Privacy protection method and device for power grid data and electronic equipment

Abstract

The invention provides a privacy protection method and device for power grid data and electronic equipment, and the method comprises the steps: obtaining a usage data set of a plurality of power grid users and the perception sensitivity of a plurality of elements in the usage data set, the usage data set comprising a plurality of usage data of the power grid users; partitioning the use data set according to the perceptual sensitivities, and determining a plurality of subsets of the use data set, the plurality of subsets having different perceptual sensitivities; determining a sampling tuple comprising the subset at least according to a preset sampling mechanism of the personalized differential privacy algorithm and the subset; determining noise output of the subset at least according to the sampling tuple and a preset calculation mechanism of a differential privacy algorithm; and according to the noise output, carrying out preset-level noise addition protection on the subset so as to carry out privacy protection on the power grid data. According to the method and the device, the problem that the power grid data privacy protection research in the prior art neglects the difference between users in privacy preference or sensitivity is solved.

Description

Technical Field

[0001] The present application relates to the field of information security technology, and in particular to a method for protecting privacy of power grid data, a device for protecting privacy of power grid data, a computer-readable storage medium, and an electronic device. Background Art

[0002] With the rapid development of smart grid and Internet of Things technologies, power grid systems are also collecting and storing a large amount of user electricity consumption data, which is essential for third parties to conduct energy management, demand forecasting and power grid optimization. However, current research on power grid data privacy protection often ignores the differences in privacy preferences or sensitivity among users, which may lead to privacy protection measures that are either too strict, limiting the rational use of data, or insufficient, putting user privacy at risk. At the same time, if the electricity consumption data that users consider to be non-sensitive or highly public is excessively noised, it will not only increase the cost of privacy protection, but may also reduce the accuracy of data analysis and the efficiency of power grid services. Summary of the invention

[0003] The present application provides a privacy protection method for power grid data, a privacy protection device for power grid data, a computer-readable storage medium and an electronic device to solve the problem in the prior art that the privacy protection of power grid data ignores the sensitivity of user perception and increases the cost of privacy protection.

[0004] According to one aspect of the present application, a method for privacy protection of power grid data is provided, the method comprising: obtaining usage data sets of multiple power grid users and perceptual sensitivities of multiple elements in the usage data sets, the usage data sets comprising multiple usage data of power grid users; partitioning the usage data sets according to the perceptual sensitivities, and determining multiple subsets of the usage data sets, the multiple subsets having different perceptual sensitivities; determining a sampling tuple including the subset at least based on a preset sampling mechanism of a personalized differential privacy algorithm and the subset; determining a noise output of the subset at least based on the sampling tuple and a preset calculation mechanism of the differential privacy algorithm; and performing a preset level of noise protection on the subset according to the noise output to perform privacy protection on the power grid data.

[0005] Optionally, partitioning the usage data set according to the perceptual sensitivity to determine multiple subsets of the usage data set includes: determining categories of the perceptual sensitivity according to the perceptual sensitivity, the categories including high sensitivity, medium sensitivity, low sensitivity and non-sensitivity; partitioning the usage data set according to the categories to determine multiple subsets of the usage data set.

[0006] Optionally, the determining of the sampling tuple including the subset at least according to the preset sampling mechanism of the personalized differential privacy algorithm and the subset includes: determining the privacy level of the subset according to the subset and the perceptual sensitivity corresponding to the subset, and determining a privacy budget according to the privacy level; determining the privacy threshold of the subset according to the privacy budget; determining the elements of the subset according to the usage data set, the privacy threshold and the privacy level to obtain the subset; and determining the sampling tuple including the subset according to the subset, the privacy budget and the privacy threshold.

[0007] Optionally, determining the noise output of the subset at least according to the sampling tuple and a preset calculation mechanism of the differential privacy algorithm includes: obtaining a query function; and determining the noise output of the subset according to the query function, the sampling tuple and the preset calculation mechanism.

[0008] Optionally, the privacy protection method further includes: determining the privacy level of the boundary between adjacent subsets at least based on a differential privacy algorithm model and a symmetric difference between adjacent subsets.

[0009] Optionally, the privacy protection method further includes: determining a privacy budget of boundaries of adjacent subsets based at least on a personalized differential privacy algorithm model, the privacy level and a symmetric difference between adjacent subsets.

[0010] Optionally, the preset sampling mechanism is: Where δ is the privacy threshold corresponding to the category of perceptual sensitivity, ∈ i is the privacy level, and ω(r) is the element in the obtained subset.

[0011] According to another aspect of the present application, a privacy protection device for power grid data is provided, the device comprising: an acquisition module, used to acquire usage data sets of multiple power grid users and perceptual sensitivities of multiple elements in the usage data sets, the usage data sets comprising multiple usage data of power grid users; a first determination module, used to partition the usage data sets according to the perceptual sensitivity, and determine multiple subsets of the usage data sets, the multiple subsets having different perceptual sensitivities; a second determination module, used to determine a sampling tuple including the subset based at least on a preset sampling mechanism of a personalized differential privacy algorithm and the subset; a third determination module, used to determine the noise output of the subset based at least on the sampling tuple and a preset calculation mechanism of the differential privacy algorithm; and a control module, used to perform a preset level of noise protection on the subset based on the noise output, so as to perform privacy protection on the power grid data.

