A power data security sharing method based on privacy calculation and related equipment

Abstract

The application belongs to the technical field of electric power data security and data sharing, and discloses an electric power data security sharing method based on privacy calculation and related equipment. The hybrid storage system solves the problem of limited storage capacity of the blockchain. The on-chain storage metadata guarantees traceability, and the off-chain storage encrypted data improves storage efficiency. The improved BFT consensus algorithm guarantees the consistency of the distributed nodes and solves the trust problem of cross-organization data sharing. The collaborative computing protocol of secret sharing and homomorphic encryption takes into account the privacy protection strength and the computing and communication efficiency. Compared with the traditional single privacy calculation scheme, the computing efficiency is improved, the communication cost is reduced, the proxy re-encryption technology realizes controllable sharing of data, and the compliance requirements are met. The application can effectively support sensitive data sharing scenes such as electric power load prediction and cross-department collaborative analysis, and can guarantee that sensitive information such as power grid topology and user power consumption behavior is not leaked.

Description

Technical Field

[0001] This invention belongs to the field of power data security and data sharing technology, specifically a method and related equipment for secure power data sharing based on privacy computing. Background Technology

[0002] With the construction of new power systems and the deep integration of the digital economy, power data, as a core production factor, is increasingly demonstrating its cross-domain collaborative value. Whether it's cross-regional power load forecasting, multi-departmental data collaborative analysis, or third-party scientific research cooperation, efficient and secure data sharing is indispensable. However, power data contains sensitive information such as grid topology and user electricity consumption behavior. Data sharing faces three core challenges: privacy risks, strict compliance constraints, and significant performance bottlenecks, severely restricting the release of the value of data elements.

[0003] Existing power data sharing solutions have significant limitations: pure homomorphic encryption schemes support direct computation of ciphertext, but suffer from severe ciphertext bloat and low computational efficiency, making them unsuitable for large-scale power data processing scenarios; traditional secret sharing schemes offer high privacy protection, but rely on complex communication protocols, and communication overhead increases exponentially with the number of participants; while blockchain technology can ensure data traceability, its storage capacity is limited and cannot support the computational needs of large-scale data; federated learning schemes avoid the transmission of raw data, but there is a risk of gradient leakage during model aggregation, and their cross-scenario adaptability is insufficient.

[0004] Meanwhile, relevant laws and regulations have imposed stricter requirements on the compliance and privacy protection of power data sharing. Existing single privacy computing technologies are insufficient to simultaneously meet the security, efficiency, and availability requirements of power data sharing. There is an urgent need to build a hybrid architecture that integrates the advantages of multiple privacy computing technologies to resolve the contradiction between the performance bottlenecks and insufficient privacy protection of traditional solutions. Summary of the Invention

[0005] This invention provides a method and related equipment for secure sharing of power data based on privacy computing, which solves the contradiction between the performance bottleneck and insufficient privacy protection of traditional solutions.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A method for secure sharing of power data based on privacy-preserving computation, comprising: Acquire raw power-sensitive data and perform data processing; The standardized raw data is split into two parts using a secret sharing protocol and stored on different distributed nodes off-chain. The metadata of the original power-sensitive data is extracted, re-encrypted, and then stored on-chain. Based on the improved BFT consensus algorithm, consistency verification is performed on distributed nodes; The identity information and computing needs submitted by data users through the blockchain are verified. If they are legitimate, the corresponding share of secret data is obtained from the off-chain storage node. Based on computational requirements, an improved Paillier homomorphic encryption algorithm is used to perform linear operations, gradient aggregation, and model training in the ciphertext domain. Parallel computation is achieved by combining the MapReduce framework to obtain the computational requirements of data users.

