Data sharing privacy protection enhancement method and system in power grid cloud environment

Abstract

The invention discloses a data sharing privacy protection enhancement method and system in a power grid cloud environment. The method comprises the steps that a trusted access control center generates a system public key, a system key, a user public and private key pair, an attribute public and private key, an attribute request private key and a re-encryption key; the data owner divides data attributes into a sensitive part and a non-sensitive part, noise meeting differential privacy protection is added to a sensitive field, the sensitive part and the non-sensitive part after noise adding are combined, and a privacy data set is obtained; the data owner uploads the encrypted privacy data set to a cloud server; the cloud service provider converts the ciphertext into the ciphertext of the data user through a re-encryption algorithm; and the data user performs analysis processing by using the decrypted privacy data set. According to the method, the time for the data owner and the data user to access the resources on the cloud is saved, the privacy protection intensity and the data availability are optimized and balanced, and the privacy and the security of the power data in the transmission, storage and application processes are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data security and privacy protection, and more specifically to a method and system for enhancing privacy protection of data sharing in a power grid cloud environment. Background Art

[0002] As an important public infrastructure, the level of management and service industry of the power grid plays a key role in the overall security and stability of the power grid system. In the context of the current digital transformation, the power grid cloud provides a strong infrastructure support for the development of digital power grids, and plays a huge role in the construction of ubiquitous power Internet of Things and smart grid services. However, with the continuous expansion of the scale of the power grid cloud platform, power grid cloud management faces a series of new challenges. The power data stored in the cloud environment contains a large amount of privacy data and sensitive information. If these data are directly hosted on the cloud server, it will bring security risks and privacy threats to the data owner and power grid enterprises to a certain extent. Therefore, how to ensure the security of privacy data in the power grid cloud environment and support relevant businesses to conduct effective data analysis and mining has become one of the hot issues in the field of data security research. Improving the management level of power grid cloud security protection will not only help enhance the defense capability of the power grid system, but also ensure the stability and reliability of power supply, thereby providing a solid guarantee for the normal operation of society.

[0003] Existing privacy-preserving data sharing technologies in power grid cloud environments can be divided into two categories: privacy-preserving methods based on cryptography and privacy-preserving methods based on permission control. Privacy-preserving methods based on cryptography encrypt the original data through technologies such as homomorphic encryption and attribute encryption to ensure the security and privacy of data in the power grid cloud during storage, application, and sharing and exchange. Privacy-preserving methods based on permission control use permission control verification technology to ensure that the identity of power grid cloud visitors is legal and valid, and visitors who pass the verification have the right to access the corresponding services and data. However, methods based on cryptography usually have the problem of reduced availability of data or business applications due to excessive encryption of data, while methods based on permission control face problems such as coarse permission division granularity and inflexible management and maintenance, which makes the use of the above two types of technologies relatively limited in the power grid cloud environment.

[0004] To address the above challenges, the Differential Privacy (DP) model based on data distortion came into being. As an alternative, the DP model theoretically defines the upper limit of privacy leakage and optimizes the trade-off between privacy and availability of published data. To protect the privacy of data in cloud environments, existing technologies combine differential privacy with mechanisms such as order-preserving encryption, homomorphic encryption, and attribute encryption to design access control schemes that meet the requirements of confidentiality, availability, and privacy. However, there is currently little research on combining the DP model with the access control mechanism to protect the security and privacy of data during storage, application, and shared exchange in the power grid cloud environment. The following problems need to be solved:

[0005] 1) The privacy protection method based on cryptography combined with the DP model has problems such as low accuracy and large statistical errors in the encrypted and noisy cloud data, which makes the published data results almost worthless.

