Industrial Internet of Things data privacy calculation method and system for equipment maintenance

Abstract

The invention discloses an industrial Internet of Things data privacy calculation method and system for equipment maintenance, and the method comprises the steps: obtaining the maintenance data of all equipment, and converting the maintenance data of each piece of equipment into a corresponding signal vector; fusing the signalized vector with a context information vector of the equipment to generate a behavior image without semantic recognition, and forming a behavior image set of all the equipment; and carrying out structural collaboration judgment on the behavior mapping elements in the behavior mapping element set, finding out conjugate mapping element pairs, carrying out life cycle control on each conjugate mapping element pair, and destroying the invalid conjugate mapping element pairs by taking the conjugate mapping element pairs of which the calling times exceed the maximum calling times as invalid conjugate mapping element pairs.

Description

Technical Field

[0001] This invention belongs to the field of equipment data privacy protection technology, and more specifically, relates to an industrial Internet of Things data privacy calculation method and system for equipment maintenance. Background Technology

[0002] In the existing technical system for privacy protection and collaborative processing in the Industrial Internet of Things (IIoT), most solutions adopt the traditional model of "encrypted transmission + access control + centralized coordination." The core idea of ​​this approach is to protect data through encryption methods (such as TLS, AES, Homomorphic Encryption, etc.), while using access control policies (RBAC, ABAC, OAuth, etc.) to manage the identities and restrict permissions of devices, users, or systems. This is further supplemented by centralized structures such as edge gateways and cloud platforms to complete information exchange and collaborative computing among multiple devices. Although this technical path has a mature standard system and a certain degree of engineering feasibility, it has exposed many systemic problems in actual deployment.

[0003] First, security relies on a central node. In existing technologies, all critical encryption key management, access control decisions, and data integration and scheduling are often concentrated on one or a few central nodes. This makes the entire system highly vulnerable to single points of failure, key leaks, or identity spoofing attacks. Once the central platform is compromised, the data security and control logic of the entire network will face the risk of collapse.

[0004] Secondly, there is a conflict between collaborative capabilities and privacy protection. In scenarios where multiple industrial devices need to make collaborative decisions and respond in unison, traditional mechanisms often require the prior "de-identification" or "partial disclosure" of some data content before data matching or policy negotiation between devices can be completed. However, such partial exposure is still essentially an infringement on the privacy of data subjects, especially in classified equipment or critical infrastructure, where any form of data sharing is considered a high-risk behavior.

[0005] Furthermore, there is a lack of effective management mechanisms for the data lifecycle. Existing technologies mostly focus on issues of "data protection" and "data access," but rarely address the mechanism of "data extinction." Once data enters a system, it may be stored in databases, log files, and cache modules for a long time. Even after a collaborative task has been completed, the relevant behavioral traces may still be reconstructed, traced, or misused, failing to meet the rigid privacy requirements of industrial systems for "clearing data upon task completion."

[0006] In summary, while existing privacy protection technologies for the Industrial Internet of Things (IIoT) are theoretically relatively sound, they still face key challenges when dealing with the demands of high collaboration, high sensitivity, and high reliability in industrial settings. These challenges include severe central dependency, conflicts between privacy and collaboration, uncontrollable data, and high deployment barriers. Summary of the Invention

[0007] To address the above technical problems, this invention proposes an industrial IoT data privacy computing method for equipment maintenance, comprising:

[0008] Acquire maintenance data for all devices and convert the maintenance data for each device into a corresponding signal vector;

[0009] The signaling vector is fused with the device's context information vector to generate a behavior map that lacks semantic recognition, and a set of behavior maps for all devices is formed.

[0010] Structural synergy is assessed for behavior mappings in the behavior mapping set to identify conjugate mapping pairs. Lifecycle control is applied to each conjugate mapping pair, and conjugate mapping pairs whose call count exceeds the maximum call count are identified as invalid conjugate mapping pairs and destroyed.