[0007] Preferably, the steps of splitting the standardized original data into two shares using a secret sharing protocol, storing them separately on different distributed nodes off-chain, extracting the metadata of the original power-sensitive data, re-encrypting it, and then storing it on-chain are as follows: For the standardized raw data, one share is randomly selected from the interval corresponding to the safe prime number. The other share is obtained by modulo operation between the raw data and the random share. For matrix-type power data, an element-level splitting method is used. The two data shares are stored on different off-chain distributed storage nodes. A single node only holds a portion of the shares and cannot deduce the original data. At the same time, metadata is extracted, encrypted using a proxy re-encryption algorithm, and then uploaded to the consortium blockchain node built on Hyperledger Fabric. Preferably, the consistency verification of distributed nodes based on the improved BFT consensus algorithm specifically involves: The fault-tolerance formula of the improved BFT algorithm is as follows: Tolerates one faulty node; The verification process is as follows: The master node rotates according to preset rules. After receiving a data sharing request, it generates a proposal containing block height, view number, proposal initiator, and encrypted metadata information. After the slave node verifies the legality of the proposal, it sends a confirmation message. After receiving 2f confirmation messages, the master node broadcasts a commit message. After each node receives 2f+1 commit messages, it confirms the block and updates its local ledger, thus completing the consensus process. When a master node failure causes a slave node to not receive a proposal within a preset timeout period, the slave node automatically initiates a view switching request. After receiving 2f identical requests, it switches to the new view.

[0008] Preferably, the steps to obtain the data user's computational requirements are as follows: Based on computational needs, the improved Paillier homomorphic encryption algorithm is used to perform linear operations, gradient aggregation, and model training in the ciphertext domain, combined with the MapReduce framework to achieve parallel computation. The corresponding secret shares are downloaded from the off-chain storage node. The participating nodes holding the two secret shares generate improved Paillier homomorphic encryption key pairs and exchange public keys. They encrypt their own vector shares using the other party's public key and send them to each other. Then, each node generates a random vector, calculates the product of its local matrix share and the other party's vector share, subtracts the random vector, and sends the encrypted result to each other a second time. The node decrypts the received encrypted intermediate result, calculates its local share by combining the product of its own matrix and vector share with its local random vector, and then superimposes the two local shares and takes the modulo of a safe prime number to obtain the product of the target matrix and vector. In the training of the power load forecasting model, based on the mini-batch gradient descent algorithm and combined with the MapReduce framework, the large-scale computing task is split into 8 parallel partitions, which are assigned to 2 Slave nodes for execution, and the Master node is responsible for task scheduling and result aggregation.

[0009] Preferably, acquiring and processing the raw power-sensitive data specifically involves: The original power-sensitive data is first standardized, and then the standardized numerical features are mapped to the [0,1] interval.

[0010] Preferably, the method further includes: when a data user needs to authorize a third party to participate in the computation, the data provider first verifies the legitimacy of the third party's identity and the scope of access permissions through the blockchain consensus node. After confirming that it meets the preset rules, the third party generates a unique re-encryption key using its own private key and the third party's public key. This key is sent through a trusted blockchain channel without disclosing the private keys of either party. After obtaining the re-encryption key, the third party can convert the original encrypted ciphertext into a new ciphertext adapted to its own public key without the data provider's participation. The conversion process does not involve exposing the original plaintext. Only the third party can decrypt the new ciphertext using its own private key to obtain the right to use the data. Unauthorized users cannot crack the ciphertext even if they obtain it.

[0011] A privacy-preserving computing-based secure power data sharing system includes: Data acquisition module: used to acquire raw power-sensitive data and perform data processing; The split storage module is used to split the standardized original data into two shares using a secret sharing protocol, and store them separately on different distributed nodes off-chain. It also extracts the metadata of the original power-sensitive data, re-encrypts it, and stores it on-chain. Verification module: Used to verify the consistency of distributed nodes based on the improved BFT consensus algorithm; Verification module: Used to verify the identity information and computing needs submitted by data users through the blockchain. If they are valid, the corresponding secret data share is obtained from the off-chain storage node. The computation module is used to perform linear operations, gradient aggregation, and model training in the ciphertext domain based on computational needs using an improved Paillier homomorphic encryption algorithm. It combines the MapReduce framework to achieve parallel computation and obtain the computational needs of data users.

