-of-use power decomposition result, which is expressed by the formula:

[0144] ;

[0145] In the formula, is the final time-of-use power decomposition result; is the corrected meteorological power component.

[0146] In summary, this embodiment provides a time-of-use power decomposition method based on sensitivity analysis.

[0147] In the embodiments of the present invention, "at least one" means one or more, and "a plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent the case of A existing alone, A and B existing simultaneously, and B existing alone. Where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the front and rear associated objects. "At least one of the following" and its similar expressions refer to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0148] Those of ordinary skill in the art can realize that the various units and algorithm steps described in the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0149] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0150] In several embodiments provided by the present invention, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs that can store program codes.

[0151] The above are only the embodiments of the present invention, and do not limit the patent scope of the present invention. All equivalent structural or equivalent process transformations made by using the contents of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, are equally included in the patent protection scope of the present invention.

Claims

1. A time-based electricity decomposition method based on sensitivity analysis, characterized in that: The method comprises: Obtain the time-sharing electricity data related to the target area, including historical time-sharing electricity data, real-time time-sharing electricity data and time-sharing meteorological data, and perform grid processing on the time-sharing electricity data related to obtain meteorological factors and time-sharing electricity-meteorological joint data sets; Based on the combined data set of time-sharing electricity and meteorology, the sensitivity index of each meteorological factor to time-sharing electricity is calculated, which is as follows: The global sensitivity of the combined time-sharing electricity and meteorological data set is calculated by the Sobol index and expressed as follows: ； In the formula, For the The global sensitivity of each meteorological factor; For the Meteorological factors; For Other meteorological factors besides is the variance; is the conditional expectation; It is the time-sharing power output variable; For a given Meteorological factors When the value of About Other meteorological factors expectations; For expectations About Meteorological factors Find the variance; Time-sharing power output variable The contrast; Calculating the Mutual Information of Meteorological Factors and Time-of-day Electricity Using Kernel Density Estimation , expressed as: ； In the formula, for and The joint probability density of and They are and The marginal probability density of and They are and Specific value of Based on global sensitivity and mutual information The sensitivity index is calculated and expressed as follows: ； In the formula, For the Sensitivity index of each meteorological factor; is the weight coefficient; is the information entropy of time-sharing electricity; And screen the meteorological factors whose sensitivity index is greater than the sensitivity threshold to obtain the key meteorological factor set; Building a basic time-sharing electricity decomposition model based on historical time-sharing electricity data, inputting real-time time-sharing electricity data into the basic time-sharing electricity decomposition model to obtain basic time-sharing electricity components; and obtaining time-sharing electricity residuals based on the basic time-sharing electricity components; The improved fully adaptive noise set decomposition algorithm ICEEMDAN is used to perform secondary modal decomposition on the time-sharing power residual to obtain a set of meteorological sensitive modal components, specifically: The first ICEEMDAN decomposition is performed on the time-sharing power residual, which is decomposed into The intrinsic mode functions and a residual component are expressed as: ； In the formula, is the first decomposition Intrinsic mode functions; is the residual component obtained from the first decomposition; All the intrinsic mode functions obtained from the first decomposition are selected for the second decomposition, which can be expressed as: ； In the formula, Express The second decomposition yields The intrinsic mode functions, is the total number of intrinsic mode functions obtained by the second decomposition; is the residual component obtained from the second decomposition; All the intrinsic mode functions obtained by the second decomposition are combined to obtain a set of meteorological sensitive mode components , expressed as: ； A MOEA / D multi-objective optimization model is constructed, and a set of meteorological sensitive modal components is used as input. The MOEA / D multi-objective optimization model is solved by using the Chebyshev decomposition method to obtain a corrected meteorological power component, wherein the MOEA / D multi-objective optimization model updates a weight vector based on a set of key meteorological factors; The basic time-sharing electricity component is superimposed with the corrected meteorological electricity component to obtain the final time-sharing electricity decomposition result.

2. The time-sharing electricity decomposition method based on sensitivity analysis according to claim 1 is characterized in that: The time-sharing meteorological data include temperature, humidity, wind speed, sunshine intensity and precipitation.