[0012] According to another aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned methods for protecting the privacy of power grid data.

[0013] According to another aspect of the present application, an electronic device is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing the privacy protection of power grid data described in any one of the above.

[0014] Applying the technical solution of the present application, the privacy protection method of the power grid data first obtains the usage data sets of multiple power grid users and the perceived sensitivity of multiple elements in the usage data sets, partitions the usage data sets according to the perceived sensitivity, determines multiple subsets of the usage data sets with different perceived sensitivities, determines a sampling tuple including the subset at least according to the preset sampling mechanism and subset of the personalized differential privacy algorithm, determines the noise output of the subset at least according to the preset calculation mechanism of the sampling tuple and the differential privacy algorithm, and performs a preset level of noise protection on the subset according to the noise output to protect the privacy of the power grid data. By introducing a preset sampling mechanism of sensitivity perception partitioning to realize the personalized differential privacy protection model of sensitivity perception partitioning (Sensitivity Perception Partition based Personalized Differential Privacy, SPPDP, SPPDP), each user in the online power grid user network can individually define the sensitivity level of their social relationship, thereby promoting personalized privacy settings. The above privacy protection method can be run within a trusted service provider, and can establish a privacy protection level according to the sensitivity settings perceived by the user. It solves the problem that the privacy protection of power grid data ignores the sensitivity perceived by the user in the prior art, thereby increasing the cost of privacy protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0016] Figure 1 A hardware structure block diagram of a mobile terminal of a method for protecting privacy of power grid data according to an embodiment of the present application is shown;

[0017] Figure 2 A schematic diagram of a process for protecting privacy of power grid data according to an embodiment of the present application is shown;

[0018] Figure 3 A schematic diagram of a privacy protection device for power grid data according to an embodiment of the present application is shown.

[0019] The above drawings include the following reference numerals:

[0020] 102, processor; 104, memory; 106, transmission device; 108, input and output devices. DETAILED DESCRIPTION

[0021] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0022] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0024] Due to the rapid popularization of smart grid technology, IoT devices and Advanced Metering Infrastructure (AMI), the power grid has become an indispensable key infrastructure in modern society, and the data of the power grid is experiencing unprecedented growth. According to the latest statistics, the number of users of smart grids worldwide has exceeded billions in 2023, a significant increase compared with ten years ago. Many power grid operators and service providers collect a large amount of electricity consumption data from individual users during their operations, and allow third parties (including research institutions and energy service companies) to obtain statistical information of these data by submitting queries to optimize energy distribution, predict demand trends and improve power grid efficiency.

[0025] However, the widespread collection and sharing of personal electricity consumption data in smart grids has caused users to have deep concerns about privacy leakage. Users do not want their sensitive electricity consumption habits, living conditions and other information to be obtained by unauthorized third parties. At the same time, the implementation of international and regional privacy laws and regulations such as the California Privacy Rights Act (CPRA), the Consumer Credit Reporting Agencies Act (CCRA) and the EU's General Data Protection Regulation (GDPR) also emphasizes the importance of personal privacy rights in grid data management. These regulations require grid operators and data processors to follow the principles of transparency, fairness and security when collecting, processing and sharing personal information, ensuring that users can control their data and understand its use.

[0026] Considering the diversity of user privacy preferences and in order to build a more adaptable and user-centric grid data privacy protection framework, researchers began to explore the application of differential privacy (DP) technology in grid data management and proposed the concept of personalized differential privacy (PDP). As an important theoretical basis of DP, personalized differential privacy integrates personalized privacy settings based on user privacy rights and emphasizes the core position of user control and authorization in grid data privacy protection. It breaks the limitation of traditional DP that provides unified global privacy parameters for all users and can flexibly adapt to different privacy needs of users.

[0027] In smart grids, the application of personalized differential privacy is mainly concentrated in two aspects: one is the personalized privacy protection in the query response of electricity consumption data for individual users by adding different degrees of noise according to the privacy preference and sensitivity level of the user; the other is the personalized privacy protection in the process of power grid data release, ensuring that the user's personal privacy can be fully respected and protected when publishing power grid operation data, user electricity consumption statistics, etc. Although there have been some research results in the former, the latter - personalized privacy protection in power grid data release - is still an area that needs to be further studied.

[0028] Giving users the ability to independently specify the sensitivity level of their personal electricity consumption data is an important manifestation of respecting personal privacy (or personalization). This not only helps to enhance users' trust in smart grids, but also promotes the legal and compliant use of grid data, laying a solid foundation for the sustainable development of smart grids.

[0029] The above-mentioned research on privacy protection of power grid data often ignores the differences in privacy preferences or sensitivity among users, which may lead to privacy protection measures that are either too strict, limiting the rational use of data, or insufficient protection, putting user privacy at risk. At the same time, if the electricity consumption data that users consider to be non-sensitive or highly public is excessively noised, it will not only increase the cost of privacy protection, but may also reduce the accuracy of data analysis and the efficiency of power grid services. In order to solve the problem that the research on privacy protection of power grid data in the prior art ignores the differences in privacy preferences or sensitivity among users, the embodiments of the present application provide a privacy protection method for power grid data, a privacy protection device for power grid data, a computer-readable storage medium, and an electronic device.