[0012] A computer device includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement steps of a method for secure sharing of power data based on privacy computing.

[0013] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a method for secure sharing of power data based on privacy computing.

[0014] A computer program product includes a computer program that, when executed by a processor, implements steps of a method for secure sharing of power data based on privacy computing.

[0015] Compared with existing technologies, this invention has the following advantages: It provides a method for secure sharing of power data based on privacy computing. The hybrid storage system solves the problem of limited storage capacity in blockchain; on-chain storage of metadata ensures traceability; off-chain storage of encrypted data improves storage efficiency; and the improved BFT consensus algorithm ensures the consistency of distributed nodes, resolving trust issues in cross-organizational data sharing. The collaborative computing protocol combining secret sharing and homomorphic encryption balances privacy protection strength with computational and communication efficiency. Compared to traditional single privacy computing schemes, it improves computational efficiency and reduces communication overhead. The proxy re-encryption technology enables controllable data sharing, meeting compliance requirements. It can effectively support sensitive data sharing scenarios such as power load forecasting and cross-departmental collaborative analysis, fully releasing data value while ensuring the non-disclosure of sensitive information such as power grid topology and user electricity consumption behavior, providing secure and reliable technical support for the market-oriented allocation of power data elements.

[0016] Furthermore, this invention constructs a hybrid storage system combining on-chain indexing and off-chain storage, employing a 2-out-of-2 secret sharing protocol to split the original power-sensitive data. Targeted splitting strategies are used for scalar and matrix data, with each node holding only a portion of the data share, making it impossible to deduce the original data and ensuring privacy and security during data storage. Simultaneously, metadata such as data indexes and access permissions are re-encrypted via a proxy before being uploaded to the blockchain. This avoids the capacity limitations of storing large amounts of original data on the blockchain and ensures full traceability of data flow through the immutability of on-chain data. The encrypted metadata further prevents the risk of leakage of access information.

[0017] Furthermore, this invention designs and improves the Byzantine fault-tolerant consensus algorithm by optimizing the three-stage process of proposal, confirmation, and submission, as well as the view switching mechanism, thereby enhancing the efficiency and stability of distributed node collaboration. The algorithm's fault tolerance capability meets the requirements. With 5 consensus nodes, it can tolerate 1 faulty node, which is suitable for the trust requirements of multiple participants in power data sharing. The view switching adopts a timeout trigger strategy. When a node does not receive a proposal, it automatically initiates a switching request. After receiving 2f identical requests, the view update is completed, which effectively reduces the view switching overhead of the traditional PBFT algorithm and ensures the continuity of the consensus process and the reliability of data flow.

[0018] Furthermore, this invention proposes a secret-sharing-homomorphic encryption collaborative computing protocol, which integrates the advantages of two privacy computing technologies while avoiding their respective limitations. The protocol avoids the high-complexity OT protocols in traditional secure multi-party computation, achieving secure multiplication of secret-shared values ​​through a combination of additive homomorphic encryption and addition masking, reducing communication overhead by 35% compared to pure MPC schemes. The improved Paillier homomorphic encryption algorithm supports addition and scalar multiplication operations in the ciphertext field, adapting to the linear computation requirements in power data collaborative analysis. Combined with the MapReduce framework, large-scale computational tasks are split and executed in parallel, improving computational efficiency by more than 40% compared to traditional fully homomorphic encryption algorithms, and supporting power data processing on a scale of millions of samples.

