A method, device, equipment and medium for constructing a transaction power data privacy index model

Abstract

The present application discloses a method, apparatus, equipment and medium for constructing a transaction power data privacy index model. In the above method, the comprehensive volatility of the transaction power data is obtained based on the high-frequency volatility and low-frequency volatility of the transaction power data curve; the Hurst index of the transaction power data is calculated using the R / S analysis method, and the non-autocorrelation parameter of the transaction power data is obtained; the fuzzy entropy value of the transaction power data is calculated, and the information uncertainty parameter of the transaction power data is obtained; and the transaction power data privacy index model is constructed based on the comprehensive volatility, non-autocorrelation parameter and information uncertainty parameter. In this process, the volatility, Hurst index and fuzzy entropy of the transaction power data curve are used to quantitatively characterize the privacy of the transaction power data between the microgrid group and the distribution network, so that the transaction power data privacy index model can be constructed to reduce the waste of resources in the privacy protection process.

Description

Technical Field

[0001] The present application relates to the technical field of data privacy security, and in particular to a method, apparatus, device, and medium for constructing a transaction power data privacy indicator model. Background Art

[0002] With the accelerated advancement of new power systems, various power equipment and corresponding measurement technologies are developing rapidly. Microgrids, as wireless network systems, not only enable communication but also leverage energy consumption data to reduce electricity costs. For example, dynamic pricing can provide accurate consumption data to microgrids or high-value real-time information to the power grid. However, the highly granular power data exchanged between microgrids and distribution networks also increases the exposure of microgrid electricity usage information. For example, data containing a large amount of private microgrid information can be used to analyze load fluctuations and infer the microgrid's electricity usage patterns and daily behaviors. Therefore, it is necessary to ensure the privacy and security of microgrid electricity usage information.

[0003] Current research on privacy protection measures for microgrid clusters only considers their own physical constraints and privacy protection needs. Various numerical values ​​are added to the transaction power data curve to achieve the purpose of privacy protection. This requires additional resources and causes resource waste in the privacy protection process. Summary of the Invention

[0004] The present application provides a method, apparatus, device and medium for constructing a privacy indicator model for transaction power data, which constructs a privacy indicator model for transaction power data based on the inherent privacy attributes of the transaction power data, namely, comprehensive volatility, non-autocorrelation and information uncertainty, thereby solving the problem of resource waste.

[0005] In a first aspect, the present application provides a method for constructing a transaction power data privacy indicator model, the method comprising:

[0006] According to the high-frequency volatility and low-frequency volatility of the trading power data curve, the comprehensive volatility of the trading power data is obtained;

[0007] Calculating the Hurst index of the trading power data using the rescaled range R / S analysis method, and obtaining the non-autocorrelation parameter of the trading power data based on the Hurst index;

[0008] Calculate the fuzzy entropy value of the transaction power data, and obtain the information uncertainty parameter of the transaction power data according to the fuzzy entropy value;

[0009] Based on the comprehensive volatility, non-autocorrelation parameter and information uncertainty parameter, a trading power data privacy index model is constructed.

[0010] Optionally, the process of obtaining the high-frequency volatility of the trading power data curve includes:

[0011] Calculating the upper and lower envelope areas of the transaction power data curve using Lebesgue integral;

[0012] The high-frequency volatility of the trading power data curve is determined by taking the average value of the upper and lower envelope areas per unit time.

[0013] Optionally, the process of obtaining the low-frequency volatility of the trading power data curve includes:

[0014] Performing state analysis on the transaction power data using a time series symbol aggregation method, extracting a time series of the transaction power data, and converting the time series into a discrete character sequence;

[0015] Calculate the slope value between the representative value of each character in the discrete character sequence and the representative value of the previous character, and obtain the low-frequency volatility of the trading power data curve based on the slope value.

[0016] Optionally, the comprehensive volatility of the trading power data is obtained according to the high-frequency volatility and the low-frequency volatility of the trading power data curve, including:

[0017] Use wavelet analysis to decompose transaction power data to obtain high-frequency information and low-frequency information, and calculate the proportion of high-frequency and low-frequency components based on the high-frequency and low-frequency information;

[0018] The high-frequency volatility and low-frequency volatility are weighted and summed according to the proportion of high and low frequency components to calculate the comprehensive volatility.

[0019] Optionally, calculate the Hurst exponent of the trading power data using the R / S analysis method, including:

[0020] According to each length L of N different lengths L, the transaction power data is divided into a plurality of interval groups, wherein the interval group includes N adjacent subintervals of length L, where N is a positive integer and L is a positive number;

[0021] Calculate the average rescaled range of N adjacent subintervals in each interval group;

[0022] The Hurst index was calculated by fitting the regression coefficients using the least squares method based on the multiple average rescaled ranges.

[0023] Optionally, calculating the fuzzy entropy value of the transaction power data includes:

[0024] Reconstruct the transaction power data into an m-dimensional phase space to obtain a vector of phase space dimension m, where m is a positive integer;

[0025] The distance between vector i and vector j in the vector of phase space dimension m is calculated by fuzzy membership function, where i and j are positive integers;

[0026] Get the function φ based on the distance between vector i and vector j m (n, r), where n represents the gradient of the similarity tolerance boundary and r represents the width of the similarity tolerance boundary;

[0027] Add 1 to the phase space dimension m and repeat the operation of obtaining the function to obtain the function φ m+1 (n,r);

[0028] The function φ m (n,r) and function φ m+1 Substitute (n, r) into the fuzzy entropy function and calculate the fuzzy entropy value.

[0029] Optionally, after constructing the model of the transaction efficiency data privacy indicator system, the following is also included:

[0030] Normalize the comprehensive volatility, non-autocorrelation and information uncertainty to obtain a normalized data set;

[0031] A three-dimensional space is established based on the normalized data set, and the area of ​​a preset size closest to the origin in the three-dimensional space is defined as a strong privacy area, the area of ​​a preset size farthest from the origin in the three-dimensional space is defined as a weak privacy area, and the remaining area in the three-dimensional space except the strong privacy area and the weak privacy area is defined as a medium privacy area.