[0030] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0031] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 1 is a hardware structure block diagram of a mobile terminal of a method for protecting the privacy of power grid data according to an embodiment of the present application. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is for illustration only and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0032] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as computer programs corresponding to the privacy protection method for power grid data in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer programs stored in the memory 104, that is, the above method is implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the mobile terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof. The transmission device 106 is used to receive or send data via a network. The above-mentioned specific examples of the network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0033] In this embodiment, a privacy protection method for power grid data running on a mobile terminal, a computer terminal or a similar computing device is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0034] Figure 2 FIG. 1 is a flow chart of a method for protecting the privacy of power grid data according to an embodiment of the present application. Figure 2 As shown, the method comprises the following steps:

[0035] Step S201 : obtaining usage data sets of a plurality of power grid users and perceived sensitivities of a plurality of elements in the usage data sets, wherein the usage data sets include a plurality of usage data of the power grid users.

[0036] Specifically, by obtaining the usage data sets of multiple power grid users and the perceived sensitivity of multiple elements in the usage data sets, it can help evaluate the impact of different data elements on user privacy, so as to take targeted privacy protection measures. By analyzing the sensitivity of user usage data sets, the data can be encrypted, anonymized or desensitized according to the importance and sensitivity of different data elements to protect user privacy information from being leaked. It can also help discover commonalities and regularities in the data sets and further improve the efficiency of data processing and analysis. By analyzing the usage data of multiple users, we can better understand users' behavioral habits and needs and provide more personalized and effective services for the power grid system.

[0037] Step S202 , partitioning the used data set according to the perceptual sensitivity, and determining a plurality of subsets of the used data set, wherein the plurality of subsets have different perceptual sensitivities.

[0038] Specifically, by using data set partitioning and determining multiple subsets with different perceptual sensitivities, it can help protect the user's privacy information. By dividing the data set into multiple subsets, each with a different perceptual sensitivity, different degrees of data protection and privacy protection measures can be taken for each subset according to actual needs, thereby effectively reducing the risk of data leakage while ensuring that the data can still be effectively utilized. For example, for subsets with higher sensitivity, stricter data encryption, access control, and anonymization measures can be taken to ensure that the data is not accessed by unauthorized personnel, while for subsets with lower perceptual sensitivity, a relatively loose protection strategy can be adopted to reduce costs to ensure the availability and effectiveness of the data.

[0039] Step S203: Determine a sampling tuple including the subset at least according to a preset sampling mechanism and subset of the personalized differential privacy algorithm.

[0040] Specifically, the preset sampling mechanism and subset of the personalized differential privacy algorithm can help determine the sampling tuples that include the subset, thereby achieving privacy protection of the data. By sampling a part of the tuples in the data set and adding noise to them, the data can be confused and disturbed, thereby protecting the privacy of the data. At the same time, the personalized differential privacy algorithm can also provide personalized protection for the data according to the user's personalized privacy needs, effectively balancing the privacy protection of the data and the accuracy of data analysis. Through the design of sampling mechanisms and subsets, the risk of data leakage can be reduced to a certain extent, protecting the user's privacy information.

[0041] Step S204, determining the noise output of the subset at least according to the sampling tuple and the preset calculation mechanism of the differential privacy algorithm.

[0042] Specifically, the differential privacy algorithm protects data privacy by adding an appropriate amount of noise to the query results while trying to maintain the effectiveness of data analysis. By determining the noise output of a subset, the user's electricity usage data privacy can be effectively protected, preventing malicious users or attackers from obtaining the user's personal information or behavioral habits by analyzing the data. By adding noise output, data desensitization can be achieved, thereby ensuring that the user's privacy is effectively protected while maintaining the effectiveness and accuracy of data analysis.

[0043] Step S205 , performing noise protection of a preset level on the subset according to the noise output, so as to perform privacy protection on the power grid data.

[0044] Specifically, by performing a preset level of noise protection on the subset according to the noise output, the privacy of the data can be effectively protected and sensitive information can be prevented from being leaked. By adding noise to the data set, the data can be blurred, making it difficult for attackers to accurately infer the information of the original data, thereby protecting the privacy of users, increasing data security, reducing the risk of data leakage, and ensuring that the privacy of users is not violated. Noise protection can also help data analysts analyze and mine data without leaking sensitive information, thereby maintaining the availability and integrity of the data.

[0045] Through this embodiment, a sampling mechanism including a personalized differential privacy algorithm, a differential privacy algorithm and a sensitivity-aware partition is introduced to implement SPPDP, and a privacy threshold optimization strategy is proposed to optimize the privacy mechanism, thereby achieving an optimal trade-off between personalized privacy protection and improving data utility in the online power grid user network. SPPDP enables each user in the online power grid user network to individually define the sensitivity level of their social relationships, thereby promoting personalized privacy settings. The above method can be run within a trusted service provider and can establish a privacy protection level based on the sensitivity settings perceived by the user. It solves the problem in the prior art that the power grid data privacy protection ignores the sensitivity perceived by the user, thereby increasing the cost of privacy protection.

