Equipment data trend privacy protection method and system

Abstract

The invention discloses an equipment data trend privacy protection method and system, and the method comprises the steps: enabling a block chain to serve as a data sharing platform, and enabling a differential privacy mechanism to be embedded into a smart contract of the block chain; the device user adds differential privacy disturbance to the collected device multi-dimensional fault data and then uploads the data to the block chain; the block chain receives an uploading transaction of a device user, and after verification is passed, a block is generated and added into the block chain; the device provider sends a query transaction to the block chain to obtain device multi-dimensional fault data of the device user; and the block chain responds to the query, and returns the multi-dimensional fault data of the equipment added with the disturbance to the equipment provider. According to the method, the sensitive information in the equipment fault data is ensured to be fully encrypted and protected by utilizing a differential privacy mechanism, the blank of data trend privacy protection in the prior art is filled, and safe data sharing between an equipment provider and a user is promoted.

Description

Technical Field

[0001] The present invention belongs to the field of data security technology, and in particular relates to a device data trend privacy protection method and system. Background Art

[0002] With the advancement of science and technology and the development of the times, the manufacturing industry has gradually moved from traditional manufacturing to intelligent manufacturing to realize the digitization, networking and intelligence of the manufacturing process. With the rise of intelligent manufacturing, data, as an invisible asset, not only has commercial research value, but also can promote the transformation of the manufacturing industry. In this context, Machine as a Service (MaaS) has emerged as an emerging industrial equipment service model. In order to focus more on research in their own fields and to further reduce costs, enterprises adopt the MaaS model to participate in production. Under the MaaS model, the original equipment manufacturer (Original Equipment Manufacturer, OEM) rents production equipment to the machine user (Machine User, MU) to help enterprises quickly build factories and start production, reduce production costs, and improve production efficiency. At the same time, OEM collects data information through the built-in data sensors of the equipment, understands the performance defects of the equipment, optimizes the performance of the equipment and upgrades after-sales service. However, equipment failure data contains the production privacy of the enterprise. For example, OEM can infer the privacy of the enterprise's production capacity, efficiency, etc. by analyzing various types of equipment failure data.

[0003] At present, Differential Privacy (DP) technology is a good choice to solve such problems. Differential privacy is a privacy protection technology proposed by Dwork that resists differential attacks. Depending on whether the cloud server is trustworthy, DP is divided into Central Differential Privacy (CDP) and Local Differential Privacy (LDP). However, in the actual application of CDP, the credibility of centralized cloud servers is difficult to guarantee. At present, trustworthiness is mainly achieved by decentralization technology, such as blockchain. Unlike CDP, LDP does not rely on trusted cloud servers. It allows users to independently allocate and add noise locally to meet users' personalized privacy protection needs.

[0004] In the MaaS model of smart manufacturing, various types of failures may occur in equipment, and each type of failure constitutes a dimension of failure data. When multi-dimensional failure data is displayed through a bar chart or a line chart, a failure data trend will be formed. Since the failure data trend is highly related to the privacy of the enterprise, analyzing the failure data trend will reveal the production privacy of the enterprise. For example, there is a situation where the enterprise shares equipment failure data with the OEM, among which the frequency of high-temperature failures is the highest. When the OEM analyzes the disturbance failure data trend, it finds that the frequency of high-temperature failures is still higher than other failures. From this, the OEM can infer that the company has recently processed a large number of high-temperature processed products. However, previous solutions based on differential privacy have not considered privacy leakage issues from data trends. Summary of the invention

[0005] In order to solve the above-mentioned problems existing in actual industrial production scenarios, the present invention provides a device data trend privacy protection method and system for intelligent manufacturing scenarios, which can effectively protect the privacy of equipment failure data trends of individual enterprises, and also provide original equipment manufacturers with a reliable basis for equipment improvement.

[0006] The technical solution provided by the present invention is specifically as follows:

[0007] A device data trend privacy protection method comprises the steps of:

[0008] S1. Use blockchain as a data sharing platform and embed differential privacy mechanism into smart contracts of blockchain;

[0009] S2. The equipment user adds differential privacy perturbations to the collected multi-dimensional equipment fault data and uploads it to the blockchain;

[0010] S3, the blockchain receives the uploaded transaction from the device user, generates a block after verification and adds it to the blockchain;

[0011] S4. The device provider sends a query transaction to the blockchain to obtain the multi-dimensional fault data of the device user;

[0012] S5. The blockchain responds to the query and returns the multi-dimensional fault data of the device after the disturbance is added to the device provider.