[0032] In a second aspect, the present application provides a device for constructing a transaction power data privacy indicator model, the device comprising:

[0033] an obtaining unit, configured to obtain a comprehensive volatility of the trading power data according to the high-frequency volatility and the low-frequency volatility of the trading power data curve;

[0034] a calculation unit, configured to calculate the Hurst index of the trading power data using a rescaled range R / S analysis method, and obtain a non-autocorrelation parameter of the trading power data based on the Hurst index;

[0035] The calculation unit is further used to calculate the fuzzy entropy value of the transaction power data and obtain the information uncertainty parameter of the transaction power data according to the fuzzy entropy value;

[0036] The construction unit is used to construct a trading power data privacy indicator model based on the comprehensive volatility, non-autocorrelation parameter and information uncertainty parameter.

[0037] Optionally, the computing unit is further configured to:

[0038] Calculating the upper and lower envelope areas of the transaction power data curve using Lebesgue integral;

[0039] The determination unit is used to determine the high-frequency volatility of the trading power data curve by taking the average value of the upper and lower envelope areas within a unit time.

[0040] Optionally, the device further comprises:

[0041] a processing unit, configured to perform state analysis on the transaction power data using a time series symbol aggregation method, extract a time series of the transaction power data, and convert the time series into a discrete character sequence;

[0042] The calculation unit is further used to calculate the slope value between the representative value of each character in the discrete character sequence and the representative value of the previous character, and obtain the low-frequency volatility of the transaction power data curve according to the slope value.

[0043] Optionally, the obtaining unit is specifically configured to:

[0044] Use wavelet analysis to decompose transaction power data to obtain high-frequency information and low-frequency information, and calculate the proportion of high-frequency and low-frequency components based on the high-frequency and low-frequency information;

[0045] The high-frequency volatility and low-frequency volatility are weighted and summed according to the proportion of high and low frequency components to calculate the comprehensive volatility.

[0046] Optionally, when the calculation unit calculates the Hurst index of the trading power data using the rescaled range R / S analysis method, it is specifically used to:

[0047] According to each length L of N different lengths L, the transaction power data is divided into a plurality of interval groups, wherein the interval group includes N adjacent subintervals of length L, where N is a positive integer and L is a positive number;

[0048] Calculate the average rescaled range of N adjacent subintervals in each interval group;

[0049] The Hurst index was calculated by fitting the regression coefficients using the least squares method based on the multiple average rescaled ranges.

[0050] Optionally, when the calculation unit calculates the fuzzy entropy value of the transaction power data, it is specifically used to:

[0051] Reconstruct the transaction power data into an m-dimensional phase space to obtain a vector of phase space dimension m, where m is a positive integer;

[0052] The distance between vector i and vector j in the vector of phase space dimension m is calculated by fuzzy membership function, where i and j are positive integers;

[0053] Get the function φ based on the distance between vector i and vector j m (n, r), where n represents the gradient of the similarity tolerance boundary and r represents the width of the similarity tolerance boundary;

[0054] Add 1 to the phase space dimension m and repeat the operation of obtaining the function to obtain the function φ m+1 (n,r);

[0055] The function φ m (n,r) and function φ m+1 Substitute (n, r) into the fuzzy entropy function and calculate the fuzzy entropy value.

[0056] Optionally, the device further comprises:

[0057] A normalization processing unit is used to normalize the comprehensive volatility, non-autocorrelation and information uncertainty to obtain a normalized data set;

[0058] A definition unit is used to establish a three-dimensional space based on a normalized data set, and define the area of ​​a preset size closest to the origin in the three-dimensional space as a strong privacy area, the area of ​​a preset size farthest from the origin in the three-dimensional space as a weak privacy area, and the remaining area in the three-dimensional space other than the strong privacy area and the weak privacy area as a medium privacy area.

[0059] In a third aspect, the present application further provides an electronic device, comprising a memory and a processor:

[0060] Memory is used to store computer programs;

[0061] The processor is configured to execute the method provided in the first aspect according to the computer program.

[0062] In a fourth aspect, the present application further provides a computer-readable storage medium, which is used to store a computer program, and the computer program is used to execute the above-mentioned first aspect and the provided method.

[0063] It can be seen that this application has the following beneficial effects:

[0064] The present application provides a method for constructing a privacy index system for transaction power data. First, the comprehensive volatility of the transaction power data is obtained based on the high-frequency volatility and low-frequency volatility of the transaction power data curve. Then, the Hurst index of the transaction power data is calculated using the rescaled range R / S analysis method, and the Hurst index is used to obtain the non-autocorrelation parameter of the quantitative transaction power data. Then, the fuzzy entropy value of the transaction power data is calculated, and the information uncertainty parameter of the transaction power data is obtained from the fuzzy entropy value. Finally, a transaction power data privacy index model is constructed based on the comprehensive volatility, non-autocorrelation and information uncertainty. In this process, the volatility, Hurst index and fuzzy entropy of the transaction power data curve are used to quantitatively characterize the privacy of the transaction power data between the microgrid group and the distribution network. Therefore, a transaction power data privacy index model can be constructed based on the inherent privacy attributes of the transaction power data, thereby reducing resource waste in the privacy protection process. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0066] Figure 1 A flowchart of a method for constructing a transaction power data privacy indicator model in an embodiment of the present application;

[0067] Figure 2 This is a flow chart of an embodiment of a method for constructing a transaction power data privacy indicator model in an embodiment of the present application;

[0068] Figure 3 This is a flow chart of constructing a transaction power data privacy indicator system in an embodiment of the present application;

[0069] Figure 4 This is a schematic diagram of strong and weak privacy partitioning of transaction power data in an embodiment of the present application;