[0006] 2) Most existing methods design access control mechanisms based on homomorphic encryption algorithms to improve the security of power grid cloud storage and application data. However, there are problems such as lack of hierarchical and classified management of cloud data and coarse granularity of access to sensitive data, which cannot effectively meet the requirements of data value release and security assurance. Summary of the invention

[0007] In view of this, in order to improve the privacy of data in the power grid cloud environment, the present invention combines the differential privacy model and attribute encryption technology, constructs a proxy re-encryption scheme, and designs and implements a data sharing privacy protection enhancement method and system in the power grid cloud environment. This method aims to enhance the privacy and security of multiple users in the power grid cloud during data sharing and publishing, and solve the problem that the existing technology has low availability of published data results and cannot meet the needs of data value release.

[0008] In order to achieve the above object, the present invention adopts the following technical solution:

[0009] In a first aspect, an embodiment of the present invention provides a method for enhancing privacy protection of data sharing in a power grid cloud environment, comprising the following steps:

[0010] Step 1: System initialization: the trusted access control center generates the system public key, system secret key, user public-private key pair, attribute public-private key pair, attribute request private key, and re-encryption key;

[0011] Step 2: Data preparation: The data owner classifies the data, divides the data attributes into sensitive and non-sensitive parts, adds noise that satisfies differential privacy protection to the sensitive fields, and merges the sensitive and non-sensitive parts after adding noise to obtain a privacy-protected data set.

[0012] Step 3: Data upload: the data owner uploads the encrypted privacy dataset to the cloud server;

[0013] Step 4: Data processing: the cloud service provider converts the ciphertext into the data user's ciphertext through a re-encryption algorithm;

[0014] Step 5: Data usage: The data user receives the ciphertext and decrypts it, and uses the decrypted privacy-preserving data set for analysis and processing.

[0015] Furthermore, the step 1 comprises:

[0016] The trusted access control center selects a bilinear group G0 of order p and a random number y, β, α i ,r∈ R Z p , Z p is a finite field, R represents the set of real numbers; calculate h = g β , Y=e(g,g) y , we get the system public key MPK = G0, g, h, Y, the system secret key MSK = (β, y), and the attribute public key and attribute private key SK i =α i ; Wherein, the generator of G0 is g, e represents the bilinear mapping function, and i represents each attribute i∈U' from the data user attribute set U';

[0017] After receiving the data user's request, the trusted access control center uses the key generation algorithm KeyGen(MSK,U') to calculate the user's public key User private key And set the property list

[0018] The trusted access control center uses the attribute to request the private key generation algorithm Calculate the attribute request private key Update property list

[0019] The re-encryption key generation algorithm ReKeyGen is used to calculate the re-encryption key through the attribute set and access policy to ensure refined access control.

[0020] Furthermore, the step 2 comprises:

[0021] According to the sensitivity of the data, the data owner uses the k-anonymity algorithm to divide the data into sensitive and non-sensitive parts; the sensitive part includes the fields in the data with high privacy risks;

[0022] The noise is generated using the Laplace mechanism and the noise vector is added to the sensitive data; the noise generation formula is: N is the noise vector, s' is the scale parameter, and rn is the random noise drawn from a distribution with scale s';

[0023] The noisy sensitive data is merged with the non-sensitive data to form a privacy-preserving dataset.

[0024] Furthermore, the step 3 comprises:

[0025] The data owner uses a public key encryption algorithm to encrypt the private data set and generate ciphertext; the encryption algorithm is based on homomorphic encryption and attribute encryption;

[0026] The encrypted ciphertext is securely uploaded to the cloud server and stored in an encrypted database.

[0027] Furthermore, the step 4 comprises:

[0028] The cloud service provider first verifies the data user's request and checks whether its attribute set satisfies the access policy W set by the data owner;

[0029] If the attribute set To meet the access policy W, the re-encryption algorithm ReEncrypt is used to convert the data owner's ciphertext into a ciphertext suitable for the data user.

[0030] Furthermore, the step 5 comprises:

[0031] The data user uses his private key to decrypt the re-encrypted ciphertext and restore the private data;

[0032] After decryption, the data user preprocesses the obtained privacy dataset using a normalization function;

[0033] The processed data is used for training and testing machine learning models.