[0013] Furthermore, step S1 includes an initialization phase:

[0014] The blockchain generates cryptographic materials and generates node and channel configuration files;

[0015] Arrange nodes, assign permissions, and complete identity authentication for equipment users MU and equipment providers OEM:

[0016] The equipment user MU and the equipment provider OEM sign a data sharing contract to describe the shared data and transaction specifications;

[0017] Define upload transaction f i * ={Data i ′, User, time, upload} and query transaction q = {User, start_t, end_t, n, query}, where Data′ represents the main data, User represents the data uploader, time is the submission time, upload represents the transaction type is data upload; start_t and end_t represent the start time of the query data, n is the amount of data to be queried, and query represents the transaction type is data query;

[0018] Start the Orderer node as the initial node of the blockchain network, create a new channel through the channel configuration file, add the Peer node to the channel, and install and instantiate the chain code.

[0019] Furthermore, in step S2, the equipment user MU collects the number of occurrences of various types of equipment failures and forms an array Data i ={x i1 ,x i2 ,L,x iM}, x ij (1≤i≤N,1≤j≤M) represents the number of occurrences of the jth type of failure of the i-th MU. MU uses the NM mechanism to i Perform local differential privacy perturbation, and then perturb the array Data i ′={x i ' 1 ,x i ' 2 ,L,x i ' M} is encapsulated into an upload transaction; the NM mechanism controls the size of parameter b through privacy parameters ε and δ to generate two disturbance probabilities r and t, thereby controlling the disturbance intensity:

[0020]

[0021] In the formula, x ij is the data before disturbance, x i ' j is the perturbed data, that is, x i ' j =x ij +NMnoise.

[0022] Furthermore, step S3 includes transaction generation and endorsement as well as node sorting and broadcast verification:

[0023] Transaction generation and endorsement: The uploaded transaction will be submitted to the designated endorsement node. Members of the same channel will be required to perform endorsement, and the execution result will be sent to other endorsement nodes that need to endorse the transaction;

[0024] Node sorting and broadcast verification: The sorting node receives the endorsement results from the endorsing node and sorts them. The sorting node packages the sorted endorsement results into a block and broadcasts it to all verification nodes. The verification node receives the block and verifies the transaction processing contained in the block. If the verification passes, the verification node adds the block to the local blockchain and updates the status ledger record. Finally, the submission success notification is returned to the MU.

[0025] Furthermore, in step S5, the blockchain calls the ledger block according to the parameters of the query transaction, and the smart contract perturbs the various types of fault data being queried through the differential privacy mechanism. If the OEM only queries the fault data of a certain MU, strong noise is directly injected into the various types of fault data of the MU to disrupt its data trend, and the disturbed data is returned as the query result.

[0026] Furthermore, if the OEM queries the fault data of some MUs with which it has signed a sharing contract, then:

[0027] First, N strong noises are generated for each type of fault data, denoted as {noise 1j ,noise 2j ,L,noise Nj}; Then calculate the mean value of the same type of fault noise Then go to the mean noise ij =noise ij -ave_noise j so that The noise ij Traverse and add various fault data of the corresponding MU, and finally aggregate the queried fault data according to the fault type and return it as the query result. In this case, the aggregated data does not meet The condition causes the noise to accumulate in the summation process, making the equipment failure data trend of a single MU hidden in the aggregate trend of the queried equipment failure data.

[0028] Furthermore, if the OEM queries the fault data of all MUs with which it has signed a sharing contract, then:

[0029] First, N strong noises are generated for each type of fault data, denoted as {noise 1j ,noise 2j ,L,noise Nj}; Then calculate the mean value of the same type of fault noise Then go to the mean noise ij =noise ij -ave_noise j so that The noise ij Traverse and add to the various fault data of the corresponding MU, and finally aggregate the queried fault data according to the fault type and return it as the query result. In this case, the aggregated data meets The condition causes the noise to cancel each other out during the summation process, which can show the equipment failure data trend of a single MU.