[0046] It should be noted that the data model of online power grid users can be concisely represented as a tuple D, D = {r1, r2, ..., r n}, each record r i With a user u i In association, tuple D also represents the user's power grid dataset. The graph model adopts an undirected graph paradigm and implements SPDP through the sensitivity-aware sampling mechanism (SASM).

[0047] Specifically, the above-mentioned sensitivity-aware partitioned personalized differential privacy model SPPDP is a differential privacy algorithm based on personalized needs. It dynamically adjusts the level of privacy protection according to the user's privacy sensitivity and data partitioning to achieve personalized privacy protection. It aims to find a better balance between protecting user privacy and improving the value of data use. Personalization is reflected in the seamless integration of the user's self-perception of the sensitivity of electricity consumption data into the differential privacy protection framework of the power grid. Perceived sensitivity means that the user can set the privacy level related to electricity consumption behavior more carefully according to his own situation, and establish a personalized sensitivity classification for his own electricity consumption data, which reflects the user's personal privacy tendency and may be affected by cultural habits, personal values ​​and social environment. Therefore, it is unique and specific for each user.

[0048] Specifically, after obtaining the usage data sets of multiple power grid users, the power grid data set D is input, and the user's perception sensitivity set Then, initialize the sensitivity profile of dataset D, i.e., the sensitive subset and non-sensitive subsets

[0049] Specifically, the sensitivity-aware partitioned personalized differential privacy model SPPDP refers to a given dataset D and a set of sensitive and privacy-aware mechanisms The obtained personalized privacy budget Φ, a random mechanism Satisfies Φ-SPPDP if and The following inequality holds:

[0050]

[0051] Among them, the neighborhood It means to let Represents the dataset D = {r1, r2, ..., r n}Associated user set The perceptual sensitivity of the dataset D' is the neighborhood dataset of D, and there are sensitive records r∈D s So that D'=D{r} or D'=D∪{r}, expressed as in, It is the neighborhood dataset of dataset D.

[0052] Specifically, the above-mentioned noise protection protects the privacy of data by adding a certain amount of noise to the original data, which can effectively prevent the data from being identified or reconstructed by malicious attackers, thereby protecting the user's privacy information. There are many ways to implement noise protection, such as adding random noise to the power grid data, disturbing the timestamp or location information of the data, etc. These noises can blur the characteristics of the original data, thereby achieving the effect of protecting data privacy.

[0053] In the specific implementation process, the data set is partitioned according to the perceptual sensitivity, and multiple subsets of the data set are determined. This can be achieved through the following steps: according to the perceptual sensitivity, the category of perceptual sensitivity is determined, and the categories include high sensitivity, medium sensitivity, low sensitivity and non-sensitivity; the data set is partitioned according to the category, and multiple subsets of the data set are determined.

[0054] In this scheme, by partitioning the used data set according to perceived sensitivity, it can help determine data of different sensitivity levels in the data set and take targeted privacy protection measures. By dividing the data set into high-sensitivity, medium-sensitivity, low-sensitivity and non-sensitive subsets, different privacy protection measures can be taken for each subset to ensure that highly sensitive data receives more stringent privacy protection, while non-sensitive data can take more relaxed privacy protection measures. This can protect data privacy while maintaining data availability and effectiveness as much as possible, effectively improving the privacy protection level of power grid data while reducing the cost and complexity of data processing and management.

[0055] Specifically, based on the sensitivity The steps to partition the data set D are as follows: traverse each record r in the data set D i ∈D, and classify the sensitivity of the edges. If P(r i )=4, then D hs =D hs ∪r; if P(r i )=3, then D ms =D ms ∪r i ; If P(r i )=2, then D ls =D ls ∪r i ; If P(r i )=1, then D ns =D ns ∪r i Therefore, based on the user's perceptual sensitivity set , we can get a sensitivity partition of the data set D, that is, D = D s ∪D ns =D hs ∪Dms ∪D ls ∪D ns .

[0056] Specifically, the above-mentioned perception sensitivity refers to the user u i The perceptual sensitivity of i :r i →{1,2,3,4}, where 1,2,3,4 correspond to the four sensitive states of non-sensitivity (ns), low-sensitivity (ls), medium-sensitivity (ms) and high-sensitivity (hs), respectively. Function P i r i as input and determine its sensitivity level.

[0057] That is, the user's D={r1,r2,…,r n} is divided into the following four different subsets: High sensitivity subset D hs = {r|r∈D, P(r) = 4}, medium sensitivity subset D ms = {r|r∈D, P(r) = 3}, low sensitivity subset D ls = {r|r∈D, P(r)=2} and non-sensitive subset D ns ={r|r∈D,P(r)=1}, so we can get the data set D={r1,r2,…,r n The sensitivity of} is configured as D = D s ∪D ns =D hs ∪D ms ∪D ls ∪D ns .

[0058] In the specific implementation process, at least according to the preset sampling mechanism and subset of the personalized differential privacy algorithm, the sampling tuple including the subset is determined, which can be achieved through the following steps: determine the privacy level of the subset according to the subset and the perceptual sensitivity corresponding to the subset, and determine the privacy budget according to the privacy level; determine the privacy threshold of the subset according to the privacy budget; determine the elements of the subset according to the used data set, the privacy threshold and the privacy level to obtain the subset; determine the sampling tuple including the subset according to the subset, the privacy budget and the privacy threshold.