[0011] Furthermore, converting the maintenance data of each device into a corresponding signaling vector includes: converting each behavior type in the maintenance data of each device into a corresponding behavior vector; and weighting and fusing the behavior vectors according to a time window to generate a new behavior vector as a signaling vector.

[0012] Furthermore, fusing the signaling vector with the device's context information vector includes fusing the signaling vector of each behavior type in the maintenance data of each device with the corresponding category's context information vector. The signaling vector and the context information vector are added together and then input into the tanh function to complete the fusion.

[0013] Furthermore, generating behavior maps that lack semantic recognition capabilities includes: using the result of fusing the signaling vector with the device's context information vector as the behavior map for each behavior type in the maintenance data of each device.

[0014] Furthermore, it also includes: collecting each behavior type of each device once every time window, converting each behavior type of the current device collected in each time window into a signal vector of the current device in each time window, and finally converting the signal vector of the current device in each time window into behavior images of multiple time windows.

[0015] Furthermore, before performing structural synergy judgment on the behavioral mappings in the behavioral mapping set, the process includes: combining the behavioral mappings of multiple time windows of the current device into a one-dimensional vector sequence, dividing the one-dimensional vector sequence into multiple sub-one-dimensional vector sequences according to a preset window length, performing a Fourier transform on each sub-one-dimensional vector sequence, and generating the spectral tensor of each sub-one-dimensional vector sequence.

[0016] Furthermore, structural synergy judgment is performed on the behavioral mappings in the behavioral mapping set to find conjugate mapping pairs. This includes: unifying the size of each spectral tensor and performing a normalization operation to generate a new spectral tensor; identifying pairs of new spectral tensors between different devices whose similarity exceeds a preset threshold as conjugate mapping pairs; clustering the conjugate mapping pairs according to the labeled device fault categories; and having the user take corresponding device maintenance measures based on the device fault categories.

[0017] Furthermore, lifecycle control for each conjugate pair includes: starting from the establishment time of each conjugate pair, if the current time exceeds a preset survival time threshold, the corresponding conjugate pair is marked as expired.

[0018] This invention also proposes an industrial IoT data privacy computing system for equipment maintenance, comprising:

[0019] The signal vector generation module is used to acquire maintenance data for all devices and convert the maintenance data for each device into a corresponding signal vector.

[0020] The behavior mapping generation module is used to fuse the signaling vector with the device's context information vector to generate a behavior mapping that does not have semantic recognition, and form a set of behavior mappings for all devices.

[0021] The module for generating conjugate mapping pairs is used to determine the structural synergy of behavioral mapping pairs in the behavioral mapping set, identify conjugate mapping pairs, control the lifecycle of each conjugate mapping pair, and destroy conjugate mapping pairs that exceed the maximum number of calls as invalid conjugate mapping pairs.

[0022] Furthermore, converting the maintenance data of each device into a corresponding signaling vector includes: converting each behavior type in the maintenance data of each device into a corresponding behavior vector; and weighting and fusing the behavior vectors according to a time window to generate a new behavior vector as a signaling vector.

[0023] In summary, the technical solutions conceived by this invention have the following beneficial effects compared with the prior art:

[0024] This invention not only constructs a closed-loop mechanism from perception to collaboration to data self-destruction at the technical implementation level, but also achieves a multiple balance of "high security + low intrusion + collaboration" in its architectural design, making it particularly suitable for scenarios such as industrial control, remote operation and maintenance, and intelligent manufacturing where privacy protection requirements are extremely high. Attached Figure Description

[0025] Figure 1 This is a flowchart of the method in Embodiment 1 of the present invention;

[0026] Figure 2This is a system structure diagram of Embodiment 2 of the present invention. Detailed Implementation

[0027] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0028] The method provided by this invention can be implemented in a terminal environment that may include one or more of the following components: a processor, a storage medium, and a display screen. The storage medium stores at least one instruction, which is loaded and executed by the processor to implement the method described in the following embodiments.