[0030] A device data trend privacy protection system based on the above method includes an OEM end, a blockchain and several MU ends. The OEM is responsible for renting the equipment to the MU with which it has reached a data sharing contract. The MU shares the daily fault conditions of the equipment with the OEM through the blockchain as a third-party data sharing platform; the MU end includes a local differential privacy module and a data upload interface. The MU end uses the local differential privacy module to add differential privacy noise to the collected device fault data key and then uploads it to the blockchain through the data upload interface; the OEM end requests data access from the blockchain through a Web interface; the blockchain smart contract retrieves the stored data according to the access request, and responds to the request after differentially perturbing the retrieved data through the differential privacy processing module integrated in the chain code.

[0031] Preferably, the OEM end also includes a data display module for displaying the data queried from the blockchain in the form of a bar chart.

[0032] Preferably, the blockchain adopts Hyperledger Fabric.

[0033] Compared with the prior art, the present invention has at least the following beneficial effects:

[0034] 1. Effectively protect data privacy: By perturbing the data before uploading it to the sharing platform, the differential privacy mechanism is used to ensure that sensitive information in the equipment failure data is fully encrypted and protected, effectively avoiding the risk of privacy leakage even in a highly complex manufacturing environment.

[0035] 2. Ensure data integrity and authenticity: Using blockchain as the infrastructure for data storage and management not only enhances data security, but also ensures data integrity and immutability, allowing all participants to obtain authentic and reliable data records, thereby promoting the establishment of cooperative trust.

[0036] 3. Improve the utilization of data value: By sharing equipment failure data, manufacturers can gain important insights into the operating status of equipment, which is of great value in predicting potential failures, optimizing production processes, improving equipment availability and reducing maintenance costs.

[0037] In summary, the present invention promotes secure data sharing between equipment providers and users, which not only fills the gap in data trend privacy protection in the prior art, but also provides a more secure and reliable solution for equipment data management and industrial automation through a series of technical innovation measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0039] Figure 1 is a schematic diagram of a system framework provided by an embodiment of the present invention;

[0040] Figure 2 It is a schematic diagram of a system workflow provided by an embodiment of the present invention;

[0041] Figure 3 This is a schematic diagram of a data aggregation process during OEM query provided by an embodiment of the present invention;

[0042] Figure 4 is a schematic diagram of data aggregation in different situations provided by an embodiment of the present invention;

[0043] Figure 5 is a bar graph of query responses provided by an embodiment of the present invention;

[0044] Figure 6 This is a diagram of system throughput and latency provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, other embodiments obtained by ordinary technicians in this field without making creative work are all within the scope of protection of the present invention.

[0046] Embodiment 1

[0047] This embodiment provides a device data trend privacy protection method, refer to Figure 1 and Figure 2 , mainly including the steps:

[0048] Step 1: Initialization phase

[0049] Select Hyperledger Fabric blockchain as the third-party server to deploy nodes, assign permissions, and complete identity authentication for the device user MU and the device provider OEM. Embed the differential privacy mechanism into the blockchain smart contract and run it in the query phase. The specific process is as follows:

[0050] Step 1.1: The blockchain generates encryption materials, including identity certificates and private keys, and generates node and channel configuration files, completing the first data synchronization and consensus operation with all client nodes.

[0051] Step 1.2: OEM and MU sign a data sharing contract, which includes the sharing objectives, background and interests; describes the specifications of shared data and transactions, including data format, transaction format, sharing frequency and noise addition standards.

[0052] Step 1.3: Define upload transaction f i * ={Data i ′, User, time, upload}, where Data′ represents the main data, User represents the data uploader, time represents the submission time, and upload represents the transaction type is data upload; define query transaction q = {User, start_t, end_t, n, query}, where start_t and end_t represent the time period of data to be queried, n is the number of MUs to be queried, and query represents the transaction type is query.

[0053] Step 1.4: Start the Orderer node as the initial node of the blockchain network. Create a new channel through the channel configuration file and add the Peer node to the channel. Install and instantiate the chain code.

[0054] Step 1.5: After the network nodes complete the first consensus, the blockchain system is successfully started.

[0055] Step 2: Data submission phase

[0056] After the system initialization is completed, the system enters the data submission stage. The specific process is as follows:

[0057] Step 2.1: MU collects the number of occurrences of various types of equipment failures, such as wear failures, strain failures, etc., through equipment sensors, and summarizes them into statistical data, which is expressed as Data i ={x i1 ,x i2 ,L x iM}, x ij(1≤i≤N,1≤j≤M) represents the number of occurrences of the j-th type of fault of the i-th MU.