[0059] In this scheme, by determining the privacy level and privacy budget of the subset, the appropriate privacy threshold can be set according to specific needs, so as to maintain the validity of the data as much as possible while protecting user privacy. According to the privacy threshold and the elements of the subset, the sampling tuple including the subset can be determined to further obfuscate the data to increase privacy protection. Through the personalized differential privacy algorithm, the validity and availability of the data can be retained as much as possible under the premise of protecting user privacy, thereby achieving a balance in data analysis and application. By determining the sampling mechanism and subset, the dimension and sensitivity of the data can be effectively reduced, the risk of data leakage can be reduced, and the security and privacy protection level of the data can be improved.

[0060] Specifically, the above privacy-aware mechanism is a privacy-aware mechanism on the online power grid user network. The record r in D can be i The sensitivity P(r i ) is mapped to the corresponding privacy level ∈ i , and obtain a personalized privacy budget Φ = {∈ i} 1≤i≤n , the operation process of this mechanism is as follows:

[0061] 1. When P(r i )=1or r i ∈D ns ,

[0062] 2. When P(r i ) = 2 or r i ∈D ls ,

[0063] 3. When P(r i )=3or r i ∈D ms ,∈ i In the interval [ b ,∈ c ) to randomly select a value.

[0064] 4. When P(r i ) = 4 or r i ∈D hs ,∈ i In the interval [∈ a ,∈ b ) to randomly select a value.

[0065] Among them, ∈ a ,∈ b ,∈ c >0 is an adjustable parameter and needs to satisfy the condition ∈ a ≤∈ b ≤ c .

[0066] Specifically, when implementing privacy-aware mechanisms After that, the corresponding privacy protection level and privacy budget value are shown in Table 1:

[0067] Table 1 Privacy level customization

[0068]

[0069] Among them, ∈ a =0.5,∈ b =5 and∈ c =10, and Φ = Φ hs ∪Φ ms ∪Φ ls .

[0070] Specifically, the sampling mechanism based on edge sensitivity partitioning is as follows: First, the edge set is divided into four partitions D = D based on the collaborative sensitivity between users. hs ∪D ms ∪D ls ∪D ns , where three sensitive edge partitions D hs , D ms and D ls Corresponding to different privacy budget ranges, non-sensitive data partition D ns Then we can directly publish the statistics. Secondly, we calculate the privacy threshold of each partition based on the privacy budget of each edge partition. hs Based on the privacy budget Φ hs =[0.5,5) is denoted as δ hs , Partition D ms The privacy threshold is denoted as δ ms , Partition D ls The privacy threshold is denoted as δ ls , a sampling mechanism SM is executed for each edge in each edge sensitivity partition f (D,Φ hs ,δ hs ), SM f (D,Φ ms ,δ ms ) and SM f (D,Φ ls ,δ ls ).

[0071] In the specific implementation process, the noise output of the subset is determined at least based on the sampling tuple and the preset calculation mechanism of the differential privacy algorithm, which can be achieved through the following steps: obtaining the query function f; determining the noise output of the subset based on the query function f, the sampling tuple and the preset calculation mechanism.

[0072] In this scheme, the above-mentioned query function can ensure the availability and accuracy of data while protecting user privacy by adopting differential privacy algorithm and noise output. By performing noise output processing on the data through query function and subset, the leakage of sensitive information and re-identification of data can be effectively prevented, thereby protecting user privacy.

[0073] The above query function f may be a query for edge weights.

[0074] It should be noted that the preset sampling mechanism is for the query function Dataset A sampling threshold δ and a personalized privacy budget Φ = {∈1,…,∈ i ,…,∈ n}, let RS(D,Φ,δ) represent the probability of independently sampling each tuple D in the dataset D hs , D ms and D ls process. Data for the entire power grid.

[0075] Specifically, according to the determined RS(D,Φ,δ) data set D, input it into the query function f, and get The sampling result corresponding to the data set D is Returns the noise output of the query function f in, Represents the sampling mechanism of sensitivity-aware partitioning.

[0076] In a specific implementation process, the above privacy protection method further includes the following steps: determining the privacy level of the boundary of adjacent subsets at least based on the differential privacy algorithm model and the symmetric difference of adjacent subsets.

[0077] In this scheme, by using the differential privacy algorithm model and calculating the symmetric difference of adjacent subsets, the degree of privacy protection between different data sets can be evaluated and the privacy level between data sets can be determined. This ensures the accuracy of data analysis results while effectively protecting the user's privacy information, preventing privacy leakage and abuse, and ensuring that power grid data will not leak the user's personal privacy information during use and sharing, thereby improving data security and privacy protection.

[0078] Specifically, the above random algorithm Satisfy∈ i -Edge differential privacy, for any adjacent datasets D1 and D2 and any subset Z of Range(M) (data sequence), the algorithm satisfy:

[0079]

[0080] Among them, |D1ΔD2|=1, Pr(x) represents the probability distribution, and x is the variable.

[0081] In a specific implementation process, the above privacy protection method further includes the following steps: determining the privacy budget of the boundary of adjacent subsets based at least on the personalized differential privacy algorithm model, the privacy level and the symmetric difference of adjacent subsets.