[0070] Figure 5 This is a schematic diagram of the structure of an apparatus for constructing a transaction power data privacy indicator model in an embodiment of the present application;

[0071] Figure 6 This is a structural diagram of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0072] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0073] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0074] To ensure the privacy and security of microgrid electricity usage information, domestic and international scholars have proposed various privacy protection methods, including cryptography-based privacy protection methods, data aggregation-based privacy protection methods, anonymization-based privacy protection methods, and differential privacy-based privacy protection methods. General user-side microgrid privacy protection measures, such as cryptography, place high computational demands on user communication equipment. By introducing blockchain technology to enable multi-party review and encryption / decryption of data, this protects data while incurring significant additional costs for key management. However, mathematical protection methods based solely on differential privacy introduce virtual noise into the load curve, impacting the stable operation of the entire power grid system. Energy storage-based privacy protection methods, on the other hand, utilize energy storage charging and discharging behavior to mask users' true electricity usage information, place low demands on communication equipment, and enable user autonomy.

[0075] It can be seen from this that current research by domestic and foreign research institutions on the privacy of transaction power data curves between user-side microgrids and distribution networks mostly focuses on improving privacy protection methods. Existing privacy protection methods can effectively improve the security of microgrid transaction power data, but most algorithms only consider their own physical constraints and privacy protection requirements. They all add various numerical values ​​to the transaction power data curve to achieve the purpose of privacy protection, resulting in the need to add additional resources and causing resource waste in the privacy protection process.

[0076] Because the inherent regularity of transaction power data curves is highly correlated with privacy, this privacy, distinct from privacy achieved through specific privacy protection methods, specifically refers to the inherent privacy properties of transaction power data curves due to their original features, shape, and other privacy-related attributes. Therefore, when considering privacy, the characteristics of the transaction power data curves themselves should also be considered. And, like regularity, the privacy of transaction power data curves cannot be measured solely based on a single metric. Furthermore, transaction power data is a time-series dynamic data that is continuously updated over time. Therefore, updated data inevitably incorporates the characteristics of previous data, leading to temporal coupling between privacy and regularity. Therefore, it is urgent to address the resource waste inherent in existing privacy protection of transaction power data curves between microgrids and distribution networks, which fails to consider inherent privacy properties such as their original features and shape.

[0077] In an embodiment of the present application, a privacy index model for trading power data is constructed by using the inherent privacy attributes of the trading power data, namely, comprehensive volatility, non-autocorrelation, and information uncertainty, thereby solving the problem of resource waste. In specific implementation, the method may include, for example: first, according to the high-frequency volatility and low-frequency volatility of the trading power data curve, the comprehensive volatility of the trading power data is obtained, and then the Hurst index of the trading power data is calculated by using the rescaled range R / S analysis method, and the non-autocorrelation parameter of the quantitative trading power data is obtained from the Hurst index, and then the fuzzy entropy value of the trading power data is calculated, and the information uncertainty parameter of the trading power data is obtained from the fuzzy entropy; finally, according to the comprehensive volatility, non-autocorrelation, and information uncertainty, a privacy index model for trading power data is constructed.

[0078] It can be seen that the method provided by the implementation of this application realizes the quantitative characterization of the privacy of transaction power data between microgrid groups and distribution networks through the volatility, Hurst index and fuzzy entropy of the transaction power data curve, so as to construct a transaction power data privacy indicator model based on the inherent privacy attributes of the transaction power data, thereby reducing resource waste in the privacy protection process.

[0079] To facilitate understanding of the specific implementation of a method for constructing a transaction power data privacy indicator model provided in an embodiment of the present application, it will be described below with reference to the accompanying drawings.

[0080] It should be noted that the subject implementing the method for constructing a transaction power data privacy indicator model may be the device for constructing a transaction power data privacy indicator model provided in the embodiments of the present application, and the device for constructing a transaction power data privacy indicator model may be carried in an electronic device or a functional module of an electronic device. The electronic device in the embodiments of the present application may be any device capable of implementing the method for constructing a transaction power data privacy indicator model in the embodiments of the present application, for example, an Internet of Things (IoT) device.

[0081] Figure 1 A flow chart of a method for constructing a transaction power data privacy index model provided in an embodiment of the present application. The method can be applied to a device for constructing a transaction power data privacy index model, which can be, for example, Figure 5 The device 500 for constructing a transaction power data privacy index model shown in FIG. 5 may also be integrated into a Figure 6 Functional modules in the electronic device 600 are shown.

[0082] like Figure 1 As shown, the method includes the following S101 to S104:

[0083] S101: Obtaining a comprehensive volatility of the trading power data according to the high-frequency volatility and the low-frequency volatility of the trading power data curve.

[0084] In order to construct a privacy index model for trading power data, it is first necessary to obtain the comprehensive volatility of the trading power data based on the high-frequency volatility and low-frequency volatility of the trading power data curve; calculate the Hurst index of the trading power data using the rescaled range R / S analysis method, and obtain the non-autocorrelation parameter of the trading power data based on the Hurst index; then calculate the fuzzy entropy value of the trading power data, and obtain the information uncertainty parameter of the trading power data based on the fuzzy entropy value; finally, construct a privacy index model for trading power data based on the comprehensive volatility, non-autocorrelation parameter and information uncertainty parameter. Therefore, in the embodiment of the present application, the comprehensive volatility of the trading power data is obtained through S101, which makes preliminary preparations for the subsequent construction of the privacy index model for trading power data.

[0085] In the embodiment of the present application, before executing S101, a process of obtaining the high-frequency volatility of the trading power data curve is also included. The specific process is as follows:

[0086] First, the Lebesgue integral is used to calculate the upper and lower envelope areas of the trading power data curve. The high-frequency volatility of the trading power data curve is then determined by taking the average of the upper and lower envelope areas per unit time. Using the Lebesgue integral instead of the conventional Riemann integral to calculate the upper and lower envelope areas of the trading power data curve resolves the problem of discontinuous integration of the trading power data curve.