[0029] A processor may include one or more processing cores. The processor uses various interfaces and lines to connect various parts of the terminal, and performs various functions and processes data by running or executing instructions, programs, code sets or instruction sets stored in the storage medium, and by calling data stored in the storage medium.

[0030] Storage media can include random access memory (RAM) or read-only memory (ROM). Storage media can be used to store instructions, programs, code, code sets, or instructions.

[0031] The display screen is used to show the user interface of each application.

[0032] In addition, those skilled in the art will understand that the structure of the terminal described above does not constitute a limitation on the terminal. The terminal may include more or fewer components, or combine certain components, or have different component arrangements. For example, the terminal may also include radio frequency circuits, input units, sensors, audio circuits, power supplies, and other components, which will not be described in detail here.

[0033] Example 1

[0034] like Figure 1 As shown, this embodiment proposes an industrial IoT data privacy computing method for equipment maintenance, including:

[0035] Step 101: Obtain maintenance data for all devices and convert the maintenance data for each device into a corresponding signal vector;

[0036] Preferably, in this embodiment, the equipment maintenance data can be multiple behavior types such as maintenance history, operation log, anomaly record, component replacement, and operating status.

[0037] Specifically, each behavior type in the maintenance data of each device is transformed into a corresponding behavior vector; the behavior vectors are then weighted and fused according to a time window to generate a new behavior vector, which is then used as a signaling vector.

[0038] Specifically, it also includes: collecting each behavior type of each device once every time window, converting each behavior type of the current device collected in each time window into a signal vector of the current device in each time window, and finally converting the signal vector of the current device in each time window into behavior images of multiple time windows.

[0039] Preferably, this embodiment converts each behavior type in the maintenance data of each device into a corresponding behavior vector in the following way:

[0040] Assign a fixed numerical token to each behavior type, and compress the numerical token to the range (-1, 1) using the tanh function. The compressed value is then a behavior vector. Here is an example:

[0041] Behavioral types Numeric token tanh compressed value Alarm incident 1 ≈0.76 runtime error 1.5 ≈0.90 Regular maintenance 2 ≈0.96 shutdown -1 ≈-0.76 Restart -1.5 ≈-0.90 Communication interruption 0.8 ≈0.66 Power fluctuations 0.5 ≈0.46 Too high temperature 1.2 ≈0.83 Diagnosis completed -0.8 ≈-0.66 Abnormal self-recovery -0.5 ≈-0.46

[0042] Suppose that within a time window, the device experiences: alarm events, shutdown, scheduled maintenance, communication interruption, and power fluctuations. Then, the token value for each behavior type within that time window is [1.0, -1.0, 2.0, 0.8, 0.5]. After compression using the tanh function, the resulting behavior vector is [0.76, -0.76, 0.96, 0.66, 0.46]. If the device only experiences three behavior types within a time window, such as alarm events, shutdown, and scheduled maintenance, then the last two digits of the behavior vector after tanh compression are padded with 0.00.

[0043] Preferably, in this embodiment, the behavior vectors are weighted and fused according to a time window to generate a new behavior vector, specifically as follows:

[0044]

[0045] in, Let N be the new behavior vector, N be the behavior vectors of the past N times (i.e., the behavior vectors of the past N time windows), and α be the initial weight. For the i-th row vector, the smaller the index i, the newer the time, and vice versa. α i α is the weight of the i-th row vector; the larger the value of index i, the greater the weight of α. i The smaller the value.

[0046] This embodiment provides the following example to explain the meaning of the above formula:

[0047]

[0048] As can be seen from the table If it is a five-dimensional vector, then in index 1... The value of the first dimension and the corresponding α i The value after multiplication is the same as that in index 2. The value of the first dimension and the corresponding α i Add the values ​​after multiplication until all values ​​in sequence 5 have been accumulated. The value of the first dimension and the corresponding α i The value after multiplication is used as The value of the first dimension, that is:

[0049] 0.8·0.7+0.64·0.5+0.512·0.3+0.4096·0.2+0.32768·0.1=1.148288, as The value of the first dimension, and so on, for the 5... The values ​​of the second dimension are accumulated using the above method to obtain... The value of the second dimension is finally obtained. The values ​​of all five dimensions.