[0058] Algorithm 1: Adding local noise to data

[0059]

[0060] Step 2.2: To prevent the fault data from leaking production privacy, MU uses the differential privacy NM mechanism to perform local differential privacy perturbations on the fault data.

[0061] The NM mechanism controls the magnitude of the two disturbance probabilities r and t through ε and δ. i ' j |x ij ) represents the probability of data perturbation change, ε and δ represent the differential privacy noise generation parameters, and these two parameters are used to control the data perturbation probability, as shown in Algorithm 1. Parameter b is used to compare the difference between the original value of the data and the perturbation value. It decreases as the privacy parameter ε increases. By adjusting the size of b, data availability and privacy can be balanced. The mechanism generates differential privacy noise NMnoise and perturbation fault data x that conform to the Gaussian distribution through ε and δ. i ' j =x ij +NMnoise. The mechanism then compares the difference between the original data value and the perturbation value, and maintains the perturbed data in the nearby interval (|x ij -x i ' j |≤b) to improve data availability; maintain the perturbed data in a farther interval (|x ij -x i ' j |>b) to enhance data privacy (refer to the paper “Mean Estimation Mechanism under (ε, δ)-Local Differential Privacy Model”).

[0062] Finally, each data is traversed and noised through the NM mechanism, and finally the disturbance fault array is obtained, namely Data i ={x i1 ,x i2 ,L x iM}Perturbation is Data i ′={x i ' 1 ,x i ' 2 ,L,x i ' M}.

[0063] Step 2.3: Data i ′,User,time is encapsulated into a transaction f i* .

[0064] Step 2.4: MU will f i * Sent to the blockchain platform for OEM inquiry.

[0065] Step 3: Transaction verification and storage phase

[0066] The specific steps are as follows:

[0067] Step 3.1: Transaction generation and endorsement: Whether querying data or uploading data, it can be regarded as a transaction. First, the transaction will be submitted to the designated endorsement node, and the request will include the chain code to be executed and the corresponding parameters. This process will not change the state of the ledger. Members of the same channel will be required to perform endorsement. The execution result will then be sent to other endorsement nodes that need to endorse the transaction.

[0068] Step 3.2: Node sorting and broadcast verification: The sorting node receives enough endorsement results from the endorsing nodes and sorts them. The sorting node packages the sorted endorsement results into a block and broadcasts it to all verification nodes. The verification node receives the block and verifies the transaction processing contained in the block. If the verification passes, the verification node adds the block to the local blockchain and updates the status ledger record. Finally, the submission success notification is returned to the MU.

[0069] Step 4: Data query phase, refer to Figure 3 The specific steps are as follows:

[0070] Step 4.1: The OEM sends a query transaction q = {User, start_t, end_t, n, query} to the blockchain to obtain various types of fault data of MU equipment, analyze equipment defects, and optimize equipment performance and after-sales service quality.

[0071] Step 4.2: To prevent OEMs from obtaining the failure data trends of individual users, the smart contract provides fine-grained privacy protection for the failure data trends through a differential privacy mechanism. The specific steps are as shown in Algorithm 2.

[0072] Step 4.3: OEM sends a data query request q = {User, start_t, end_t, n, query} to the blockchain, where the parameters represent: which MUs to query, what time period to query the uploaded data, and how many MUs to query.

[0073] Algorithm 2 Classification query algorithm based on DP

[0074]

[0075] Step 4.4: The blockchain calls the ledger block according to the transaction parameters, and the smart contract disturbs various fault data in it through the differential privacy mechanism. Figure 3 and Figure 4 , the disturbance operation is specifically divided into the following three cases:

[0076] Case 1: The OEM only queries the fault data of a certain MU. In this case, data aggregation is not required, and strong noise is directly injected into various data to disrupt their data trends and return them as query results.

[0077] Case 2: The OEM queries the data of some MUs that have signed a sharing agreement with it. In this case, the differential privacy mechanism first generates N strong noises for each type of fault data, denoted as {noise 1j ,noise 2j ,L,noise Nj}; Then the mean value of the same type of fault noise is calculated by removing the mean value Then through noise ij =noise ij -ave_noise j To achieve Secondly, the noise ij The traversal is added to various types of fault data, and this operation is repeated in a loop. Finally, the queried fault data is aggregated according to the fault type, that is, Because only part of the data is queried, the aggregated data does not meet the requirements. The condition causes the noise to accumulate in the summation process and cannot be eliminated. Ultimately, all types of fault data will be affected by strong noise, thereby changing the aggregation trend of the equipment fault data in the queried area, making the equipment fault data trend of a single MU hidden in the aggregation trend of the queried equipment fault data.