[0034] In a second aspect, an embodiment of the present invention further provides a data sharing privacy protection enhancement system in a power grid cloud environment, comprising:

[0035] System initialization module: The trusted access control center generates system public key, system secret key, user public-private key pair, attribute public-private key pair, attribute request private key and re-encryption key;

[0036] Data preparation module: The data owner classifies the data, divides the data attributes into sensitive parts and non-sensitive parts, adds noise that satisfies differential privacy protection to the sensitive fields, merges the sensitive and non-sensitive parts after adding noise, and obtains a privacy-protected data set;

[0037] Data upload module: The data owner uploads the encrypted privacy data set to the cloud server;

[0038] Data processing module: The cloud service provider converts the ciphertext into the ciphertext of the data user through the re-encryption algorithm;

[0039] Data usage module: Data users receive ciphertext and decrypt it, and use the decrypted privacy data set for analysis and processing.

[0040] It can be seen from the above technical solution that compared with the prior art, it has the following advantages:

[0041] The present invention discloses a method for enhancing privacy protection of data sharing in a power grid cloud environment. Compared with the existing privacy protection data sharing method in a cloud environment, the present invention saves the time of data owners and data users in accessing cloud resources, optimizes the balance between privacy protection strength and data availability, and improves the privacy and security of power data during transmission, storage, and application.

[0042] Furthermore, the beneficial effects achieved by the present invention are as follows:

[0043] (1) A K-anonymity-based preprocessing method is proposed to divide the original data into sensitive and non-sensitive parts, and to manage the data in a hierarchical and classified manner, thereby achieving fine-grained data privacy protection control and improving the accuracy of published data.

[0044] (2) Combining differential privacy and ciphertext policy-based attribute encryption (CP-ABE) technology, an efficient proxy re-encryption scheme for cloud environments is proposed. On the basis of enhancing the privacy and security of pre-processed cloud data, it is suitable for different data publishing and sharing scenarios, and solves the problem of poor availability of data published by current privacy computing technology in cloud environments.

[0045] Through the mutual constraints between cloud service providers and trusted access control centers, data storage and permission authentication are separated to enhance the privacy and confidentiality of data during transmission, storage and use. At the same time, attributes are divided according to the sensitivity level of the original data, and privacy protection strategies are flexibly set to support privacy data protection and application needs in the power grid cloud environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0047] Figure 1This is a flow chart of the data sharing privacy protection enhancement method in the power grid cloud environment provided by the present invention.

[0048] Figure 2 Schematic diagram of the data sharing privacy protection enhancement system in the power grid cloud environment provided by the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] The embodiment of the present invention discloses a method for enhancing privacy protection of data sharing in a power grid cloud environment, referring to Figure 1 As shown, the following steps are included:

[0051] Step 1: System setup. The trusted access control center initializes the system and uses the key generation algorithm KeuGen, the re-encryption key generation algorithm ReKeyGen, and the request attribute private key generation algorithm RequestAttributeSK to generate the system public key, system key, user public-private key pair, attribute public-private key pair, attribute request key, and re-encryption key respectively.

[0052] The specific process of step 1 includes:

[0053] First, the trusted access control center selects a bilinear group G0 with a prime order p and a random number y,β,α i ,r∈ R Z p , Z p is a finite field, R is a real number set. Calculate h = g β , Y=e(g,g) y , we get the system public key MPK = G0, g, h, Y, the system secret key MSK = (β, y), and the attribute public key and attribute private key SK i =α i . Among them, the generator of G0 is g, e represents the bilinear mapping function, and i represents each attribute i∈U' from the data user attribute set U'.

[0054] Then, after receiving the data user's request, the trusted access control center uses the key generation algorithm KeyGen(MSK,U') to calculate the user's public key User private key And set the property list Empty Set Among them, U' is the attribute set that the data user needs to submit when initiating a private key request.