[0050] Step 102: The signaling vector is fused with the device's context information vector to generate a behavior map that does not have semantic recognition, and a set of behavior maps for all devices is formed.

[0051] Specifically, fusing the signaling vector with the device's context information vector includes: fusing the signaling vector of each behavior type in the maintenance data of each device with the corresponding category's context information vector. The signaling vector and the context information vector are added together and then input into the tanh function to complete the fusion.

[0052] This embodiment describes the relationship between the signaling vector and the device context information vector using the following table:

[0053]

[0054]

[0055] The vector values ​​for the corresponding context information can be assigned manually or looked up in a table. The following is an example of this embodiment, but this embodiment is not limited to this example. Users can customize the lookup table for the vector values ​​of context information. The specific table is as follows:

[0056]

[0057]

[0058] Specifically, generating behavior maps that lack semantic recognition capabilities includes: using the result of fusing the signaling vector with the device's context information vector as the behavior map for each behavior type in the maintenance data of each device.

[0059] Step 103: Perform structural coordination judgment on the behavior mappings in the behavior mapping set, find conjugate mapping pairs, perform lifecycle control on each conjugate mapping pair, and destroy conjugate mapping pairs whose call count exceeds the maximum call count as invalid conjugate mapping pairs.

[0060] Specifically, before performing structural synergy judgment on the behavior mappings in the behavior mapping set, the process includes: combining the behavior mappings of multiple time windows of the current device into a one-dimensional vector sequence, dividing the one-dimensional vector sequence into multiple sub-one-dimensional vector sequences according to a preset window length, performing a Fourier transform on each sub-one-dimensional vector sequence, and generating the spectral tensor of each sub-one-dimensional vector sequence.

[0061] Specifically, the structural synergy judgment of behavioral mappings in the behavioral mapping set is performed to find conjugate mapping pairs. This includes: unifying the size of each spectral tensor (e.g., interpolation resampling to 64×32) and performing normalization operations (e.g., 0 mean, 1 variance, or min-max) to generate new spectral tensors; identifying pairs of new spectral tensors between different devices whose similarity exceeds a preset threshold as conjugate mapping pairs; clustering the conjugate mapping pairs according to the labeled device fault categories; and having users take corresponding device maintenance measures based on the device fault categories.

[0062] Specifically, lifecycle control for each conjugate pair includes: starting from the establishment time of each conjugate pair, if the current time exceeds a preset survival time threshold, the corresponding conjugate pair is marked as expired.

[0063] Example 2

[0064] like Figure 2 As shown, this embodiment proposes an industrial IoT data privacy computing system for equipment maintenance, including:

[0065] The signal vector generation module is used to acquire maintenance data for all devices and convert the maintenance data for each device into a corresponding signal vector.

[0066] Specifically, converting the maintenance data of each device into a corresponding signaling vector includes: converting each behavior type in the maintenance data of each device into a corresponding behavior vector; and weighting and fusing the behavior vectors according to a time window to generate a new behavior vector as a signaling vector.

[0067] The behavior mapping generation module is used to fuse the signaling vector with the device's context information vector to generate a behavior mapping that does not have semantic recognition, and form a set of behavior mappings for all devices.

[0068] Specifically, fusing the signaling vector with the device's context information vector includes: fusing the signaling vector of each behavior type in the maintenance data of each device with the corresponding category's context information vector. The signaling vector and the context information vector are added together and then input into the tanh function to complete the fusion.

[0069] Specifically, generating behavior maps that lack semantic recognition capabilities includes: using the result of fusing the signaling vector with the device's context information vector as the behavior map for each behavior type in the maintenance data of each device.

[0070] The module for generating conjugate mapping pairs is used to determine the structural synergy of behavioral mapping pairs in the behavioral mapping set, identify conjugate mapping pairs, control the lifecycle of each conjugate mapping pair, and destroy conjugate mapping pairs that exceed the maximum number of calls as invalid conjugate mapping pairs.