[0087] The above-mentioned method of using Lebesgue integral to calculate the area of ​​the upper and lower envelopes of the trading power data curve may include: obtaining a trading power time series data sequence based on the trading power data, and normalizing the trading power time series data sequence, and then respectively obtaining the maximum and minimum values ​​of the trading power in each time period based on the normalized trading power time series data sequence, interpolating the maximum and minimum value sequences into two time series of length T through a linear interpolation method, respectively serving as the upper and lower envelopes of the trading power time series data sequence, and finally using Lebesgue integral to obtain the areas of the upper and lower envelopes.

[0088] Furthermore, before executing S101 in the embodiment of the present application, a process of obtaining the low-frequency volatility of the trading power data curve is also included. The specific process is as follows:

[0089] The trading power data is analyzed using a time series symbol aggregation method to extract the time series of the trading power data and convert it into a discrete character sequence. The slope between the representative value of each character in the discrete character sequence and the representative value of the previous character in each time period is calculated, and the low-frequency volatility of the trading power data curve is derived from the slope. In this process, the trading power data is analyzed using a time series symbol aggregation method, and the time series features of the extracted trading power data are converted into a discrete character sequence, thereby achieving a dimensionality reduction effect. The low-frequency volatility of the trading power data curve can then be calculated.

[0090] The above-mentioned method of calculating the low-frequency volatility of the trading power data curve using time series symbol aggregation can include: first normalizing the trading power data curve and performing load segment aggregation approximation processing, and then symbolizing the trading power data curve into a discrete character string, wherein the trading power data curve after load segment aggregation approximation conversion is divided into intervals according to the amplitude size, and each character represents a certain amplitude range; finally, the slope average of the character representative value of each time period and the character representative value of the previous period is used to characterize the magnitude of the low-frequency volatility.

[0091] In the above method, the specific idea of ​​approximating the load segment aggregation of the transaction power data curve is to replace the amplitude of the decrease in the same time interval with the average value. The calculation method is as follows:

[0092]

[0093] in, is the average value of the data in the i-th segment; k i is the time point of the i-th load data; j is the index of the load data after normalization; i is the index of the load data after load segment aggregation and approximate transformation.

[0094] In the above method, the formula for calculating the low-frequency volatility of the trading power data curve is as follows:

[0095]

[0096] Where r2 is the low-frequency volatility of the transaction power data curve; n is the total number of symbol sequence stages; y i is the representative value of the character in the i-th period; t is the total number of time point bits contained in each stage of the symbol sequence.

[0097] The high-frequency volatility and low-frequency volatility of the trading power data curve obtained above can be used to execute S101. As an example, S101 may include: using wavelet analysis to decompose the trading power data to obtain high-frequency information and low-frequency information, and calculating the proportion of high- and low-frequency components based on the high-frequency information and the low-frequency information; finally, performing weighted summation of the high-frequency volatility and the low-frequency volatility according to the proportion of the high- and low-frequency components to calculate the comprehensive volatility.

[0098] The specific process of S101 can include: first, selecting sym wavelet as the basis function, using discrete wavelet transform method to calculate the approximate signal A3 and detail signals D1, D2 and D3 of the transaction power data curve through three decompositions. Among them, the approximate signal A3 is high-frequency information; the detail signals D1, D2 and D3 are low-frequency information. Then, the high-frequency information and low-frequency information are substituted into the formula , calculate the proportion of high- and low-frequency components k. Finally, substitute high-frequency volatility, low-frequency volatility, and the proportion of high- and low-frequency components into the comprehensive volatility formula for weighted summation, thereby defining an indicator for measuring the comprehensive volatility of the trading power data curve. The comprehensive volatility formula is as follows:

[0099]

[0100] Among them, Vol is the comprehensive volatility of the trading power data; k is the proportion of high-frequency and low-frequency components in the trading power data curve; r1 is the high-frequency volatility of the trading power data curve; r2 is the low-frequency volatility of the trading power data curve.

[0101] During this process, when calculating the upper and lower envelope areas of the trading power data curve, the Lebesgue integral was used to address the breakpoint integral problem of the trading power data curve as a discontinuous curve. Furthermore, the time series symbol aggregation method was used to analyze the state of the trading power data, achieving dimensionality reduction and further calculating its low-frequency volatility. Finally, wavelet analysis was used to extract high-frequency and low-frequency information through wavelet basis functions and scaling functions, respectively. Based on the quantitative characterization of high-frequency and low-frequency volatility, the discrete wavelet transform was used to calculate the proportion of the high- and low-frequency components of the curve, more comprehensively and accurately quantifying the comprehensive volatility of the trading power data.

[0102] S102: Calculate the Hurst index of the trading power data using the rescaled range R / S analysis method, and obtain the non-autocorrelation parameter of the trading power data based on the Hurst index.

[0103] As an example, S102 may include: dividing the trading power data into multiple interval groups according to each length L of N different lengths L, the interval group including N adjacent sub-intervals of length L, where N is a positive integer and L is a positive number; calculating the average rescaled range of the N adjacent sub-intervals in each interval group; fitting the regression coefficients using the least squares method based on the multiple average rescaled ranges to calculate the Hurst index; S1024, obtaining the non-autocorrelation parameter of the trading power data according to the Hurst index.

[0104] In the above method, the transaction power data is first divided into multiple interval groups and then the average rescaled range of N adjacent sub-intervals in each interval group is calculated. Alternatively, a length L is set first and then the average rescaled range of N adjacent sub-intervals under the corresponding length L is calculated. After the length L is changed, the average rescaled range of N intervals under the corresponding length L is calculated again. The specific calculation order is not limited in this embodiment of the application. Among them, the average rescaled range (R / S) is calculated. l The formula is as follows:

[0105]

[0106] Among them, R n is the range of the nth interval; S n is the standard deviation of the nth interval; N is the number of adjacent subintervals.