[0055] Secondly, the trusted access control center uses the attribute to request the private key generation algorithm Calculate the attribute request private key Update property list

[0056] Finally, the trusted access control center is based on the attribute set and satisfy the access policy W, select a random number d,g1∈ R Z p , using the re-encryption key generation algorithm ReKeyGen(MPK,W,W',O i ) Calculate the ciphertext under access policy W' and the access control tree in, is the generator of the number of embedded attributes n, O i Request the private key for the attribute, that is, the root node value of the access control tree T, O i,j is the leaf node set, v i,j Generate the re-encryption key RK for the leaf node value AS→W' = <C,T,g r >, where n = |AS| = |W| represents the number of attributes in the user attribute set, and W' represents the new access policy.

[0057] Step 2: Data preparation. The data owner divides the data attributes into sensitive and non-sensitive parts according to the data classification standard, adds noise that satisfies differential privacy protection to the sensitive fields, and merges the sensitive and non-sensitive parts after adding noise to obtain a privacy-protected data set.

[0058] Step 2: Data preparation. The specific process includes:

[0059] First, after receiving the data classification standard sent by the trusted access center, the data owner uses the k-anonymity algorithm to divide the data set D = {D1, D2, ..., D x The sensitivity level of}(i∈[1,x]) divides it into sensitive parts and non-sensitive parts That is, D1, D2, ..., D x Divided into Among them, D i ∈D represents a data record of the data set D. Each data record D1,D2,…,D xContains x Attributes t x represents the number of attributes in each group, p,q,k,s∈Z, represents the sensitive part of the dataset, Represents the non-sensitive part of the data set. NS is the abbreviation of non-sensitive.

[0060] Second, to enhance the privacy of sensitive data, the data owner uses the probability density function Generate Laplace noise N = {N1, N2, ..., N b}(i∈[1,b], N represents the noise vector set, N i Represented as each data record D i The added noise vector, s' is the scale parameter, and rn represents the random noise drawn from a distribution with scale s', which is added to the sensitive part In the example above, we can obtain data that satisfies differential privacy protection. i∈[1,b]. After adding noise, and Merge to obtain a privacy-preserving dataset

[0061] Step 3: Data upload. The data owner uses the encryption algorithm Encrypt to encrypt the privacy dataset and upload the encrypted result to the cloud server;

[0062] Step 3 The specific process includes:

[0063] The data owner selects element l∈ R Z p , using encryption algorithm Calculate ciphertext And upload it to the cloud server. represents the privacy-protected dataset after adding Laplace noise, W represents the access policy, and PK i (i∈[1,F]) represents the attribute public key of each data owner, C2=g l , C4=(h) l =g βl , p∈W represents the access policy.

[0064] Step 4: Data processing. Based on the request sent by the data user, the cloud service provider uses the re-encryption algorithm ReEncrypt to convert the data owner's ciphertext into the data user's ciphertext and sends it to the data user;

[0065] Step 4 The specific process includes:

[0066] After receiving the ciphertext uploaded by the data owner and the data user's request, the cloud service provider checks whether the requested data user has the attribute set Satisfy access policy W. If the access conditions are met, use the re-encryption algorithm The data owner's ciphertext Convert to ciphertext for data users Where T represents the access control tree, n represents the number of attributes in the user attribute set, l∈ R Z p is the random number selected in step 3, d∈ R Z p is the random number selected in step 1.

[0067] Step 5: Data use. After receiving the ciphertext, the data user uses the decryption algorithm Decrypt to decrypt it, and processes and analyzes the decrypted privacy data set based on the machine learning algorithm.

[0068] Step 5 The specific process includes:

[0069] First, the data user uses the decryption algorithm Decrypt the ciphertext, assuming And AS=W,

[0070] Secondly, data users obtain privacy-protected data sets. Then, use the normalization function Perform preprocessing to obtain the preprocessed data set T A is a training sample with A attributes, μ and σ are the mean and standard deviation of the training samples respectively. Belong to l * ≤n type label in

[0071] Then, the data user will use the preprocessed data Divide into training data and test data And use the training data The classification model is trained with machine learning algorithms, and the accuracy of the classification model is measured by testing data. For evaluation, data sample Assign class labels to classification models during testing.

