Power grid demand response user screening method and system based on data security sharing

Abstract

The invention discloses a power grid demand response user screening method and system based on data security sharing, and relates to the technical field of user screening and guiding, and the method comprises the steps: a charging station carries out the clustering of charging station data samples based on a K-means clustering algorithm; the charging station encrypts a clustering result by using a Paillier homomorphic encryption algorithm and uploads the clustering result to the power grid cloud platform; the power grid cloud platform calculates the encrypted clustering result uploaded by the charging station; the charging station decrypts the calculation result ciphertext and calculates a new common clustering center; the charging station continues iterative optimization according to the new clustering center until the clustering center does not change any more, and a clustering result is output; and the power grid cloud platform screens electric vehicle demand response users according to the charging modes of the users, and formulates a demand response guide strategy. And the user charging mode is extracted by using the user charging information of the plurality of charging stations in the area, so that the method has important engineering significance for fully excavating the potential of the electric vehicle participating in network load interaction.

Description

Technical Field

[0001] This invention relates to the field of user screening and guidance technology, and in particular to a user screening method and system for power grid demand response based on secure data sharing. Background Technology

[0002] Demand response is a crucial method for power grid load regulation, and selecting demand response users is essential for supporting the grid's demand response implementation. Electric vehicles, as a widely discussed flexible load regulation resource, are a key participant in demand response. Therefore, how to extract the charging patterns of electric vehicle users through methods such as charging data clustering to support the grid in selecting demand response users has become a pressing issue.

[0003] Within a local area, a single electric vehicle user may charge at multiple charging stations. Therefore, user charging data is scattered across these stations. Extracting user charging patterns requires relatively complete charging data. Clustering user charging information from a single station to obtain charging patterns may lead to incomplete evaluation results. Furthermore, collecting charging information from multiple stations for evaluation could result in user privacy leaks. Therefore, how to extract user charging patterns from multiple charging stations within a region without compromising user privacy, using privacy-preserving computational methods and clustering algorithms, to help the power grid cloud platform screen electric vehicle demand response users and formulate demand response guidance strategies, has become a pressing issue.

[0004] Based on this, the present invention proposes a power grid demand response user screening method based on secure data sharing. Under the premise of ensuring user privacy and security, it makes full use of user charging information from multiple charging stations in the region and extracts user charging patterns through methods such as clustering, thereby supporting the power grid to complete demand response user screening. Summary of the Invention

[0005] In view of the above-mentioned problems, the present invention is proposed.

[0006] Therefore, the problem to be solved by this invention is: how to extract user charging patterns by making full use of user charging information from multiple charging stations in a region without disclosing user privacy.

[0007] To address the aforementioned technical problems, this invention provides the following technical solution: a power grid demand response user screening method based on secure data sharing, comprising: charging stations clustering charging station data samples using the K-means clustering algorithm; the charging stations encrypting the clustering results using the Paillier homomorphic encryption algorithm and uploading them to the power grid cloud platform; the power grid cloud platform calculating the encrypted clustering results uploaded by the charging stations; the charging stations decrypting the encrypted calculation results and calculating new common cluster centers; the charging stations iteratively optimizing based on the new cluster centers until the cluster centers no longer change, and outputting the clustering results; and the power grid cloud platform screening electric vehicle demand response users based on their charging modes and formulating demand response guidance strategies.

[0008] As a preferred embodiment of the power grid demand response user screening method based on data security sharing described in this invention, wherein: the clustering of charging station data samples includes a region having a number of charging stations of l z If user s charges at charging station k a times, then user s has a charging time data samples (X) at charging station k. k,1 ,X k,2 ,…,X k,a Each data sample represents the start and end times of each charge cycle (t). c ,t o The time granularity is 1440 points per day, or 1 point per minute. The power grid cloud platform is set to have m clusters and the initial cluster centers are set to (U1, U2, ..., U...). m The process involves distributing m cluster centers to each charging station. Each charging station calculates the Manhattan distance from each of its sample data points to a given cluster center. If a sample has the closest Manhattan distance to a given cluster center, it is assigned to that cluster. The sample data from each charging station is then clustered to obtain new cluster centers for each station and the number of samples in each cluster. The new cluster centers and the number of samples in each cluster for charging station k are denoted as (Uk). k，1 U k,2 ,…,U k,m ) and (N k，1 N k,2 ,…,N k,m Let the sum of the charging time data in each cluster be (P). k，1 ,P k,2 ,…,P k,m This includes data on the start and end times of charging.