[0071] Specifically, it also includes: collecting each behavior type of each device once every time window, converting each behavior type of the current device collected in each time window into a signal vector of the current device in each time window, and finally converting the signal vector of the current device in each time window into behavior images of multiple time windows.

[0072] Specifically, before performing structural synergy judgment on the behavior mappings in the behavior mapping set, the process includes: combining the behavior mappings of multiple time windows of the current device into a one-dimensional vector sequence, dividing the one-dimensional vector sequence into multiple sub-one-dimensional vector sequences according to a preset window length, performing a Fourier transform on each sub-one-dimensional vector sequence, and generating the spectral tensor of each sub-one-dimensional vector sequence.

[0073] Specifically, the structural synergy judgment of behavioral mappings in the behavioral mapping set and the identification of conjugate mapping pairs include: unifying the size of each spectral tensor and performing a normalization operation to generate a new spectral tensor; identifying pairs of new spectral tensors between different devices whose similarity exceeds a preset threshold as conjugate mapping pairs; clustering the conjugate mapping pairs according to the labeled device fault categories; and having users take corresponding device maintenance measures based on the device fault categories.

[0074] Specifically, lifecycle control for each conjugate pair includes: starting from the establishment time of each conjugate pair, if the current time exceeds a preset survival time threshold, the corresponding conjugate pair is marked as expired.

[0075] Example 3

[0076] This invention also proposes a storage medium storing multiple instructions for implementing the aforementioned industrial IoT data privacy computation method for equipment maintenance.

[0077] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0078] Optionally, in this embodiment, the storage medium is configured to store program code for performing the following method steps: Step 101, acquire maintenance data of all devices, and convert the maintenance data of each device into a corresponding signal vector;

[0079] Specifically, converting the maintenance data of each device into a corresponding signaling vector includes: converting each behavior type in the maintenance data of each device into a corresponding behavior vector; and weighting and fusing the behavior vectors according to a time window to generate a new behavior vector as a signaling vector.

[0080] Step 102: The signaling vector is fused with the device's context information vector to generate a behavior map that does not have semantic recognition, and a set of behavior maps for all devices is formed.

[0081] Specifically, fusing the signaling vector with the device's context information vector includes: fusing the signaling vector of each behavior type in the maintenance data of each device with the corresponding category's context information vector. The signaling vector and the context information vector are added together and then input into the tanh function to complete the fusion.

[0082] Specifically, generating behavior maps that lack semantic recognition capabilities includes: using the result of fusing the signaling vector with the device's context information vector as the behavior map for each behavior type in the maintenance data of each device.

[0083] Step 103: Perform structural coordination judgment on the behavior mappings in the behavior mapping set, find conjugate mapping pairs, perform lifecycle control on each conjugate mapping pair, and destroy conjugate mapping pairs whose call count exceeds the maximum call count as invalid conjugate mapping pairs.

[0084] Specifically, it also includes: collecting each behavior type of each device once every time window, converting each behavior type of the current device collected in each time window into a signal vector of the current device in each time window, and finally converting the signal vector of the current device in each time window into behavior images of multiple time windows.

[0085] Specifically, before performing structural synergy judgment on the behavior mappings in the behavior mapping set, the process includes: combining the behavior mappings of multiple time windows of the current device into a one-dimensional vector sequence, dividing the one-dimensional vector sequence into multiple sub-one-dimensional vector sequences according to a preset window length, performing a Fourier transform on each sub-one-dimensional vector sequence, and generating the spectral tensor of each sub-one-dimensional vector sequence.

[0086] Specifically, the structural synergy judgment of behavioral mappings in the behavioral mapping set and the identification of conjugate mapping pairs include: unifying the size of each spectral tensor and performing a normalization operation to generate a new spectral tensor; identifying pairs of new spectral tensors between different devices whose similarity exceeds a preset threshold as conjugate mapping pairs; clustering the conjugate mapping pairs according to the labeled device fault categories; and having users take corresponding device maintenance measures based on the device fault categories.