[0072] Finally, data users will Output as class label vector and use Calculate the accuracy of the classification model, where CI represents the number of correctly classified samples and TI represents the total number of test samples.

[0073] The above five steps cover the entire process of enhancing privacy protection of data sharing in the entire power grid cloud environment, including the entire process from key generation, data encryption to re-encryption and data analysis. This method separates data storage and permission authentication through the mutual constraints of cloud service providers and trusted access control centers to enhance the privacy and confidentiality of data during transmission, storage and use. At the same time, the attributes are divided according to the sensitivity level of the original data, and the privacy protection strategy is flexibly set to support the privacy data protection and application needs in the power grid cloud environment.

[0074] For example:

[0075] Assume that there is a power grid company (data owner) that wants to share power consumption data with multiple research institutions (data users) for analysis in order to improve the efficiency of the power grid system. However, the power consumption data contains a large amount of personal privacy information, such as users' power usage habits, location data, etc. In order to protect user privacy and ensure security during data sharing, the company uses the data sharing privacy protection enhancement method in the power grid cloud environment of the present invention.

[0076] 1: System initialization

[0077] The power grid company decided to use a privacy protection method based on differential privacy and attribute encryption. Before sharing data, the system needs to be initialized:

[0078] Public and private key generation: The trusted access control center of the power grid company generates the system public key MPK and master key MSK for encryption and decryption operations. Each research institution will also obtain the corresponding user public and private key pair.

[0079] For example, the power grid company generates a public key MPK = G0, g, h, Y, and generates a user public key for a research institution. and private key

[0080] Attribute private key generation: Access control policies are set for different research institutions, and attribute private keys are generated based on the attributes of the research institutions.

[0081] For example, a research institute has access to power data analysis, and the system generates a corresponding attribute private key for it. And use it for subsequent data decryption.

[0082] 2: Data preparation

[0083] The power grid company classifies the user's electricity consumption data and processes the sensitive parts with differential privacy protection:

[0084] Data classification and grading: The company divides data into sensitive and non-sensitive parts based on the sensitivity of user data.

[0085] For example, a user's home address and specific time of electricity usage are classified as sensitive data, while the average daily electricity consumption is non-sensitive data.

[0086] Adding differential privacy noise: To ensure privacy, the power grid company adds differential privacy noise to users’ sensitive data.

[0087] For example, assuming a user's daily electricity consumption is 12kWh, the company adds Laplace noise to this data, making the final published value 120.5kWh, ensuring that attackers cannot infer the user's specific electricity consumption through the data.

[0088] Privacy dataset generation: The noisy sensitive data is merged with the non-sensitive data to generate a privacy dataset

[0089] 3: Data upload

[0090] The power grid company uses an encryption algorithm to encrypt the privacy data set and upload it to the cloud server:

[0091] Encrypted private data: The power grid company uses the public key MPK to encrypt the private data set and generate ciphertext

[0092] For example, the encrypted power consumption data contains four ciphertext parts: C1, C2, C3, and C4.

[0093] Upload to cloud server: The encrypted data is uploaded to the power grid cloud server for analysis by multiple research institutions.

[0094] 4: Data Processing

[0095] A research institute wants to analyze the data, so it sends an access request to the power grid company. The cloud service provider performs re-encryption operations based on its attributes and permissions:

[0096] Request verification: The cloud server verifies whether the attribute set of the research institution meets the access control policy set by the power grid company.

[0097] For example, the attribute of a research institute is "power data analysis", and the cloud server checks whether it has this permission.

[0098] Re-encrypt ciphertext: After verification, the cloud server uses the re-encryption key to Converted to ciphertext that can be decrypted by research institutions

[0099] 5: Data Usage

[0100] After receiving the re-encrypted ciphertext, the research institute decrypts and analyzes the data:

[0101] Ciphertext decryption: Research institutions use their private keys Ciphertext Decrypt and restore the private power data.