[0009] As a preferred embodiment of the power grid demand response user screening method based on secure data sharing described in this invention, the encryption of the clustering results includes encrypting the number of samples and the sum of charging time data in each cluster of user s' charging time data samples using the Paillier homomorphic encryption algorithm, expressed as:

[0010] V k,j =E(U k,j )

[0011] M k,j =E(N) k,j )

[0012] R k,j =E(P) k,j )

[0013] E(·)=g · ×r n modn 2

[0014] Where E(·) is Paillier homomorphic encryption, V k,j For U k,j The encryption result, M k,j For N k,j The encryption result, R k,j For P k,j The encryption result is given by g, which is part of the public key, n, which is another part of the public key, r, which is a randomly selected integer, and mod, which is the modulo operation. The charging station encrypts the mobile phone number of user s, which serves as the basis for the power grid cloud platform to identify the ciphertext of user s from each charging station. The charging station then uploads the encrypted data to the power grid cloud platform.

[0015] As a preferred embodiment of the power grid demand response user screening method based on secure data sharing described in this invention, the calculation of the encrypted clustering results uploaded by charging stations includes: the power grid cloud platform associates the encrypted data of user s's mobile phone number with the encrypted data of user s's charging time uploaded by each charging station, and performs homomorphic computation on the encrypted data, expressed as:

[0016]

[0017]

[0018] The power grid cloud platform will calculate the encrypted result M j R j Distribute to each charging station.

[0019] As a preferred embodiment of the power grid demand response user screening method based on data security sharing described in this invention, the decryption of the ciphertext of the calculation result includes each charging station decrypting the ciphertext of the calculation result sent by the power grid cloud platform, represented as:

[0020]

[0021]

[0022]

[0023] Where D(·) represents the decryption calculation, and N j Let P be the number of samples in the j-th common cluster. j Let L be the sum of charging times in the j-th common cluster, summing the start and end charging times respectively, where L is the decryption function and λ is a part of the private key; the calculation of the new common cluster center includes each charging station calculating the new common cluster center based on the decrypted data, expressed as:

[0024]

[0025] Among them, U j Let j be the j-th common cluster center.

[0026] As a preferred embodiment of the power grid demand response user screening method based on data security sharing described in this invention, the iterative optimization includes: after each charging station obtains a new cluster center, the entire process is repeated, with each iteration reclassifying samples and calculating common cluster centers. When the cluster centers no longer change, the iteration ends, and the cluster centers (U1, U2, ..., U...) are output. m The m charging modes for user s are given, and the number of samples (N1, N2, ..., N) corresponding to each cluster is output. m The sum of the charging time data (P1, P2, ..., P) and the charging time data (P1, P2, ..., P) m Based on the number of charges under each charging mode, the m charging modes (U1, U2, ..., U...) of user s are calculated. m The probability of occurrence is calculated and expressed as:

[0027]

[0028] Each charging station uploads the final charging mode clustering results of users to the power grid cloud platform.

[0029] As a preferred embodiment of the power grid demand response user screening method based on data security sharing described in this invention, the step of formulating the demand response guidance strategy includes defining a load gap ΔS in the region during time period t on a certain day. tTo guide electric vehicle users to participate in demand response, the overlap between the charging mode j of electric vehicle user s and time period t is defined as the proportion of the time period in which charging mode j overlaps with time period t to the total charging time of charging mode j, expressed as:

[0030]

[0031] Where, λ j Δτ represents the overlap between charging mode j and time period t. j τ is the duration of the period when charging mode j overlaps with time period t. j Define the charging duration for charging mode j; define the response potential ρ of user s charging mode j in the demand response process. j Let the product of the overlap between charging mode j and time period t and the probability of charging mode j occurring be expressed as:

[0032] ρ j =ω j λ j

[0033] The charging mode with the highest user response potential is defined as the unrecommended charging mode for demand response. The response potential of user s in this demand response process is defined as the maximum value of the response potential of each charging mode, expressed as:

[0034] ρ0=max(ρ j )

[0035] The response potential of all electric vehicle users in the region is ranked. Based on the load gap size and the ranking results of user demand response potential, a list of demand response users is selected. For some electric vehicle users on the demand response user list, users are guided not to charge during the load gap period through electricity price compensation. The charging information of users on the same day is collected through charging stations to verify whether users respond in accordance with the grid requirements. Compensation is given to users who actually participate in the response.

[0036] Another objective of this invention is to provide a system for selecting users of power grid demand response based on secure data sharing, which solves the problem of selecting users of power grid demand response based on secure data sharing by constructing a power grid demand response user selection system.

[0037] To address the aforementioned technical problems, this invention provides the following technical solution: a power grid demand response user screening system based on secure data sharing, comprising a data acquisition module, a data processing and clustering module, an encryption and data security module, a cloud platform computing module, a decryption and optimization module, and a demand response strategy formulation module. The data acquisition module collects data from charging stations as the basis for analysis and processing. The data processing and clustering module uses the K-means clustering algorithm to cluster data samples from charging stations, identifying different charging modes. The encryption and data security module uses the Paillier homomorphic encryption algorithm to encrypt the clustering results. The cloud platform computing module performs calculations on the encrypted clustering results on the power grid cloud platform. The decryption and optimization module decrypts the encrypted calculation results returned by the cloud platform, calculates new common cluster centers based on the decrypted data, and continuously iterates and optimizes the clustering process. The demand response strategy formulation module, based on the user's charging mode, allows the power grid cloud platform to screen suitable electric vehicle demand response users and formulate corresponding demand response guidance strategies.

[0038] A computer device includes a memory and a processor, the memory storing a computer program, characterized in that the processor executes the computer program to implement the steps of the power grid demand response user screening method based on secure data sharing as described above.

[0039] A computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of the power grid demand response user screening method based on secure data sharing as described above.

[0040] The beneficial effects of this invention are as follows: The power grid demand response user screening method based on secure data sharing provided by this invention utilizes homomorphic encryption algorithm and k-means clustering algorithm to extract user charging patterns by making full use of user charging information from multiple charging stations in the region without leaking user privacy. This provides a reference for the power grid cloud platform to guide users to participate in demand response through electricity price compensation strategy, and has important engineering significance for fully tapping the potential of electric vehicles to participate in grid-load interaction. Attached Figure Description

[0041] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 The flowchart illustrates the overall process of the power grid demand response user selection method based on secure data sharing, as provided in the first embodiment of the present invention.

[0043] Figure 2 This is a structural diagram of the power grid demand response user screening method based on secure data sharing provided in the first embodiment of the present invention.

[0044] Figure 3 This is a structural diagram of a power grid demand response user screening system based on secure data sharing, provided in the second embodiment of the present invention. Detailed Implementation

[0045] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0046] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0047] Example 1

[0048] Reference Figure 1 and Figure 2 This is the first embodiment of the present invention, which provides a power grid demand response user screening method based on secure data sharing, including: charging stations clustering charging station data samples using the K-means clustering algorithm; charging stations encrypting the clustering results using the Paillier homomorphic encryption algorithm and uploading them to the power grid cloud platform; the power grid cloud platform calculating the encrypted clustering results uploaded by the charging stations; the charging stations decrypting the encrypted calculation results and calculating new common cluster centers; the charging stations iteratively optimizing based on the new cluster centers until the cluster centers no longer change, and outputting the clustering results; and the power grid cloud platform screening electric vehicle demand response users based on their charging modes and formulating demand response guidance strategies.

[0049] The data-secure sharing-based power grid demand response user screening technology architecture primarily involves operators of various charging stations and the power grid cloud platform. The power grid cloud platform mainly comprises power grid entities tasked with guiding electric vehicle users to participate in demand response, such as power trading centers and the marketing departments of power supply bureaus.