[0019] Furthermore, this invention employs proxy re-encryption technology to achieve controllable sharing of power data, constructing a complete authorization chain encompassing key generation, ciphertext conversion, and authorization / decryption. The data provider generates a unique re-encryption key using its own private key and the authorized user's public key, completing authorization without needing to know the authorized user's private key. The authorization process does not involve decrypting the original data, ensuring the data remains encrypted throughout. Authorized users convert the original ciphertext into ciphertext corresponding to their own public key using the re-encryption key, which can only be decrypted and used with their own private key; unauthorized users, even if they obtain the ciphertext, cannot decrypt it. Key operation logs throughout the authorization process are stored on the blockchain in real time, covering information such as permission changes and ciphertext conversion trajectories. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart of a method for securely sharing power data based on privacy computing, according to an embodiment of the present invention. Figure 2This is a schematic diagram of a method for securely sharing power data based on privacy computing, according to an embodiment of the present invention. Figure 3 This is a block diagram of a power data security sharing system based on privacy computing, according to an embodiment of the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0023] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0024] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0025] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0026] like Figure 1 As shown, this invention provides a method for secure sharing of power data based on privacy computing, comprising: S1: Acquire raw power-sensitive data and perform data processing; S2: The standardized original data is split into two shares using a secret sharing protocol and stored in different distributed nodes off-chain. The metadata of the original power-sensitive data is extracted, re-encrypted, and then stored on-chain. S3: Based on the improved BFT consensus algorithm, perform consistency verification on distributed nodes; S4: Verify the identity information and computing needs submitted by data users through the blockchain. If they are valid, obtain the corresponding share of secret data from the off-chain storage node. S5: Based on computational requirements, the improved Paillier homomorphic encryption algorithm is used to perform linear operations, gradient aggregation, and model training in the ciphertext domain. Parallel computation is achieved by combining the MapReduce framework to obtain the computational requirements of data users.

[0027] The hybrid storage system solves the problem of limited storage capacity in blockchain. On-chain storage of metadata ensures traceability, while off-chain storage of encrypted data improves storage efficiency. The improved BFT consensus algorithm ensures the consistency of distributed nodes and solves the trust problem in cross-organizational data sharing. The collaborative computing protocol of secret sharing and homomorphic encryption balances the strength of privacy protection with computing and communication efficiency. Compared with traditional single privacy computing solutions, it improves computing efficiency and reduces communication overhead. The proxy re-encryption technology enables controllable data sharing and meets compliance requirements.

[0028] The detailed method is as follows, such as Figure 2 As shown: Acquiring and processing raw power-sensitive data specifically includes: The raw power-sensitive data is first standardized, and then the standardized numerical features are mapped to the [0,1] interval. The standardized raw data is split into two parts using a secret sharing protocol and stored separately on different off-chain distributed nodes. The metadata of the original power-sensitive data is extracted, re-encrypted, and then stored on-chain. The specific steps are as follows: After collecting sensitive data such as power load data and equipment operating parameters, the data provider first performs standardization processing, mapping numerical features to the [0,1] interval to adapt to subsequent calculation requirements. Then, a 2-out-of-2 secret sharing protocol is used to split the standardized raw data, starting with secure prime numbers (e.g., p=2¹) for scalar data. 048 +3) Within the corresponding interval, one share is randomly selected, and the other share is obtained by modulo operation between the original data and the random share. For matrix-type power data, an element-level splitting method is adopted. The two data shares are stored in different off-chain distributed storage nodes. A single node only holds a portion of the shares and cannot reverse-engineer the original data. At the same time, metadata such as data index and access permission rules are extracted, encrypted by a proxy re-encryption algorithm, and uploaded to the consortium chain node built on Hyperledger Fabric. This forms a collaborative mode of on-chain index and off-chain storage, which not only avoids the capacity limitation of blockchain storage of large-scale original data, but also ensures the traceability of data flow through the on-chain immutability feature.

[0029] Details are as follows: First, implement the 2-out-of-2 secret sharing protocol for the raw power data. The data owner generates two secret shares. and ,satisfy:

[0030] in For the sake of safe prime numbers, from Random selection Where mod p is the remainder when divided by the safe prime number p; for matrix-type power data The share matrix is ​​obtained by element-level partitioning. and ,satisfy Then, the Paillier homomorphic encryption algorithm is improved by randomly selecting two large prime numbers p and q during the key generation stage, satisfying... Calculate the modulus and private key parameters Randomly select generator Define public key and private key ,in , The encryption process for plaintext Select random numbers The ciphertext calculation process is as follows:

[0031] It satisfies additive homomorphism and scalar multiplication homomorphism; for ciphertext c, the plaintext calculation process is as follows:

[0032] Based on the improved BFT consensus algorithm, the consistency verification of distributed nodes is specifically performed as follows: Improve the Byzantine Fault Tolerant (BFT) consensus algorithm to make its fault tolerance meet the requirements. Where n is the number of consensus nodes, and the consensus process is as follows: After receiving the computation request, the master node generates a proposal. Where h is the block height and v is the view number; after verifying the validity of the proposal, the slave node sends a confirmation message. After receiving the 2f confirmation message, the master node broadcasts a commit message. After receiving 2f+1 commit messages, the node confirms the block and updates its local ledger, thus completing consensus.

[0033] Based on computational requirements, the improved Paillier homomorphic encryption algorithm is used to perform linear operations, gradient aggregation, and model training in the ciphertext domain. Parallel computation is then achieved using the MapReduce framework. The specific steps to obtain the computational requirements of the data user are as follows: The corresponding secret shares are downloaded from the off-chain storage node. The participating nodes holding the two secret shares generate improved Paillier homomorphic encryption key pairs and exchange public keys. They encrypt their own vector shares using the other party's public key and send them to each other. Then, each node generates a random vector, calculates the product of its local matrix share and the other party's vector share, subtracts the random vector, and sends the encrypted result to each other a second time. The node decrypts the received encrypted intermediate result, calculates its local share by combining the product of its own matrix and vector share with its local random vector, and then superimposes the two local shares and takes the modulo of a safe prime number to obtain the product of the target matrix and vector. In the training of the power load forecasting model, based on the mini-batch gradient descent algorithm and combined with the MapReduce framework, the large-scale computing task is split into 8 parallel partitions, which are assigned to 2 Slave nodes for execution, and the Master node is responsible for task scheduling and result aggregation.

[0034] Details: The implementation of a secret-sharing, homomorphic encrypted collaborative computation protocol in matrix-vector multiplication involves two nodes each holding fragments of the target matrix and vector. The matrix and vector are formed by superimposing these corresponding fragments. The two nodes first exchange encryption keys and send each other encrypted vector fragments. Then, they generate random vectors, subtract the random vector from the product of their respective matrix fragments and the other node's vector fragment, and then encrypt and send these fragments back to each other. Next, they decrypt intermediate results, combine their own fragment products with the random vector to calculate the local result, and finally, they superimpose the local results from both nodes and take the remainder according to a secure value to obtain the final product. Afterwards, gradient descent and model updates are performed, based on a mini-batch gradient descent algorithm, for mini-batch samples. and the split model parameters Gradient ciphertext aggregation is achieved through a collaborative computing protocol, and the gradient calculation satisfies:

[0035]

[0036]

[0037] Nodes protect gradient shares through homomorphic encryption, according to the formula:

[0038] Update local parameters, where Let t be the learning rate and t be the number of iterations.

[0039] Power data is shared in a controlled manner by using proxy re-encryption technology. Only authorized users can obtain the right to use the data by decrypting it with a private key. The entire data flow log is stored on the blockchain to ensure traceability. The data provider first verifies the identity and access rights of authorized users through blockchain consensus nodes. After confirming compliance with preset rules, the provider generates a unique re-encryption key using its own private key and the authorized user's public key. This key is then sent to the authorized user via a trusted blockchain channel, ensuring that neither the data provider's nor the authorized user's private key is disclosed throughout the process. Once the authorized user obtains the re-encryption key, they can convert the original encrypted ciphertext into a new ciphertext adapted to their own public key without the data provider's involvement. The conversion process does not involve the decryption or exposure of the original plaintext data. Only authorized users can decrypt the new ciphertext using their own private key to obtain legitimate data usage rights. Unauthorized users, even if they obtain the original or converted ciphertext, cannot decrypt it to obtain the plaintext due to the lack of the corresponding private key and re-encryption key. Key operation logs throughout the authorization and ciphertext conversion process are stored on the blockchain in real time, ensuring that changes in data sharing permissions and the entire ciphertext flow are fully traceable.