[0082] In this scheme, the privacy of power grid data can be effectively protected through a personalized differential privacy algorithm model to ensure that personal sensitive information is not leaked when the data is used. By determining the privacy budget of the boundaries of adjacent subsets, it can help determine the degree of privacy leakage that can be allowed during data publishing or querying, thereby better balancing the needs of data privacy protection and data analysis.

[0083] Specifically, the personalized differential privacy algorithm is a random algorithm Satisfies Φ-personalized differential privacy PDP if for any adjacent datasets D1 and D1ΔD2=r, and any subset Z of Range(M), algorithm satisfy: Among them, Δ represents the symmetric difference, r∈D represents the difference with user u i The associated record, ∈ i Represents user u i privacy requirements.

[0084] In the specific implementation process, the above preset sampling mechanism is: Where δ is the privacy threshold corresponding to the category of perceptual sensitivity, ∈ i is the privacy level, and ω(r) is the element in the obtained subset.

[0085] In this scheme, by setting different privacy thresholds and privacy levels, the sampling quantity and quality of data can be controlled, thereby protecting user privacy while ensuring the validity and availability of data, effectively reducing the risk of sensitive data leakage, and ensuring the accuracy of data analysis. By randomly selecting subset elements in the data set, the infringement on individual users can be reduced, and the privacy of users can be protected to a certain extent.

[0086] Specifically, according to the privacy threshold δ and the privacy level ∈ i , determine the element r in each sensitivity set D, according to the privacy level ∈ i The corresponding privacy budget Φ, data set D and privacy threshold δ get the sampling result RS(D,Φ,δ). The above sampling mechanism Satisfies Φ-personalized differential privacy, where is any delta-DP mechanism for computing a function f, where D = {r1, r2, …, r n}and Respectively represent the user set The associated record set and perceived sensitivity, r i Indicates that the user u i The associated records, P i Indicated by u i Assign to r i perceptual sensitivity.

[0087] As can be seen from the above, in the embodiments of the present invention, a pioneering study is conducted on sensitivity-aware personalization in an online power grid user network from the perspective of edge granularity, enabling users to perform fine-grained control over the privacy of their social relationships based on the sensitivity perceived by the user; a novel privacy framework, namely the SPPDP model, is proposed, which integrates the user-perceived sensitivity into the edge differential privacy (DP) framework to provide personalized privacy protection for social relationships, and a sensitivity-based privacy-aware mechanism is designed and embedded into a trusted service provider to effectively establish a privacy protection level based on the user's sensitivity setting; a sampling mechanism for sensitivity-aware partitioning is introduced to implement SPPDP, and a privacy threshold optimization strategy is proposed to optimize the privacy mechanism, thereby achieving an optimal trade-off between personalized privacy protection and improved data utility in an online power grid user network; the reliability of the proposed SPPDP is strictly determined, and a series of experiments are conducted using real data sets to verify that, compared with the most advanced DP mechanism, SPPDP provides substantial utility improvement for specific graph queries by utilizing non-sensitive social relationships.

[0088] That is, the technical solution provided by the embodiment of the present invention comprehensively considers multiple factors and introduces a fine-grained, sensitivity-aware partitioned personalized differential privacy model SPPDP for the online power grid user network. Specifically, SPPDP enables each user in the online power grid user network to individually define the sensitivity level of their social relationships, thereby promoting personalized privacy settings. A privacy-aware mechanism is designed, which runs within a trusted service provider and can establish a privacy protection level based on the user-perceived sensitivity setting. In addition, a sensitivity-aware partitioned sampling mechanism is proposed to implement SPPDP. Finally, through theoretical analysis and comprehensive comparative experiments on benchmark data sets, the personalized privacy protection and utility improvement achieved by the SPPDP mechanism are strictly verified.

[0089] The embodiment of the present application also provides a privacy protection device for power grid data. It should be noted that the privacy protection device for power grid data in the embodiment of the present application can be used to execute the determination method for power supply restoration of a faulty distribution network provided in the embodiment of the present application. The device is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.

[0090] The following is an introduction to the privacy protection device for power grid data provided in an embodiment of the present application.

[0091] Figure 3 Schematic diagram of a privacy protection device for power grid data according to an embodiment of the present application. Figure 3 As shown, the device includes: an acquisition module 31 , a first determination module 32 , a second determination module 33 , a third determination module 34 and a control module 35 .

[0092] The acquisition module 31 is used to acquire usage data sets of multiple power grid users and the perceived sensitivity of multiple elements in the usage data sets, where the usage data sets include multiple usage data of the power grid users.

[0093] The first determination module 32 is used to partition the used data set according to the perceptual sensitivity, and determine a plurality of subsets of the used data set, wherein the plurality of subsets have different perceptual sensitivities.

[0094] The second determination module 33 is used to determine a sampling tuple including a subset at least according to a preset sampling mechanism and a subset of the personalized differential privacy algorithm.

[0095] The third determination module 34 is used to determine the noise output of the subset at least according to the sampling tuple and the preset calculation mechanism of the differential privacy algorithm.

[0096] The control module 35 is used to perform noise protection of a preset level on the subset according to the noise output, so as to perform privacy protection on the power grid data.