[0087] Specifically, lifecycle control for each conjugate pair includes: starting from the establishment time of each conjugate pair, if the current time exceeds a preset survival time threshold, the corresponding conjugate pair is marked as expired.

[0088] Example 4

[0089] This invention also proposes an electronic device, including a processor and a storage medium connected to the processor. The storage medium stores multiple instructions, which can be loaded and executed by the processor to enable the processor to execute the aforementioned industrial IoT data privacy computing method for equipment maintenance.

[0090] Specifically, the electronic device in this embodiment can be a computer terminal, which may include one or more processors and a storage medium.

[0091] The storage medium can be used to store software programs and modules, such as the industrial IoT data privacy computing method for equipment maintenance in this embodiment of the invention. The corresponding program instructions / modules allow the processor to execute various functional applications and data processing by running the software programs and modules stored in the storage medium, thus realizing the aforementioned industrial IoT data privacy computing method for equipment maintenance. The storage medium may include high-speed random access storage media, and may also include non-volatile storage media, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state storage media. In some instances, the storage medium may further include storage media remotely configured relative to the processor, which can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0092] The processor can call the information and application stored in the storage medium through the transmission system to execute the following method steps: Step 101, acquire the maintenance data of all devices and convert the maintenance data of each device into a corresponding signal vector;

[0093] Specifically, converting the maintenance data of each device into a corresponding signaling vector includes: converting each behavior type in the maintenance data of each device into a corresponding behavior vector; and weighting and fusing the behavior vectors according to a time window to generate a new behavior vector as a signaling vector.

[0094] Step 102: The signaling vector is fused with the device's context information vector to generate a behavior map that does not have semantic recognition, and a set of behavior maps for all devices is formed.

[0095] Specifically, fusing the signaling vector with the device's context information vector includes: fusing the signaling vector of each behavior type in the maintenance data of each device with the corresponding category's context information vector. The signaling vector and the context information vector are added together and then input into the tanh function to complete the fusion.

[0096] Specifically, generating behavior maps that lack semantic recognition capabilities includes: using the result of fusing the signaling vector with the device's context information vector as the behavior map for each behavior type in the maintenance data of each device.

[0097] Step 103: Perform structural coordination judgment on the behavior mappings in the behavior mapping set, find conjugate mapping pairs, perform lifecycle control on each conjugate mapping pair, and destroy conjugate mapping pairs whose call count exceeds the maximum call count as invalid conjugate mapping pairs.

[0098] Specifically, it also includes: collecting each behavior type of each device once every time window, converting each behavior type of the current device collected in each time window into a signal vector of the current device in each time window, and finally converting the signal vector of the current device in each time window into behavior images of multiple time windows.

[0099] Specifically, before performing structural synergy judgment on the behavior mappings in the behavior mapping set, the process includes: combining the behavior mappings of multiple time windows of the current device into a one-dimensional vector sequence, dividing the one-dimensional vector sequence into multiple sub-one-dimensional vector sequences according to a preset window length, performing a Fourier transform on each sub-one-dimensional vector sequence, and generating the spectral tensor of each sub-one-dimensional vector sequence.

[0100] Specifically, the structural synergy judgment of behavioral mappings in the behavioral mapping set and the identification of conjugate mapping pairs include: unifying the size of each spectral tensor and performing a normalization operation to generate a new spectral tensor; identifying pairs of new spectral tensors between different devices whose similarity exceeds a preset threshold as conjugate mapping pairs; clustering the conjugate mapping pairs according to the labeled device fault categories; and having users take corresponding device maintenance measures based on the device fault categories.

[0101] Specifically, lifecycle control for each conjugate pair includes: starting from the establishment time of each conjugate pair, if the current time exceeds a preset survival time threshold, the corresponding conjugate pair is marked as expired.

[0102] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0103] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0104] In the several embodiments provided by this invention, it should be understood that the disclosed technical content can be implemented in other ways. The system embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between units or modules, and may be electrical or other forms.