3. The time-sharing electricity decomposition method based on sensitivity analysis according to claim 1 is characterized in that: The gridding process of the time-sharing electricity-related data to obtain the meteorological factors and the time-sharing electricity-meteorological joint data set is specifically: The time-sharing meteorological data is gridded according to the city dimension to obtain gridded meteorological indicators, namely meteorological factors; the historical time-sharing electricity data and real-time time-sharing electricity data are gridded according to the city dimension and industry dimension, and the time-sharing electricity-meteorological joint data set is constructed in combination with meteorological factors, where the grid resolution is 1km×1km; The time-sharing electricity-meteorological joint data set is subjected to data preprocessing, including data alignment, outlier removal and feature standardization.

4. The time-sharing electricity decomposition method based on sensitivity analysis according to claim 1 is characterized in that: The basic time-sharing electricity decomposition model based on historical time-sharing electricity data is constructed as follows: The basic time-sharing electricity decomposition model is a quantile regression model, which is expressed as follows: ； In the formula, for The time quantile is Basic time-sharing power; during training, for The time quantile is The historical time-sharing power consumption, real-time monitoring, for The time quantile is Real-time time-sharing power consumption; and for The time quantile is The regression coefficient of and Respectively represent Cities and industries; and They are the total number of cities and the total number of industries respectively; for The time quantile is No. The coefficient of the variable of each city; for The time quantile is No. The coefficient of the industry variable for each industry; for The time quantile is The error term of is the quantile; The basic time-sharing electricity decomposition model is trained based on historical time-sharing electricity data to obtain a trained basic time-sharing electricity decomposition model, and real-time time-sharing electricity data is input into the basic time-sharing electricity decomposition model to obtain real-time basic time-sharing electricity components.

5. The time-sharing electricity decomposition method based on sensitivity analysis according to claim 4 is characterized in that: The time-sharing power residual obtained based on the basic time-sharing power component is expressed as follows: ； In the formula, for Time-sharing power residual at the moment; for Real-time time-sharing power consumption at all times.

6. The time-sharing electricity decomposition method based on sensitivity analysis according to claim 5 is characterized in that: The constructed MOEA / D multi-objective optimization model is expressed as follows: ； In the formula, The residual fitting error is minimized; For the Meteorological factors in Nonlinear mapping function of time; The length of the real-time time-sharing power signal; For the The weight coefficient of each meteorological factor; The goal is to maximize the consistency between weights and sensitivity coefficients; is the total number of meteorological factors.

7. The time-sharing electricity decomposition method based on sensitivity analysis according to claim 6 is characterized in that: The Chebyshev decomposition method is used to solve the MOEA / D multi-objective optimization model, specifically: Defining Chebyshev distance , expressed as: ； In the formula, is a decision variable, used to refer to the The weight coefficient of each meteorological factor; is a set of multi-objective trade-off weights, ; is the target index; is the set of ideal points, ; is the decision variable The corresponding objective function value; A set of uniformly distributed initial weight vectors is generated according to the set of key meteorological factors, which can be expressed as follows: ； In the formula, is the initial weight vector set; is the bit index of the initial weight vector; is the total number of initial weight vectors, each of which corresponds to a subproblem; Calculate the weight vector in the initial weight vector set The distance to other weight vectors, select the initial weight vector whose distance meets the distance threshold to form the corresponding neighborhood ; Randomly generate an initial population set , expressed as: ； In the formula, corresponds to the initial weight vector The solution; From the neighborhood Randomly select two solutions, perform genetic operations on the two selected solutions, obtain new solutions, and calculate the objective function value of the new solution; Perform iterative solution, including calculating the Chebyshev distances of the current solution and the new solution relative to the multi-objective trade-off weights and the ideal point for each subproblem in the neighborhood in each iteration, and replacing the current solution with the new solution if the Chebyshev distance corresponding to the new solution is less than or equal to the Chebyshev distance corresponding to the current solution; If the objective function value is less than the ideal point, then the ideal point is equal to the objective function value; Repeat the above iterative solution process until the preset termination condition is met, then stop the iteration. At this time, the solution in the population is the approximate optimal solution of the MOEA / D multi-objective optimization model, and the corrected meteorological power component is obtained based on the approximate optimal solution.

8. The time-sharing electricity decomposition method based on sensitivity analysis according to claim 7 is characterized in that: The final time-sharing power decomposition result is expressed as follows: ； In the formula, The final time-sharing power decomposition result; is the corrected meteorological electricity component.

Images