[0040] When a data user needs to authorize a third party to participate in computation, the data provider first verifies the legitimacy of the third party's identity and the scope of their access permissions through the blockchain consensus node. After confirming that it meets the preset rules, the third party generates a unique re-encryption key using its own private key and the third party's public key. This key is sent through a trusted blockchain channel without revealing the private keys of either party. After obtaining the re-encryption key, the third party can convert the original encrypted ciphertext into a new ciphertext adapted to its own public key without the data provider's participation. The conversion process does not involve exposing the original plaintext. Only the third party can decrypt the new ciphertext using its own private key to obtain the right to use the data. Unauthorized users cannot decrypt the ciphertext even if they obtain it.

[0041] like Figure 3 As shown, this embodiment of the invention provides a power data security sharing system based on privacy computing, comprising: Data acquisition module: used to acquire raw power-sensitive data and perform data processing; The split storage module is used to split the standardized original data into two shares using a secret sharing protocol, and store them separately on different distributed nodes off-chain. It also extracts the metadata of the original power-sensitive data, re-encrypts it, and stores it on-chain. Verification module: Used to verify the consistency of distributed nodes based on the improved BFT consensus algorithm; Verification module: Used to verify the identity information and computing needs submitted by data users through the blockchain. If they are valid, the corresponding secret data share is obtained from the off-chain storage node. The computation module is used to perform linear operations, gradient aggregation, and model training in the ciphertext domain based on computational needs using an improved Paillier homomorphic encryption algorithm. It combines the MapReduce framework to achieve parallel computation and obtain the computational needs of data users.

[0042] A computer device is provided according to an embodiment of the present invention. This computer device includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in the various method embodiments described above. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the various device embodiments described above.

[0043] The computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention.

[0044] The computer device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device may include, but is not limited to, a processor and memory.

[0045] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0046] The memory can be used to store the computer program and / or module, and the processor implements various functions of the computer device by running or executing the computer program and / or module stored in the memory, and by calling the data stored in the memory.

[0047] If the modules / units integrated into the computer device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory, random access memory, electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0048] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the scope of the invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0049] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can be appropriately combined to form other embodiments that can be understood by those skilled in the art. The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A method for secure sharing of power data based on privacy computing, characterized in that, include: Acquire raw power-sensitive data and perform data processing; The standardized raw data is split into two parts using a secret sharing protocol and stored on different distributed nodes off-chain. The metadata of the original power-sensitive data is extracted, re-encrypted, and then stored on-chain. Based on the improved BFT consensus algorithm, consistency verification is performed on distributed nodes; The identity information and computing needs submitted by data users through the blockchain are verified. If they are legitimate, the corresponding share of secret data is obtained from the off-chain storage node. Based on computational requirements, an improved Paillier homomorphic encryption algorithm is used to perform linear operations, gradient aggregation, and model training in the ciphertext domain. Parallel computation is achieved by combining the MapReduce framework to obtain the computational requirements of data users.

2. The method for secure sharing of power data based on privacy computing according to claim 1, characterized in that, The standardized raw data is split into two parts using a secret sharing protocol and stored separately on different off-chain distributed nodes. The metadata of the original power-sensitive data is extracted, re-encrypted, and then stored on-chain. The specific steps are as follows: For the standardized raw data, one share is randomly selected from the interval corresponding to the safe prime number. The other share is obtained by modulo operation between the raw data and the random share. For matrix-type power data, an element-level splitting method is adopted. The two data shares are stored in different off-chain distributed storage nodes. A single node only holds a part of the share and cannot reverse the original data. At the same time, metadata is extracted, encrypted by a proxy re-encryption algorithm, and uploaded to the consortium chain node built on Hyperledger Fabric.