[0107] After calculating the average rescaled range under different subinterval lengths L, due to lg(R / S) l There is a linear relationship between it and lgl, such as lg(R / S) l =lg(K)+Hlgl, where K is a constant. Therefore, the approximate value of the Hurst exponent H can be calculated by linear fitting using least squares.

[0108] In this process, the Hurst index of the trading power data is calculated using the rescaled range R / S analysis method, and the Hurst index obtains the non-autocorrelation parameter of the trading power data. That is, the Hurst index quantifies the non-autocorrelation of the trading power data, which can effectively characterize the privacy of the trading power data due to its own intrinsic characteristics, curve shape and other characteristics.

[0109] S103: Calculate the fuzzy entropy value of the transaction power data, and obtain the information uncertainty parameter of the transaction power data according to the fuzzy entropy value.

[0110] As an example, S103 may include: S1031, reconstructing the transaction power data into an m-dimensional phase space to obtain a vector of the phase space dimension m, where m is a positive integer; S1032, calculating the distance between vector i and vector j in the vector of the phase space dimension m by using a fuzzy membership function, where i and j are positive integers; S1033, obtaining a function φ based on the distance between vector i and vector j. m (n, r), where n represents the gradient of the similarity tolerance boundary and r represents the width of the similarity tolerance boundary; S1034, add 1 to the phase space dimension m and repeat the operation of obtaining the function to obtain the function φ m+1 (n,r); S1035, the function φ m (n,r) and function φ m+1 Substitute (n, r) into the fuzzy entropy function and calculate the fuzzy entropy value.

[0111] In the above S1031, the transaction power data is reconstructed into an m-dimensional phase space, and the calculation formula for obtaining the vector of the phase space dimension m is as follows:

[0112]

[0113] Wherein, u(i) is the i-th transaction power data; i, j, m = 1, 2, ...; u0(i) is the mean of the transaction power data.

[0114] The step of calculating the distance between vector i and vector j in the vector of phase space dimension m by using the fuzzy membership function in S1032 may include: first defining the distance is a vector With vector The maximum distance between corresponding elements in , The calculation formula is as follows:

[0115]

[0116] Then the vector is measured by the fuzzy membership function With vector The distance between The specific calculation formula is as follows:

[0117]

[0118] Where n represents the gradient of the similarity tolerance boundary; r represents the width of the similarity tolerance boundary.

[0119] In the above S1033, the function φ is obtained based on the distance between vector i and vector j. m (n,r) can include: according to distance Get the function φm(n,r). The specific method of obtaining it is as follows:

[0120]

[0121] In the above S1035, the above φ is obtained. m (n,r) and φ m+1 (n, r) can be substituted into the fuzzy entropy function to calculate the fuzzy entropy value. The specific fuzzy entropy function is as follows:

[0122] FuzzyEn(m,n,r)=lnφ m (n,r)-lnφ m+1 (n,r)

[0123] In this process, the concept of fuzzy membership function is used to make the value of fuzzy entropy change continuously and smoothly with the change of parameters during the calculation process. In addition, the fuzzy entropy eliminates the influence of baseline drift and slight data fluctuations through mean operation, making the description of information uncertainty of trading power data more accurate.

[0124] S104: Construct a transaction power data privacy indicator model based on the comprehensive volatility, non-autocorrelation parameter, and information uncertainty parameter.

[0125] As an example, S104 may include: constructing a function model of the transaction power data privacy indicator by integrating volatility, non-autocorrelation parameter, and information uncertainty parameter. The specific function model is as follows:

[0126] InPrivacy=[Vol,nonACor,FuzzyEn]

[0127] Among them, Vol quantifies the comprehensive volatility of trading power data, nonACor quantifies the non-autocorrelation of trading power data, and FuzzyEn quantifies the information uncertainty of trading power data.

[0128] Because the three indicators all have the same directional characteristics—that is, larger values ​​indicate weaker privacy—we can normalize the combined volatility, non-autocorrelation, and information uncertainty to obtain a normalized dataset. Based on this normalized dataset, a three-dimensional space is then established. The area of ​​a preset size closest to the origin in the three-dimensional space is defined as a strong privacy zone, the area of ​​a preset size farthest from the origin is defined as a weak privacy zone, and the remaining area in the three-dimensional space, excluding the strong and weak privacy zones, is defined as a medium privacy zone. For example, the 1 / 8th of the square closest to the origin in the three-dimensional space can be defined as a strong privacy zone, the 1 / 8th of the square farthest from the origin as a weak privacy zone, and the remaining area as a medium privacy zone.

[0129] In this process, the comprehensive volatility, non-autocorrelation and information uncertainty of trading power data as a time series are comprehensively considered, and a trading power data privacy indicator system in three-dimensional space is constructed to facilitate a clearer observation of the privacy strength and weakness partitioning of different trading power data curves.

[0130] It can be seen that the embodiment of the present application first obtains the comprehensive volatility of the trading power data based on the high-frequency volatility and low-frequency volatility of the trading power data curve, and then uses the rescaled range R / S analysis method to calculate the Hurst index of the trading power data, and uses the Hurst index to obtain the non-autocorrelation parameter of the quantitative trading power data, and then calculates the fuzzy entropy value of the trading power data, and uses the fuzzy entropy value to obtain the information uncertainty parameter of the trading power data; finally, based on the comprehensive volatility, non-autocorrelation and information uncertainty, a trading power data privacy index model is constructed. In this process, the envelope area, time series symbol aggregation, wavelet analysis, Hurst index and fuzzy entropy are used to jointly realize the quantitative characterization of the privacy of the trading power data between the microgrid group and the distribution network, so that the trading power data privacy index model can be constructed based on the inherent privacy attributes of the trading power data, reducing the waste of resources in the privacy protection process.

[0131] In order to make the method provided in the embodiment of the present application clearer and easier to understand, Figure 2 A specific example of this method is described below.