[0050] Charging station operators possess a large amount of user charging data and have the right to decide whether to share data resources, allowing sharing only under the premise of data security. Their main responsibilities include: ① Receiving homomorphic encryption public keys; ② Clustering charging data samples based on initial cluster centers; ③ Encrypting the clustering results and sending the encrypted ciphertext to the privacy computing service platform; ④ Receiving and decrypting the homomorphic computation results returned by the privacy computing service platform, and clustering charging data samples based on the new cluster centers; ⑤ Uploading the final user charging mode clustering results to the power grid cloud platform.

[0051] The power grid cloud platform is primarily responsible for: ① receiving encrypted data from each charging station; ② broadcasting encrypted parameters; ③ performing homomorphic computation; ④ acting as the central node for each charging station, providing relay for communication between charging stations; and ⑤ developing electricity price supplementation strategies based on the clustering results of electric vehicle users' charging patterns to guide users to participate in demand response.

[0052] like Figure 2 As shown, the power grid cloud platform acts as a central node. Each charging station uploads the clustering results of a user's start and end charging times in encrypted form to the power grid cloud platform. The power grid cloud platform performs homomorphic computation on the common cluster centers and returns the results to each charging station. Each charging station decrypts the data to obtain the new common cluster centers. This process is iterated until the common cluster centers remain unchanged, thus obtaining the user's charging pattern. The charging stations need to upload the final user charging pattern clustering results to the power grid cloud platform. Based on the user's charging pattern, combined with the demand response period and the user's response potential, the power grid cloud platform filters electric vehicle demand response users and formulates demand response guidance strategies.

[0053] Step 1: Each charging station clusters its data samples using the K-means clustering algorithm.

[0054] Let the number of charging stations in this area be l. z If user s charges at charging station k a times, then user s has a charging time data samples (X) at charging station k. k,1 ,X k,2 ,…,X k,a Each data sample represents the start and end times of each charge cycle (t). c ,t o The time granularity is 1440 points per day, or 1 point per minute. The power grid cloud platform sets the number of clusters to m, and the initial cluster centers to (U1, U2, ..., U...). mThe process involves distributing m cluster centers to each charging station. Each charging station calculates the Manhattan distance from each of its sample data points to a given cluster center. If a sample has the closest Manhattan distance to a given cluster center, it is assigned to that cluster. This method allows for the clustering of sample data from each charging station, resulting in new cluster centers for each station and the number of samples in each cluster. The new cluster centers and the number of samples in each cluster for charging station k are denoted as (Uk, Uk, Uk, Uk). k，1 U k,2 ,…,U k,m ) and (N k，1 N k,2 ,…,N k,m Let the sum of the charging time data in each cluster be (P). k，1 ,P k,2 ,…,P k,m This includes data on the start and end times of charging.

[0055] Step 2: Each charging station uses the Paillier homomorphic encryption algorithm to encrypt the clustering results and upload them to the power grid cloud platform.

[0056] Each charging station uses the Paillier homomorphic encryption algorithm to encrypt the number of samples in each cluster and the sum of charging time data for user s, as shown below:

[0057] V k,j =E(U k,j (1)

[0058] M k,j =E(N) k,j (2)

[0059] R k,j =E(P) k,j (3)

[0060] E(·)=g · ×r n modn 2 (4)

[0061] Where E(·) is Paillier homomorphic encryption, V k,j For U k,j The encryption result, M k,j For N k,j The encryption result, R k,j For P k,j The encryption result is given by g, which is part of the public key, n, r, a randomly selected integer, and mod, which is the modulo operation.

[0062] At the same time, each charging station encrypts the user's mobile phone number as the basis for the power grid cloud platform to consistently identify the encrypted data from each charging station. Each charging station then uploads the encrypted data to the power grid cloud platform.

[0063] Step 3: The power grid cloud platform calculates the encrypted clustering results uploaded by each charging station.

[0064] The power grid cloud platform uses the encrypted mobile phone number of user s to associate with the encrypted clustering results of user s's charging time data uploaded by each charging station, and performs homomorphic computation on the encrypted data, which is represented as:

[0065]

[0066]

[0067] The power grid cloud platform will calculate the above encrypted result M j R j Distribute to each charging station.