[0078] Case 3: OEM queries the data of all MUs that have signed a sharing agreement with it, and the smart contract repeats the steps in Case 2. This case can satisfy Condition, the noise cancels each other out when the data is aggregated, showing the trend of MU failure data. The key to distinguishing Case 2 from Case 3 in Algorithm 2 is n and N. The relationship between n and N determines the noise. ij Whether the same type of fault data can be offset when aggregated.

[0079] Step 5: In the data response phase, the blockchain encapsulates the perturbed data query results into a transaction and feeds back the query results to the OEM in the form of a bar chart through the client web page. After receiving the response, the OEM analyzes it and uses the equipment failure information to optimize the equipment design, update the equipment version, and improve the quality of after-sales maintenance services.

[0080] Embodiment 2

[0081] Based on the above method, this embodiment provides a device data trend privacy protection system, referring to Figure 1 The system includes OEM, MU and blockchain. OEM is the original equipment manufacturer, responsible for leasing equipment to MU for industrial production. MU is the equipment user. MU and OEM reach a data sharing agreement to share daily equipment failures and help OEM improve equipment performance. Blockchain, as a third-party data storage server, is responsible for data storage, sharing, noise addition, and distribution of permissions to participants.

[0082] MU maintains statistics on the raw data collected from the device sensors, adds differential privacy noise to the raw statistical data through the local differential privacy module, and then uploads it to the blockchain through the MU client interface. The blockchain uses the storage module to reach consensus, sort, and store the data. OEM requests data access through the OEM client web interface. The blockchain smart contract receives and retrieves the data based on the name and date parameters of the access request. The data is processed by the differential privacy processing module integrated in the chain code and returned to the web page, and displayed in the form of a bar chart. Figure 5 As shown, a bar chart of query responses for three situations is shown.

[0083] The above-mentioned system can be deployed to implement the device data trend privacy protection method described in Example 1, and has the corresponding functional modules and beneficial effects of the method. For technical details that are not fully described in this embodiment, please refer to the device data trend privacy protection method provided in Example 1 of the present invention. Through this system, multiple MUs collect multi-dimensional fault data of leased equipment, and upload the fault data to the blockchain after differential privacy perturbation; for data query requests initiated by the OEM, the system adds differential privacy noise to the queried user data to perturb the multi-dimensional fault data trend; if all user data are queried, the noise will cancel each other out when all user data are aggregated, and finally the true fault data trend will be displayed. This system takes into account the characteristics of multi-dimensional data with data trends from the data level, and designs algorithms to protect data trend privacy. In addition, if Figure 6 As shown, the system also has efficient data transmission performance.

[0084] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may also be combined, the steps may be implemented in any order, and there are many other changes in different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person of ordinary skill in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A device data trend privacy protection method, characterized in that: Includes steps: S1. Use blockchain as a data sharing platform and embed differential privacy mechanism into smart contracts of blockchain; S2. The equipment user adds differential privacy perturbations to the collected multi-dimensional equipment fault data and uploads it to the blockchain; S3, the blockchain receives the uploaded transaction from the device user, generates a block after verification and adds it to the blockchain; S4. The device provider sends a query transaction to the blockchain to obtain the multi-dimensional fault data of the device user; S5. The blockchain responds to the query and returns the multi-dimensional fault data of the device after adding the disturbance to the device provider.

2. A device data trend privacy protection method as claimed in claim 1, characterized in that: Step S1 includes an initialization phase: The blockchain generates cryptographic materials and generates node and channel configuration files; Arrange nodes, assign permissions, and complete identity authentication for equipment users MU and equipment providers OEM: The equipment user MU and the equipment provider OEM sign a data sharing contract to describe the shared data and transaction specifications; Define upload transaction f i * ={Data i ′, User, time, upload} and query transaction q = {User, start_t, end_t, n, query}, where Data′ represents the main data, User represents the data uploader, time is the submission time, upload represents the transaction type is data upload; start_t and end_t represent the start time of the query data, n is the amount of data to be queried, and query represents the transaction type is data query; Start the Orderer node as the initial node of the blockchain network, create a new channel through the channel configuration file, add the Peer node to the channel, and install and instantiate the chain code.