[0097] Through this embodiment, first, the acquisition module acquires the usage data sets of multiple power grid users and the perceived sensitivity of multiple elements in the usage data sets. The first determination module partitions the usage data sets according to the perceived sensitivity and determines multiple subsets of the usage data sets with different perceived sensitivities. The second determination module determines the sampling tuple including the subset at least according to the preset sampling mechanism and subset of the personalized differential privacy algorithm. The third determination module determines the noise output of the subset at least according to the sampling tuple and the preset calculation mechanism of the differential privacy algorithm. The control module performs a preset level of noise protection on the subset according to the noise output to protect the privacy of the power grid data. By introducing a preset sampling mechanism of sensitivity-aware partitioning to realize the personalized differential privacy protection model of sensitivity-aware partitioning, each user in the online power grid user network can individually define the sensitivity level of its social relationship, thereby promoting personalized privacy settings. The above privacy protection method can be run within a trusted service provider and can establish a privacy protection level according to the sensitivity settings perceived by the user. The problem of increasing the cost of privacy protection due to the neglect of user-perceived sensitivity in the privacy protection of power grid data in the prior art is solved.

[0098] In some instances, the first determination module includes a first determination unit and a second determination unit, the first determination unit being used to determine a category of perceptual sensitivity based on the perceptual sensitivity, the categories including high sensitivity, medium sensitivity, low sensitivity, and non-sensitivity; the second determination unit being used to partition the used data set according to the category and determine multiple subsets of the used data set.

[0099] In this scheme, by partitioning the used data set according to perceived sensitivity, it can help determine data of different sensitivity levels in the data set and take targeted privacy protection measures. By dividing the data set into high-sensitivity, medium-sensitivity, low-sensitivity and non-sensitive subsets, different privacy protection measures can be taken for each subset to ensure that highly sensitive data receives more stringent privacy protection, while non-sensitive data can take more relaxed privacy protection measures. This can protect data privacy while maintaining data availability and effectiveness as much as possible, effectively improving the privacy protection level of power grid data while reducing the cost and complexity of data processing and management.

[0100] In some instances, the second determination module includes a third determination unit, a fourth determination unit, a fifth determination unit, and a sixth determination unit, the third determination unit being used to determine the privacy level of the subset based on the subset and the perceptual sensitivity corresponding to the subset, and to determine the privacy budget based on the privacy level; the fourth determination unit being used to determine the privacy threshold of the subset based on the privacy budget; the fifth determination unit being used to determine the elements of the subset based on the use data set, the privacy threshold, and the privacy level to obtain the subset; and the sixth determination unit being used to determine the sampling tuple including the subset based on the subset, the privacy budget, and the privacy threshold.

[0101] In this scheme, by determining the privacy level and privacy budget of the subset, the appropriate privacy threshold can be set according to specific needs, so as to maintain the validity of the data as much as possible while protecting user privacy. According to the privacy threshold and the elements of the subset, the sampling tuple including the subset can be determined to further obfuscate the data to increase privacy protection. Through the personalized differential privacy algorithm, the validity and availability of the data can be retained as much as possible under the premise of protecting user privacy, thereby achieving a balance in data analysis and application. By determining the sampling mechanism and subset, the dimension and sensitivity of the data can be effectively reduced, the risk of data leakage can be reduced, and the security and privacy protection level of the data can be improved.

[0102] In some examples, the third determination module includes an acquisition unit and a seventh determination unit, the acquisition unit is used to acquire the query function; the seventh determination unit is used to determine the noise output of the subset according to the query function, the sampling tuple and the preset calculation mechanism.

[0103] In this scheme, by adopting differential privacy algorithms and noise output methods, the availability and accuracy of data can be ensured while protecting user privacy. By performing noise output processing on query functions and subsets of data, the leakage of sensitive information and re-identification of data can be effectively prevented, thereby protecting user privacy.

[0104] In some instances, the apparatus further includes a fourth determination module, which is configured to determine a privacy level of a boundary of adjacent subsets based at least on a differential privacy algorithm model and a symmetric difference between adjacent subsets.

[0105] In this scheme, by using the differential privacy algorithm model and calculating the symmetric difference of adjacent subsets, the degree of privacy protection between different data sets can be evaluated and the privacy level between data sets can be determined. This ensures the accuracy of data analysis results while effectively protecting the user's privacy information, preventing privacy leakage and abuse, and ensuring that power grid data will not leak the user's personal privacy information during use and sharing, thereby improving data security and privacy protection.

[0106] In some instances, the apparatus further includes a fifth determination module, which is configured to determine a privacy budget of a boundary of adjacent subsets based at least on a personalized differential privacy algorithm model, a privacy level, and a symmetric difference between adjacent subsets.

[0107] In this scheme, the privacy of power grid data can be effectively protected through a personalized differential privacy algorithm model to ensure that personal sensitive information is not leaked when the data is used. By determining the privacy budget of the boundaries of adjacent subsets, it can help determine the degree of privacy leakage that can be allowed during data publishing or querying, thereby better balancing the needs of data privacy protection and data analysis.

[0108] In some examples, the above preset sampling mechanism is: Where δ is the privacy threshold corresponding to the category of perceptual sensitivity, ∈ i is the privacy level, and ω(r) is the element in the obtained subset.