[0068] Step 4: Each charging station decrypts the encrypted calculation results and calculates new common cluster centers.

[0069] Each charging station decrypts the encrypted calculation results sent by the power grid cloud platform, which are represented as follows:

[0070]

[0071]

[0072]

[0073] Where D(·) represents the decryption calculation, and N j Let P be the number of samples in the j-th common cluster. j Let L be the sum of the charging times in the j-th common cluster, and let L be the sum of the start charging time and the end charging time. Let L be the decryption function, and λ be a part of the private key, which is related to n in the public key.

[0074] Each charging station calculates a new common cluster center based on the decrypted data, represented as:

[0075]

[0076] Among them, U j Let j be the j-th common cluster center.

[0077] Step 5: Each charging station continues to iterate and optimize based on the new cluster centers until the cluster centers no longer change, and then outputs the clustering results.

[0078] After each charging station obtains its new cluster centers, the above steps are repeated to cluster the samples again. Each iteration reclassifies the samples and calculates the common cluster centers. The iteration ends when the cluster centers no longer change, and the cluster centers (U1, U2, ..., U...) are output. m The m charging modes for user s are given, and the number of samples (N1, N2, ..., N) corresponding to each cluster is output. m The sum of the charging time data (P1, P2, ..., P) and the charging time data (P1, P2, ..., P) m ).

[0079] Based on the number of charges under each charging mode, the m charging modes (U1, U2, ..., U...) of user s are analyzed. m The probability of occurrence is calculated and expressed as:

[0080]

[0081] Each charging station uploads the final charging mode clustering results of users to the power grid cloud platform.

[0082] Step Six: The power grid cloud platform filters electric vehicle demand response users based on their charging modes and formulates demand response guidance strategies.

[0083] Suppose that there is still a load gap ΔS in the region during time period t on a certain day. t This requires guiding electric vehicle users to participate in demand response. The overlap between electric vehicle user s's charging mode j and time period t is defined as the proportion of the time period during which charging mode j overlaps with time period t to the total charging time of charging mode j, expressed as:

[0084]

[0085] Where, λ j Δτ represents the overlap between charging mode j and time period t. j τ is the duration of the period when charging mode j overlaps with time period t. j The charging duration for charging mode j.

[0086] Define the response potential ρ of user s charging mode j in this demand response process. j Let the product of the overlap between charging mode j and time period t and the probability of charging mode j occurring be expressed as:

[0087] ρ j =ω j λ j (13)

[0088] The charging mode with the highest user response potential is defined as the unrecommended charging mode for this demand response. Based on this, the response potential of user s in this demand response process is defined as the maximum value of the response potential of each charging mode, expressed as:

[0089] ρ0=max(ρ j (14)

[0090] The response potential of all electric vehicle users in the region is ranked. Based on the load gap size and the ranking results of user demand response potential, a list of demand response users is selected. For these electric vehicle users, users are guided not to charge during periods of load gap through electricity price compensation. The charging information of users on the same day is collected through charging stations to verify whether users respond in accordance with the grid requirements. Electricity price compensation is given to users who actually participate in the response through charging stations. The grid cloud platform gives corresponding compensation to charging stations based on the user electricity price compensation.

[0091] Example 2

[0092] Reference Figure 3 This is the second embodiment of the present invention, which differs from the previous embodiment in that it provides a power grid demand response user screening system based on secure data sharing, including: a data acquisition module, a data processing and clustering module, an encryption and data security module, a cloud platform computing module, a decryption and optimization module, and a demand response strategy formulation module.

[0093] The data acquisition module collects data from the charging station as the basis for analysis and processing.

[0094] The data processing and clustering module uses the K-means clustering algorithm to cluster data samples from charging stations and identify different charging modes.

[0095] The encryption and data security module uses the Paillier homomorphic encryption algorithm to encrypt the clustering results.

[0096] The cloud platform computing module performs calculations on the encrypted clustering results on the power grid cloud platform.