3. The method for secure sharing of power data based on privacy computing according to claim 1, characterized in that, Based on the improved BFT consensus algorithm, the consistency verification of distributed nodes is specifically performed as follows: The fault-tolerance formula of the improved BFT algorithm is as follows: Tolerates one faulty node; The verification process is as follows: The master node rotates according to preset rules. After receiving a data sharing request, it generates a proposal containing block height, view number, proposal initiator, and encrypted metadata information. After the slave node verifies the legality of the proposal, it sends a confirmation message. After receiving 2f confirmation messages, the master node broadcasts a commit message. After each node receives 2f+1 commit messages, it confirms the block and updates its local ledger, thus completing the consensus process. When a master node failure causes a slave node to not receive a proposal within a preset timeout period, the slave node automatically initiates a view switching request. After receiving 2f identical requests, it switches to the new view.

4. The method for secure sharing of power data based on privacy computing according to claim 1, characterized in that, Based on computational requirements, the improved Paillier homomorphic encryption algorithm is used to perform linear operations, gradient aggregation, and model training in the ciphertext domain. Parallel computation is then achieved using the MapReduce framework. The specific steps to obtain the computational requirements of the data user are as follows: The corresponding secret shares are downloaded from the off-chain storage node. The participating nodes holding the two secret shares generate improved Paillier homomorphic encryption key pairs and exchange public keys. They encrypt their own vector shares using the other party's public key and send them to each other. Then, each node generates a random vector, calculates the product of its local matrix share and the other party's vector share, subtracts the random vector, and sends the encrypted result to each other a second time. The node decrypts the received encrypted intermediate result, calculates its local share by combining the product of its own matrix and vector share with its local random vector, and then superimposes the two local shares and takes the modulo of a safe prime number to obtain the product of the target matrix and vector. In the training of the power load forecasting model, based on the mini-batch gradient descent algorithm and combined with the MapReduce framework, the large-scale computing task is split into 8 parallel partitions, which are assigned to 2 Slave nodes for execution, and the Master node is responsible for task scheduling and result aggregation.

5. A method for secure sharing of power data based on privacy computing according to claim 1, characterized in that, The specific steps for acquiring and processing raw power-sensitive data are as follows: The original power-sensitive data is first standardized, and then the standardized numerical features are mapped to the [0,1] interval.

6. A method for secure sharing of power data based on privacy computing according to claim 1, characterized in that, Also includes: When a data user needs to authorize a third party to participate in the calculation, the data provider first verifies the legitimacy of the third party's identity and the scope of access permissions through the blockchain consensus node. After confirming that it meets the preset rules, the provider generates a unique re-encryption key using its own private key and the third party's public key. This key is sent through the blockchain's trusted channel without disclosing the private keys of either party throughout the process. Once a third party obtains the re-encryption key, it can convert the original encrypted ciphertext into a new ciphertext adapted to its own public key without the data provider's involvement. The conversion process does not involve exposing the original plaintext. Only the third party can decrypt the new ciphertext using its own private key to obtain the right to use the data. Unauthorized users cannot crack the ciphertext even if they obtain it.

7. A power data secure sharing system based on privacy computing, characterized in that, include: Data acquisition module: used to acquire raw power-sensitive data and perform data processing; The split storage module is used to split the standardized original data into two shares using a secret sharing protocol, and store them separately on different distributed nodes off-chain. It also extracts the metadata of the original power-sensitive data, re-encrypts it, and stores it on-chain. Verification module: Used to verify the consistency of distributed nodes based on the improved BFT consensus algorithm; Verification module: Used to verify the identity information and computing needs submitted by data users through the blockchain. If they are valid, the corresponding secret data share is obtained from the off-chain storage node. The computation module is used to perform linear operations, gradient aggregation, and model training in the ciphertext domain based on computational needs using an improved Paillier homomorphic encryption algorithm. It combines the MapReduce framework to achieve parallel computation and obtain the computational needs of data users.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method for secure sharing of power data based on privacy computing as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of a method for secure sharing of power data based on privacy computing as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of a method for secure sharing of power data based on privacy computing as described in any one of claims 1-6.

Images