[0132] like Figure 2 As shown, the embodiment may include:

[0133] S201: Calculate the upper and lower envelope areas of the trading power data curve using the Lebesgue integral, and determine the high-frequency volatility of the trading power data curve by taking the average value of the upper and lower envelope areas per unit time.

[0134] In order to utilize the privacy of the power data between the microgrid group and the distribution network due to its inherent characteristics, curve shape and other characteristics, the privacy of the power data between the microgrid group and the distribution network needs to be quantitatively characterized. Figure 3 As shown in the flowchart of the quantitative indicator system for privacy of trading power data, the volatility of trading power data can be determined based on the high-frequency volatility of the trading power data curve. Therefore, the high-frequency volatility of the trading power data curve can be obtained in S201. The specific method for obtaining the upper and lower envelope areas of the trading power data curve can be described in the above embodiment and will not be further described here. In this process, the Lebesgue integral is used instead of the conventional Riemann integral to calculate the upper and lower envelope areas of the trading power data curve, which can solve the problem of discontinuous integration of the trading power data curve.

[0135] S202: Perform time series symbol aggregation processing on the transaction power curve, and calculate the slope value of the character representative value of each period in the discrete character sequence and the character representative value of the previous period as the low-frequency volatility of the transaction power data curve.

[0136] Combine Figure 3 As shown in the flowchart of the quantitative indicator system for privacy of transaction power data, the volatility of transaction power data can be obtained based on the low-frequency volatility of the transaction power data curve. Therefore, the low-frequency volatility of the transaction power data curve can be obtained through S202. The specific process of obtaining the low-frequency volatility of the transaction power data curve can be as shown in the above embodiment and will not be repeated here. In this process, the transaction power data is analyzed using the time series symbol aggregation method, and the time series features of the transaction power data are extracted and converted into a discrete character sequence, thereby achieving a dimensionality reduction effect. The low-frequency volatility of the transaction power data curve can be further calculated.

[0137] S203: Decompose the transaction power data using wavelet analysis to obtain high-frequency information and low-frequency information, and calculate the proportion of high-frequency and low-frequency components based on the high-frequency and low-frequency information.

[0138] Combine Figure 3As shown in the flowchart of the quantitative indicator system for the privacy of trading power data, the volatility of trading power data can be obtained based on the high-frequency volatility data and low-frequency volatility of the trading power data curve. Therefore, in order to obtain the comprehensive volatility, it is necessary to first determine the weight of the high-frequency volatility and the low-frequency volatility. Therefore, it is necessary to use the wavelet analysis method in S203 to decompose the trading power data to obtain high-frequency information and low-frequency information, and calculate the proportion of high- and low-frequency components based on the high-frequency information and low-frequency information. The specific calculation method is shown in the above embodiment and will not be repeated here. In this process, the wavelet analysis method is used to extract high-frequency information and low-frequency information respectively through the wavelet basis function and the scaling function, so as to quantify the comprehensive volatility of the trading power data more comprehensively and accurately.

[0139] S204: Calculate the comprehensive volatility by performing weighted summation of the high-frequency volatility and the low-frequency volatility according to the proportion of the high-frequency and low-frequency components.

[0140] After obtaining the proportions of high- and low-frequency components in S203, the high-frequency volatility and low-frequency volatility can be weighted and summed according to the proportions of the high- and low-frequency components to calculate the comprehensive volatility. The specific weighted summation method is described in the above embodiment and is not further described here. After obtaining the comprehensive volatility of the trading power data, the comprehensive volatility is used as the quantitative volatility of the trading power data, which can effectively characterize the privacy of the trading power data due to its inherent characteristics, curve shape, and other characteristics.

[0141] S205: Calculate the Hurst index of the trading power data using the rescaled range R / S analysis method, and obtain the non-autocorrelation parameter of the trading power data based on the Hurst index.

[0142] Combine Figure 3 As can be seen from the flowchart of the quantitative index system for the privacy of transaction power data, the non-autocorrelation of transaction power data can be obtained based on the Hurst index of the transaction power data. Therefore, the embodiment of the present application uses the rescaled range R / S analysis method to calculate the Hurst index of the transaction power data. The specific calculation method is as shown in the above embodiment and is not repeated here. In this process, the Hurst index of the transaction power data is calculated using the rescaled range R / S analysis method, and the non-autocorrelation of the transaction power data is quantified by the Hurst index, which can effectively characterize the privacy of the transaction power data due to its own inherent characteristics, curve shape, and other characteristics.

[0143] S206: Calculate the fuzzy entropy value of the transaction power data, and obtain the information uncertainty parameter of the transaction power data according to the fuzzy entropy value.

[0144] Combine Figure 3As shown in the flowchart of the quantitative indicator system for privacy of trading power data, the uncertainty of trading power data can be determined based on its fuzzy entropy value. The specific calculation method is described in the above embodiment and is not detailed here. In this process, the concept of fuzzy membership function is utilized to ensure that the fuzzy entropy value changes continuously and smoothly with parameter changes during calculation. Furthermore, the fuzzy entropy is averaged to eliminate the effects of baseline drift and slight data fluctuations, making the characterization of information uncertainty in trading power data more accurate and effectively capturing the privacy of trading power data due to its inherent characteristics, curve shape, and other features.

[0145] S207: Construct a transaction power data privacy indicator model based on the comprehensive volatility, non-autocorrelation parameter and information uncertainty parameter.

[0146] Finally, after obtaining the privacy of the trading power data due to its own intrinsic characteristics, curve shape, etc., namely the comprehensive volatility, non-autocorrelation parameter and information uncertainty parameter, a trading power data privacy indicator model can be constructed. The specific calculation method is shown in the above embodiment and will not be repeated here.

[0147] In this process, envelope area, time series symbol aggregation, wavelet analysis, Hurst exponent and fuzzy entropy are used to quantitatively characterize the privacy of power data traded between microgrids and distribution networks. This effectively characterizes the privacy of power data traded between microgrids and distribution networks due to their inherent characteristics, curve shape and other features, and combines it with predictability to reduce resource waste in the privacy protection process.