[0097] The decryption and optimization module decrypts the encrypted calculation results returned by the cloud platform, calculates new common cluster centers based on the decrypted data, and continuously iterates and optimizes the clustering process.

[0098] The demand response strategy formulation module is based on the user's charging mode. The power grid cloud platform selects suitable electric vehicle demand response users and formulates corresponding demand response guidance strategies.

[0099] If the aforementioned functions 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, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0100] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0101] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0102] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0103] Example 3

[0104] The third embodiment of the present invention differs from the first two embodiments in that it is used to verify and explain the technical effects adopted in the present invention, so as to verify the real effect of the method.

[0105] This embodiment uses both the conventional method and our invented method for simultaneous detection, and the comparison results are shown in the table below:

[0106] Table 1 Comparison between traditional methods and our inventive methods

[0107]

[0108] The above comparison results show that the charging mode extraction error of our invention is 5%, which is 63% less than the 68% of the traditional method. The demand response error is 5%, which is 49% less than the 54% of the traditional method. The technical security is much stronger than that of the traditional method. Our invention has the technical feature of multi-charging station data clustering based on homomorphic encryption technology, while the traditional method is single-charging station data clustering.

[0109] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A power grid demand response user screening method based on secure data sharing, characterized in that: include, The charging station clusters its data samples using the K-means clustering algorithm. The charging station uses the Paillier homomorphic encryption algorithm to encrypt the clustering results and upload them to the power grid cloud platform; The power grid cloud platform calculates the encrypted clustering results uploaded by charging stations; The charging station decrypts the encrypted calculation results and calculates new common cluster centers; The charging station continues to iterate and optimize based on the new cluster centers until the cluster centers no longer change, and then outputs the clustering results. The power grid cloud platform filters electric vehicle demand response users based on their charging patterns and formulates demand response guidance strategies.

2. The power grid demand response user screening method based on secure data sharing as described in claim 1, characterized in that: The clustering of charging station data samples includes a region having a number of charging stations of l. z If user s charges at charging station k a times, then user s has a charging time data samples (X) at charging station k. k,1 ,X k,2 ,…,X k,a Each data sample represents the start and end times of each charge cycle (t). c ,t o The time granularity is 1440 points per day, or 1 point per minute. The power grid cloud platform is set to have m clusters and the initial cluster centers are set to (U1, U2, ..., U...). m The m cluster centers are distributed to each charging station. The charging station calculates the Manhattan distance from each of its sample data points to the initial cluster center. If a sample has the closest Manhattan distance to a cluster center, it is assigned to that cluster. The sample data from each charging station is then clustered separately, resulting in new cluster centers for each station and the number of samples in each cluster. The new cluster centers and the number of samples in each cluster for charging station k are denoted as (Uk). k，1 U k,2 ,…,U k,m ) and (N k，1 N k,2 ,…,N k,m Let the sum of the charging time data in each cluster be (P). k，1 ,P k,2 ,…,P k,m This includes data on the start and end times of charging.

3. The power grid demand response user screening method based on secure data sharing as described in claim 2, characterized in that: The encryption of the clustering results includes using the Paillier homomorphic encryption algorithm to encrypt the number of samples in each cluster and the sum of the charging time data for user s, as shown below. V k,j =E(U k,j ) M k,j =E(N k,j ) R k,j =E(P k,j ) E(·)=g · ×r n modern 2 Where E(·) is Paillier homomorphic encryption, V k,j For U k,j The encryption result, M k,j For N k,j The encryption result, R k,j For P k,j The encryption result is given by g, which is part of the public key, n, which is another part of the public key, r, which is a randomly selected integer, and mod, which is the modulo operation. The charging station encrypts the user's mobile phone number as the basis for the power grid cloud platform to identify the encrypted data from each charging station. The charging station then uploads the encrypted data to the power grid cloud platform.

4. The power grid demand response user screening method based on secure data sharing as described in claim 3, characterized in that: The calculation of the encrypted clustering results uploaded by the charging stations includes the following steps: the power grid cloud platform associates the encrypted data of user s's mobile phone number with the encrypted data of user s's charging time uploaded by each charging station, performs homomorphic computation on the encrypted data, and represents it as follows: The power grid cloud platform will calculate the encrypted result M j R j Distribute to each charging station.