[0102] For example, after decryption, the research institute obtained the user's private electricity consumption data as 12kWh.

[0103] Data analysis: Research institutions use the decrypted electricity data for analysis, for example, to predict future electricity demand through machine learning models.

[0104] For example, the agency uses decrypted data to conduct cluster analysis to identify electricity consumption patterns in different regions to help power grid companies optimize power supply.

[0105] Through the above methods, the power grid company successfully achieved the following goals: securely shared power consumption data without leaking user privacy. Used differential privacy protection to ensure that user privacy would not be leaked even if the encrypted data was attacked. Research institutions were able to effectively analyze the shared data, increasing the value of the data.

[0106] This example shows how to ensure data security and privacy during sharing and analysis in a power grid cloud environment through differential privacy, attribute encryption, and re-encryption mechanisms. This method is suitable for data sharing scenarios between power grid companies and multiple entities such as research institutions and governments, and effectively balances data availability and privacy protection.

[0107] In addition, based on the above embodiments, the present invention also provides a data sharing privacy protection enhancement system in a power grid cloud environment, such as Figure 2 shown.

[0108] Taking the power data sharing and analysis scenario as an example, the data owner It is designed to share power data with cloud service providers (CSPs) for data users, and to provide data sharing services to various organizations, industry regulatory departments, and important enterprises. In this scenario, the CSP is an honest and curious entity, that is, it will strictly follow the protocol to store and process the data owned by the data owner. Uploaded data and provide it to data users Provide privacy-protected data for analysis, but due to curiosity about information related to the original data, try to use various means to learn the received data and spy on the privacy or sensitive information contained in the data. The Trusted Access Control Center (TA) generates keys and re-encryption keys for legitimate data owners and data users.

[0109] The specific steps include:

[0110] Step 1: System initialization. First, the trusted access control center uses the system public key, system key, public parameters, master key and key generation algorithm KeyGen to initialize the system for the data owner P. i and data user U j Generate a public-private key pair Then, the re-encryption key RK is generated using the attribute private key, access policy, system public key, and re-encryption key generation algorithm ReKeyGen AS→W' ; Finally, the trusted access control center classifies the data into classification standards, public keys, Sent to data owner P i , private key Sent to data user U j , re-encryption key RK AS→W' Sent to the cloud service provider.

[0111] Step 2: Data preparation. First, the data owner P i After receiving the data classification standard sent by the trusted access center, the k-anonymity algorithm is used to separate the data sets D i Divide sensitive parts and non-sensitive parts Then, the probability density function is used to generate Laplace noise N to protect the sensitive part D S , get data that satisfies differential privacy protection Finally, the data owner will and Merge to obtain a privacy-preserving dataset

[0112] Step 3: Upload data. Data owner P i Using Public Key Encrypt the privacy dataset using the encryption algorithm Encrypt Get the ciphertext Upload it to the cloud server.

[0113] Step 4: Data processing. First, the cloud service provider receives data from the data user U jThe access request is received and the requested data user is checked to see whether the access policy is met; then, the re-encryption algorithm ReEncrypt is used to convert the ciphertext of the data owner into Convert to ciphertext for data users Finally, the ciphertext Sent to the corresponding data user U j .

[0114] Step 5: Obtain and use data. First, the data user uses the decryption algorithm Decrypt to decrypt the ciphertext and obtain the privacy data set. Then, use the normalization function to Perform preprocessing to obtain the preprocessed data set Finally, using the training data and test data Train classification models to analyze private data. For example, integrate private electricity data with data from industries such as industry and finance to provide green transformation support for environmental pollution monitoring, green bond financing, and other fields.