[0109] In this scheme, by setting different privacy thresholds and privacy levels, the sampling quantity and quality of data can be controlled, thereby protecting user privacy while ensuring the validity and availability of data, effectively reducing the risk of sensitive data leakage, and ensuring the accuracy of data analysis. By randomly selecting subset elements in the data set, the infringement on individual users can be reduced, and the privacy of users can be protected to a certain extent.

[0110] The above-mentioned privacy protection device for power grid data includes a processor and a memory. The above-mentioned acquisition module, the first determination module, the second determination module, the third determination module and the control module are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to implement corresponding functions. The above-mentioned modules are all located in the same processor; or, the above-mentioned modules are located in different processors in any combination.

[0111] The processor includes a kernel, which calls the corresponding program unit from the memory. One or more kernels can be set, and the problem that the existing research on power grid data privacy protection ignores the differences in privacy preferences or sensitivities between users is solved by adjusting kernel parameters.

[0112] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0113] An embodiment of the present application provides a computer-readable storage medium, which includes a stored program, wherein when the program is running, a device where the computer-readable storage medium is located is controlled to execute a privacy protection method for power grid data.

[0114] An embodiment of the present application provides a processor, which is used to run a program, wherein a privacy protection method for power grid data is executed when the program is run.

[0115] A computer program product includes a non-volatile computer-readable storage medium, the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the privacy protection method for power grid data in various embodiments of the present application are implemented.

[0116] The present application also provides an electronic device, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a privacy protection method for power grid data for executing any of the above methods.

[0117] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0118] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0119] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0120] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1A function specified in one or more boxes.

[0121] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0122] In a typical configuration, a computer device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0123] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0124] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0125] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0126] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A privacy protection method for power grid data, characterized in that: The method comprises: Acquire usage data sets of a plurality of grid users and perceived sensitivities of a plurality of elements in the usage data sets, wherein the usage data sets include a plurality of usage data of the grid users; Partitioning the usage data set according to the perceptual sensitivity to determine a plurality of subsets of the usage data set, wherein the plurality of subsets have different perceptual sensitivities; Determine a sampling tuple including the subset at least according to a preset sampling mechanism of the personalized differential privacy algorithm and the subset; Determining a noise output of the subset based at least on the sampled tuple and a preset calculation mechanism of a differential privacy algorithm; The subset is subjected to noise protection at a preset level according to the noise output, so as to perform privacy protection on the power grid data.

2. The privacy protection method according to claim 1, characterized in that: Partitioning the usage data set according to the perceptual sensitivity to determine a plurality of subsets of the usage data set includes: Determining, according to the perceptual sensitivity, a category of the perceptual sensitivity, wherein the category includes high sensitivity, medium sensitivity, low sensitivity, and non-sensitivity; The usage data set is partitioned according to the categories to determine a plurality of subsets of the usage data set.

3. The privacy protection method according to claim 1, characterized in that: The determining, at least according to a preset sampling mechanism of the personalized differential privacy algorithm and the subset, a sampling tuple including the subset comprises: Determining a privacy level of the subset according to the subset and the perceptual sensitivity corresponding to the subset, and determining a privacy budget according to the privacy level; Determining a privacy threshold of the subset according to the privacy budget; Determine the elements of the subset according to the usage data set, the privacy threshold and the privacy level to obtain the subset; A sampling tuple including the subset is determined according to the subset, the privacy budget, and the privacy threshold.

4. The privacy protection method according to claim 1, characterized in that: The determining the noise output of the subset at least according to the sampling tuple and a preset calculation mechanism of the differential privacy algorithm comprises: Get the query function; The noise output of the subset is determined according to the query function, the sampling tuple and the preset calculation mechanism.

5. The privacy protection method according to claim 3, characterized in that: The privacy protection method further includes: The privacy level of the boundary between the adjacent subsets is determined at least according to the differential privacy algorithm model and the symmetric difference between the adjacent subsets.

6. The privacy protection method according to claim 3, characterized in that: The privacy protection method further includes: A privacy budget of a boundary of adjacent subsets is determined based at least on a personalized differential privacy algorithm model, the privacy level, and a symmetric difference between adjacent subsets.

7. The privacy protection method according to claim 1, characterized in that: The preset sampling mechanism is: Among them, δ is the privacy threshold corresponding to the category of perceptual sensitivity, ∈ i is the privacy level, and ω(r) is the element in the obtained subset.

8. A privacy protection device for power grid data, characterized in that: The device comprises: An acquisition module, configured to acquire usage data sets of a plurality of power grid users and perceived sensitivities of a plurality of elements in the usage data sets, wherein the usage data sets include a plurality of usage data of the power grid users; A first determining module, configured to partition the usage data set according to the perceptual sensitivity, and determine a plurality of subsets of the usage data set, wherein the plurality of subsets have different perceptual sensitivities; A second determination module, configured to determine a sampling tuple including the subset at least according to a preset sampling mechanism of the personalized differential privacy algorithm and the subset; A third determination module, configured to determine the noise output of the subset at least according to the sampling tuple and a preset calculation mechanism of the differential privacy algorithm; A control module is used to perform a preset level of noise protection on the subset according to the noise output to perform privacy protection on the power grid data.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the privacy protection method for power grid data according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing the privacy protection method of power grid data described in any one of claims 1 to 7.

Images