3. A device data trend privacy protection method as claimed in claim 2, characterized in that: In step S2, the equipment user MU collects the number of occurrences of various types of equipment failures and forms an array Data i ={x i1 ,x i2 ,L,x iM }, x ij (1≤i≤N,1≤j≤M) represents the number of occurrences of the jth type of failure of the i-th MU. MU uses the NM mechanism to i Perform local differential privacy perturbation, and then perturb the array Data i ′={x i ′1,x i ′2,L,x i ' M } is encapsulated into an upload transaction; the NM mechanism controls the size of parameter b through privacy parameters ε and δ to generate two disturbance probabilities r and t, thereby controlling the disturbance intensity: In the formula, x ij is the data before disturbance, x i ' j is the perturbed data, that is, x i ' j =x ij +NMnoise, NMnoise is differential privacy noise, P(x i ' j |x ij ) represents the probability of data disturbance change.

4. A device data trend privacy protection method as claimed in claim 2, characterized in that: Step S3 includes transaction generation and endorsement as well as node sorting and broadcast verification: Transaction generation and endorsement: The uploaded transaction will be submitted to the designated endorsement node. Members of the same channel will be required to perform endorsement, and the execution result will be sent to other endorsement nodes that need to endorse the transaction; Node sorting and broadcast verification: The sorting node receives the endorsement results from the endorsing node and sorts them. The sorting node packages the sorted endorsement results into a block and broadcasts it to all verification nodes. The verification node receives the block and verifies the transaction processing contained in the block. If the verification passes, the verification node adds the block to the local blockchain and updates the status ledger record. Finally, the submission success notification is returned to the MU.

5. A device data trend privacy protection method as claimed in claim 2, characterized in that: In step S5, the blockchain calls the ledger block according to the parameters of the query transaction, and the smart contract disturbs the various fault data being queried through the differential privacy mechanism. If the OEM only queries the fault data of a certain MU, strong noise is directly injected into the various fault data of the MU to disrupt its data trend. And the disturbed data is returned as the query result.

6. A device data trend privacy protection method as claimed in claim 5, characterized in that: If an OEM queries the fault data of some MUs with which it has signed a sharing contract, then: First, N strong noises are generated for each type of fault data, denoted as {noise 1j ,noise 2j ,L,noise Nj }; Then calculate the mean value of the same type of fault noise Then go to the mean noise ij =noise ij -ave_noise j so that The noise ij Traverse and add various fault data of the corresponding MU, and finally aggregate the queried fault data according to the fault type and return it as the query result. In this case, the aggregated data does not meet The condition causes the noise to accumulate in the summation process, making the equipment failure data trend of a single MU hidden in the aggregate trend of the queried equipment failure data.

7. A device data trend privacy protection method as claimed in claim 5, characterized in that: If an OEM queries the fault data of all MUs with which it has signed a sharing contract, then: First, N strong noises are generated for each type of fault data, denoted as {noise 1j ,noise 2j ,L,noise Nj }; Then calculate the mean value of the same type of fault noise Then go to the mean noise ij =noise ij -ave_noise j so that The noise ij Traverse and add to the various fault data of the corresponding MU, and finally aggregate the queried fault data according to the fault type and return it as the query result. In this case, the aggregated data meets The condition causes the noise to cancel each other out during the summation process, which can show the equipment failure data trend of a single MU.

8. A device data trend privacy protection system based on the method according to any one of claims 1 to 7, characterized in that: It includes an OEM end, a blockchain and several MU ends. The OEM is responsible for renting the equipment to the MU with which it has reached a data sharing contract. The MU shares the daily failure conditions of the equipment with the OEM through the blockchain as a third-party data sharing platform; the MU end includes a local differential privacy module and a data upload interface. The MU end uses the local differential privacy module to add differential privacy noise to the collected equipment failure data key, and then uploads it to the blockchain through the data upload interface; the OEM end requests data access from the blockchain through the Web interface; the blockchain smart contract retrieves the stored data according to the access request, and responds to the request after differentially perturbing the retrieved data through the differential privacy processing module integrated in the chain code.

9. The device data trend privacy protection system according to claim 8, characterized in that: The OEM end also includes a data display module for displaying the data queried from the blockchain in the form of a bar chart.

10. The device data trend privacy protection system according to claim 8, characterized in that: The blockchain adopts Hyperledger Fabric.

Images