[0148] S208: Normalizing the comprehensive volatility, non-autocorrelation, and information uncertainty to obtain a normalized data set, and establishing a three-dimensional space based on the normalized data set to define a strong privacy region, a weak privacy region, and a medium privacy region.

[0149] Since the three indicators have the same directional characteristics, that is, they all have the characteristic that the larger the indicator value, the weaker the privacy, so the comprehensive volatility, non-autocorrelation and information uncertainty can be normalized to obtain a normalized data set, and then a three-dimensional space can be established based on the normalized data set, and the strong and weak privacy partitions of the trading power data can be defined. The specific partitioning method is as follows: Figure 4 As shown, the 1 / 8 area closest to the origin in the three-dimensional space is the strong privacy area, the 1 / 8 area farthest from the origin in the three-dimensional space is the weak privacy area, and the rest is the medium privacy area.

[0150] This embodiment provides a method for constructing a privacy indicator model for transaction power data. This method utilizes envelope area, time series symbol aggregation, wavelet analysis, the Hurst exponent, and fuzzy entropy to quantitatively characterize the privacy of transaction power data between microgrids and distribution networks. This effectively characterizes the privacy inherent in transaction power data between microgrids and distribution networks due to its inherent characteristics and curve shape. This, combined with predictability, reduces resource waste during privacy protection. Furthermore, privacy-focused partitioning can be constructed in three dimensions based on comprehensive volatility, non-autocorrelation, and information uncertainty, facilitating clearer and more precise observation.

[0151] See also Figure 5 The embodiment of the present application provides a device 500 for constructing a transaction power data privacy indicator model, the device comprising:

[0152] An obtaining unit 501 is configured to obtain a comprehensive volatility of the trading power data based on the high-frequency volatility and the low-frequency volatility of the trading power data curve;

[0153] A calculation unit 502 is configured to calculate the Hurst index of the trading power data using a rescaled range R / S analysis method, and obtain a non-autocorrelation parameter of the trading power data based on the Hurst index;

[0154] The calculation unit 502 is further configured to calculate the fuzzy entropy value of the transaction power data and obtain the information uncertainty parameter of the transaction power data according to the fuzzy entropy value;

[0155] The construction unit 503 is used to construct a transaction power data privacy indicator model based on the comprehensive volatility, the non-autocorrelation parameter and the information uncertainty parameter.

[0156] Optionally, the calculation unit 502 is further configured to:

[0157] Calculating the upper and lower envelope areas of the transaction power data curve using Lebesgue integral;

[0158] The determination unit is used to determine the high-frequency volatility of the trading power data curve by taking the average value of the upper and lower envelope areas within a unit time.

[0159] Optionally, the apparatus 500 further includes:

[0160] a processing unit, configured to perform state analysis on the transaction power data using a time series symbol aggregation method, extract a time series of the transaction power data, and convert the time series into a discrete character sequence;

[0161] The calculation unit 502 is further configured to calculate a slope value between a character representative value of each time period and a character representative value of the previous time period in the discrete character sequence, and obtain a low-frequency volatility of the transaction power data curve according to the slope value.

[0162] Optionally, the obtaining unit 501 is specifically configured to:

[0163] Use wavelet analysis to decompose transaction power data to obtain high-frequency information and low-frequency information, and calculate the proportion of high-frequency and low-frequency components based on the high-frequency and low-frequency information;

[0164] The high-frequency volatility and low-frequency volatility are weighted and summed according to the proportion of high and low frequency components to calculate the comprehensive volatility.

[0165] Optionally, when the calculation unit 502 calculates the Hurst index of the trading power data using the rescaled range R / S analysis method, it is specifically used to:

[0166] According to each length L of N different lengths L, the transaction power data is divided into a plurality of interval groups, wherein the interval group includes N adjacent subintervals of length L, where N is a positive integer and L is a positive number;

[0167] Calculate the average rescaled range of N adjacent subintervals in each interval group;

[0168] The Hurst index was calculated by fitting the regression coefficients using the least squares method based on the multiple average rescaled ranges.

[0169] Optionally, when calculating the fuzzy entropy value of the transaction power data, the calculation unit 502 is specifically configured to:

[0170] Reconstruct the transaction power data into an m-dimensional phase space to obtain a vector of phase space dimension m, where m is a positive integer;

[0171] The distance between vector i and vector j in the vector of phase space dimension m is calculated by fuzzy membership function, where i and j are positive integers;

[0172] Get the function φ based on the distance between vector i and vector j m (n, r), where n represents the gradient of the similarity tolerance boundary and r represents the width of the similarity tolerance boundary;

[0173] Add 1 to the phase space dimension m and repeat the operation of obtaining the function to obtain the function φ m+1 (n,r);

[0174] The function φ m (n,r) and function φ m+1 Substitute (n, r) into the fuzzy entropy function and calculate the fuzzy entropy value.

[0175] Optionally, the apparatus 500 further includes:

[0176] A normalization processing unit is used to normalize the comprehensive volatility, non-autocorrelation and information uncertainty to obtain a normalized data set;

[0177] A definition unit is used to establish a three-dimensional space based on a normalized data set, and define the area of ​​a preset size closest to the origin in the three-dimensional space as a strong privacy area, the area of ​​a preset size farthest from the origin in the three-dimensional space as a weak privacy area, and the remaining area in the three-dimensional space other than the strong privacy area and the weak privacy area as a medium privacy area.

[0178] It should be noted that the specific implementation method and the effect achieved by the device 500 for constructing the transaction power data privacy index model can be found in the above Figure 1 or Figure 2 The relevant descriptions in the provided methods will not be repeated here.

[0179] The embodiment of the present application further provides an electronic device 600, such as Figure 6 As shown, the device 600 includes a memory 601 and a processor 602:

[0180] The memory 601 is used to store computer programs;

[0181] The processor 602 is used to execute the above Figure 1 or Figure 2 Provided method.