[0115] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0116] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for enhancing privacy protection of data sharing in a power grid cloud environment, characterized in that: The following steps are involved: Step 1: System initialization: the trusted access control center generates the system public key, system secret key, attribute public-private key pair, user public-private key pair, attribute request private key, and re-encryption key; Step 2: Data preparation: The data owner classifies the data, divides the data attributes into sensitive and non-sensitive parts, adds noise that satisfies differential privacy protection to the sensitive fields, and merges the sensitive and non-sensitive parts after adding noise to obtain a privacy-protected data set. Step 3: Data upload: the data owner uploads the encrypted privacy dataset to the cloud server; Step 4: Data processing: the cloud service provider converts the ciphertext into the data user's ciphertext through a re-encryption algorithm; Step 5: Data usage: The data user receives the ciphertext and decrypts it, and uses the decrypted privacy-preserving data set for analysis and processing.

2. According to the method for enhancing data sharing privacy protection in a power grid cloud environment according to claim 1, it is characterized in that: The step 1 comprises: The trusted access control center selects a bilinear group G0 of order p and a random number y,β,α i ,r∈ R Z p , Z p is a finite field, R represents the set of real numbers; calculate h = g β , Y=e(g,g) y , we get the system public key MPK = G0, g, h, Y, the system secret key MSK = (β, y), and the attribute public key and attribute private key SK i =α i ; Wherein, the generator of G0 is g, e represents the bilinear mapping function, and i represents each attribute i∈U' from the data user attribute set U'; After receiving the data user's request, the trusted access control center uses the key generation algorithm KeyGen(MSK,U') to calculate the user's public key User private key And set the property list The trusted access control center uses the attribute to request the private key generation algorithm Calculate the attribute request private key Update property list The re-encryption key generation algorithm ReKeyGen is used to calculate the re-encryption key through the attribute set and access policy to ensure refined access control.

3. According to the method for enhancing data sharing privacy protection in a power grid cloud environment according to claim 2, it is characterized in that: The step 2 comprises: According to the sensitivity of the data, the data owner uses the k-anonymity algorithm to divide the data into sensitive and non-sensitive parts; the sensitive part includes the fields in the data with high privacy risks; The noise is generated using the Laplace mechanism and the noise vector is added to the sensitive data; the noise generation formula is: N is the noise vector, s' is the scale parameter, and rn is the random noise drawn from a distribution with scale s'; The noisy sensitive data is merged with the non-sensitive data to form a privacy-preserving dataset.

4. According to the method for enhancing data sharing privacy protection in a power grid cloud environment according to claim 3, it is characterized in that: The step 3 comprises: The data owner uses a public key encryption algorithm to encrypt the private data set and generate ciphertext; the encryption algorithm is based on homomorphic encryption and attribute encryption; The encrypted ciphertext is securely uploaded to the cloud server and stored in an encrypted database.

5. According to the method for enhancing data sharing privacy protection in a power grid cloud environment according to claim 4, it is characterized in that: The step 4 comprises: The cloud service provider first verifies the data user's request and checks whether its attribute set satisfies the access policy W set by the data owner; If the attribute set To meet the access policy W, the re-encryption algorithm ReEncrypt is used to convert the data owner's ciphertext into a ciphertext suitable for the data user.

6. A method for enhancing privacy protection of data sharing in a power grid cloud environment according to claim 5, characterized in that: The step 5 comprises: The data user uses his private key to decrypt the re-encrypted ciphertext and restore the private data; After decryption, the data user preprocesses the obtained privacy dataset using a normalization function; The processed data is used for training and testing machine learning models.

7. A data sharing privacy protection enhancement system in a power grid cloud environment, characterized in that: include: System initialization module: The trusted access control center generates system public key, system secret key, user public-private key pair, attribute public-private key pair, attribute request private key and re-encryption key; Data preparation module: The data owner classifies the data, divides the data attributes into sensitive parts and non-sensitive parts, adds noise that satisfies differential privacy protection to the sensitive fields, merges the sensitive and non-sensitive parts after adding noise, and obtains a privacy-protected data set; Data upload module: The data owner uploads the encrypted privacy data set to the cloud server; Data processing module: The cloud service provider converts the ciphertext into the ciphertext of the data user through the re-encryption algorithm; Data usage module: Data users receive ciphertext and decrypt it, and use the decrypted privacy data set for analysis and processing.

Images