5. The power grid demand response user screening method based on secure data sharing as described in claim 4, characterized in that: The decryption of the ciphertext of the calculation results includes each charging station decrypting the ciphertext of the calculation results sent by the power grid cloud platform, represented as follows: N j =D(M j )=N 1,j +N 2,j +…+N lz,j P j =D(R j )=P 1,j +P 2,j +…+P lz,j Where D(·) represents the decryption calculation, and N j Let P be the number of samples in the j-th common cluster. j Let L be the sum of the charging times in the j-th common cluster, and let L be the decryption function, where λ is a part of the private key. The calculation of the new common cluster center includes each charging station calculating a new common cluster center based on the decrypted data, denoted as follows: Among them, U j Let j be the j-th common cluster center.

6. The power grid demand response user screening method based on secure data sharing as described in claim 5, characterized in that: The iterative optimization includes repeating the entire process after each charging station obtains a new cluster center. Each iteration reclassifies the samples and calculates the common cluster center. The iteration ends when the cluster centers no longer change, and the cluster centers (U1, U2, ..., U...) are output. m The m charging modes for user s are given, and the number of samples (N1, N2, ..., N) corresponding to each cluster is output. m The sum of the charging time data (P1, P2, ..., P) and the charging time data (P1, P2, ..., P) m ); Based on the number of charges under each charging mode, the m charging modes (U1, U2, ..., U...) of user s are analyzed. m The probability of occurrence is calculated and expressed as, Each charging station uploads the final charging mode clustering results of users to the power grid cloud platform.

7. The power grid demand response user screening method based on secure data sharing as described in claim 6, characterized in that: The aforementioned demand response guidance strategy includes defining a load gap ΔS in the region during time period t on a certain day. t To guide electric vehicle users to participate in demand response, the overlap between charging mode j of electric vehicle user s and time period t is defined as the proportion of the time period in which charging mode j overlaps with time period t to the total charging time of charging mode j, expressed as: Where, λ j Δτ represents the overlap between charging mode j and time period t. j τ is the duration of the period when charging mode j overlaps with time period t. j The charging duration for charging mode j; Define the response potential ρ of user s charging mode j in the demand response process. j Let be the product of the overlap between charging mode j and time period t and the probability of charging mode j occurring, expressed as . r j =ω j l j The charging mode with the highest user response potential is defined as the unrecommended charging mode for demand response. The response potential of user s in this demand response process is defined as the maximum response potential of all charging modes, denoted as , ρ0=max(ρ j ) The response potential of all electric vehicle users in the region is ranked. Based on the load gap size and the ranking results of user demand response potential, a list of demand response users is selected. For some electric vehicle users on the demand response user list, users are guided not to charge during the load gap period through electricity price compensation. The charging information of users on the same day is collected through charging stations to verify whether users respond in accordance with the grid requirements. Compensation is given to users who actually participate in the response.

8. A system employing the power grid demand response user screening method based on data security sharing as described in any one of claims 1 to 7, characterized in that: It includes a data acquisition module, a data processing and clustering module, an encryption and data security module, a cloud platform computing module, a decryption and optimization module, and a demand response strategy formulation module; The data acquisition module collects data from the charging station as the basis for analysis and processing. The data processing and clustering module uses the K-means clustering algorithm to cluster data samples from charging stations and identify different charging modes; The encryption and data security module uses the Paillier homomorphic encryption algorithm to encrypt the clustering results; The cloud platform computing module performs calculations on the encrypted clustering results on the power grid cloud platform; The decryption and optimization module decrypts the encrypted calculation results returned by the cloud platform, calculates new common cluster centers based on the decrypted data, and continuously iterates and optimizes the clustering process. The demand response strategy formulation module is based on the user's charging mode. The power grid cloud platform selects suitable electric vehicle demand response users and formulates corresponding demand response guidance strategies.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the power grid demand response user screening method based on data security sharing as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the power grid demand response user screening method based on data security sharing as described in any one of claims 1 to 7.

Images