[0182] In addition, the present application also provides a computer-readable storage medium for storing a computer program for executing Figure 1 or Figure 2 Provided method.

[0183] Through the description of the above embodiments, it can be known that those skilled in the art can clearly understand that all or part of the steps in the above embodiment methods can be implemented by means of software plus a general hardware platform. Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a storage medium, such as a read-only memory (ROM) / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network communication device such as a router) to execute the methods described in each embodiment or certain parts of the embodiments of the present application.

[0184] Each embodiment in this specification is described in a progressive manner. The same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple. For the relevant parts, refer to the partial description of the method embodiment. The device embodiment described above is merely illustrative, in which the modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the goals of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0185] The above description is merely an exemplary embodiment of the present application and is not intended to limit the scope of protection of the present application.

Claims

1. A method for constructing a transaction power data privacy index model, characterized in that: include: According to the high-frequency volatility and low-frequency volatility of the trading power data curve, the comprehensive volatility of the trading power data is obtained; Calculating the Hurst index of the trading power data using a rescaled range R / S analysis method, and obtaining a non-autocorrelation parameter of the trading power data based on the Hurst index; Calculating a fuzzy entropy value of the transaction power data, and obtaining an information uncertainty parameter of the transaction power data according to the fuzzy entropy value; Constructing a transaction power data privacy indicator model based on the comprehensive volatility, the non-autocorrelation parameter, and the information uncertainty parameter; After constructing the transaction power data privacy index model, the comprehensive volatility, the non-autocorrelation and the information uncertainty parameter are normalized to obtain a normalized data set; a three-dimensional space is established based on the normalized data set, and the area of ​​a preset size closest to the origin in the three-dimensional space is defined as a strong privacy area, the area of ​​a preset size farthest from the origin in the three-dimensional space is defined as a weak privacy area, and the remaining area in the three-dimensional space other than the strong privacy area and the weak privacy area is defined as a medium privacy area.

2. The method according to claim 1, characterized in that The process of obtaining the high-frequency volatility of the trading power data curve includes: Calculating the upper and lower envelope areas of the transaction power data curve using Lebesgue integral; The high-frequency volatility of the transaction power data curve is determined by taking the average value of the upper and lower envelope areas within a unit time.

3. The method according to claim 1, characterized in that The process of obtaining the low-frequency volatility of the trading power data curve includes: Performing state analysis on the transaction power data using a time series symbol aggregation method, extracting a time series of the transaction power data, and converting the time series into a discrete character sequence; The slope value between the character representative value of each time period in the discrete character sequence and the character representative value of the previous time period is calculated, and the low-frequency volatility of the transaction power data curve is obtained according to the slope value.

4. The method according to claim 1, wherein Obtaining the comprehensive volatility of the trading power data according to the high-frequency volatility and the low-frequency volatility of the trading power data curve includes: Decomposing the transaction power data using wavelet analysis to obtain high-frequency information and low-frequency information, and calculating the proportion of high-frequency and low-frequency components based on the high-frequency and low-frequency information; The high-frequency volatility and the low-frequency volatility are weighted and summed according to the proportions of the high- and low-frequency components to calculate the comprehensive volatility.

5. The method according to claim 1, wherein The method of calculating the Hurst index of the transaction power data using the R / S analysis method includes: According to each length L of N different lengths L, the transaction power data is divided into a plurality of interval groups, wherein the interval group includes N adjacent subintervals of length L, where N is a positive integer and L is a positive number; Calculate the average rescaled range of N adjacent subintervals in each interval group; The Hurst index is calculated by fitting regression coefficients using the least squares method based on the multiple average rescaled ranges.

6. The method according to claim 1, characterized in that The calculating of the fuzzy entropy value of the transaction power data includes: Performing m-dimensional phase space reconstruction on the transaction power data to obtain a vector of phase space dimension m, where m is a positive integer; Calculating the distance between vector i and vector j in the vector of dimension m of the phase space by a fuzzy membership function, wherein i and j are positive integers; Obtain function φ based on the distance between the vector i and the vector j m (n, r), where n represents the gradient of the similarity tolerance boundary and r represents the width of the similarity tolerance boundary; Add 1 to the phase space dimension m and repeat the operation of obtaining the function to obtain the function φ m+1 (n,r); The function φ m (n, r) and the function φ m+1 (n, r) is substituted into the fuzzy entropy function to calculate the fuzzy entropy value.

7. A device for constructing a transaction power data privacy index system, characterized in that: include: an obtaining unit, configured to obtain a comprehensive volatility of the trading power data according to the high-frequency volatility and the low-frequency volatility of the trading power data curve; a calculation unit, configured to calculate the Hurst index of the trading power data using a rescaled range R / S analysis method, and obtain a non-autocorrelation parameter of the trading power data based on the Hurst index; The calculation unit is further configured to calculate a fuzzy entropy value of the transaction power data and obtain an information uncertainty parameter of the transaction power data according to the fuzzy entropy value; A construction unit, configured to construct a transaction power data privacy indicator model based on the comprehensive volatility, the non-autocorrelation parameter, and the information uncertainty parameter; a normalization processing unit, configured to perform normalization processing on the comprehensive volatility, the non-autocorrelation, and the information uncertainty parameter after constructing the transaction power data privacy indicator model, to obtain a normalized data set; A definition unit is configured to establish a three-dimensional space based on the normalized data set, and define an area of ​​a preset size closest to an origin in the three-dimensional space as a strong privacy area, an area of ​​a preset size farthest from the origin in the three-dimensional space as a weak privacy area, and the remaining area in the three-dimensional space other than the strong privacy area and the weak privacy area as a medium privacy area.

8. An electronic device, characterized in that: The device includes a memory and a processor, and the processor is configured to execute a program stored in the memory and run the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium is used to store a computer program, and the computer program is used to execute the method according to any one of claims 1 to 6.

Images