[0105] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0106] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0107] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or 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: USB flash drives, read-only storage media (ROM), random access storage media (RAM), portable hard drives, magnetic disks, optical disks, and other media capable of storing program code.

[0108] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for data privacy computation in the Industrial Internet of Things (IIoT) for equipment maintenance, characterized in that, include: Acquire maintenance data for all devices and convert the maintenance data for each device into a corresponding signal vector; The signaling vector is fused with the device's context information vector to generate a behavior map that lacks semantic recognition, and a set of behavior maps for all devices is formed. Structural synergy is assessed for behavior mappings in the behavior mapping set to identify conjugate mapping pairs. Lifecycle control is applied to each conjugate mapping pair, and conjugate mapping pairs whose call count exceeds the maximum call count are identified as invalid conjugate mapping pairs and destroyed.

2. The industrial IoT data privacy computation method for equipment maintenance as described in claim 1, characterized in that, Transforming the maintenance data of each device into a corresponding signaling vector includes: converting each behavior type in the maintenance data of each device into a corresponding behavior vector; and weighting and fusing the behavior vectors according to a time window to generate a new behavior vector as a signaling vector.

3. The industrial IoT data privacy computation method for equipment maintenance as described in claim 2, characterized in that, The fusion of signaling vectors with device context information vectors includes: fusing the signaling vector of each behavior type in the maintenance data of each device with the corresponding category of context information vector. The signaling vector and context information vector are added together and then input into the tanh function to complete the fusion.

4. The industrial IoT data privacy computation method for equipment maintenance as described in claim 1, characterized in that, Generating behavior maps that lack semantic recognition involves using the result of fusing the signaling vector with the device's context information vector as the behavior map for each behavior type in the maintenance data of each device.

5. The industrial IoT data privacy computation method for equipment maintenance as described in claim 2, characterized in that, Also includes: Each time window, each behavior type of each device is collected once. Each behavior type of the current device collected in each time window is transformed into a signal vector of the current device in each time window. Finally, the signal vector of the current device in each time window is transformed into behavior images of multiple time windows.

6. The industrial IoT data privacy computation method for equipment maintenance as described in claim 5, characterized in that, Before performing structural synergy judgment on the behavior mappings in the behavior mapping set, the process also includes: combining the behavior mappings of multiple time windows of the current device into a one-dimensional vector sequence, dividing the one-dimensional vector sequence into multiple sub-one-dimensional vector sequences according to a preset window length, performing a Fourier transform on each sub-one-dimensional vector sequence, and generating the spectral tensor of each sub-one-dimensional vector sequence.

7. The industrial IoT data privacy computation method for equipment maintenance as described in claim 6, characterized in that, The structural synergy assessment of behavioral mappings in the behavioral mapping set and the identification of conjugate mapping pairs include: unifying the size of each spectral tensor and performing a normalization operation to generate a new spectral tensor; identifying pairs of new spectral tensors between different devices whose similarity exceeds a preset threshold as conjugate mapping pairs; clustering the conjugate mapping pairs according to the labeled device fault categories; and having users take corresponding device maintenance measures based on the device fault categories.

8. The industrial IoT data privacy computation method for equipment maintenance as described in claim 1, characterized in that, Lifecycle control for each conjugate pair includes: starting from the establishment time of each conjugate pair, if the current time exceeds a preset survival time threshold, the corresponding conjugate pair is marked as expired.

9. An industrial IoT data privacy computing system for equipment maintenance, characterized in that, include: The signal vector generation module is used to acquire maintenance data for all devices and convert the maintenance data for each device into a corresponding signal vector. The behavior mapping generation module is used to fuse the signaling vector with the device's context information vector to generate a behavior mapping that does not have semantic recognition, and form a set of behavior mappings for all devices. The module for generating conjugate mapping pairs is used to determine the structural synergy of behavioral mapping pairs in the behavioral mapping set, identify conjugate mapping pairs, control the lifecycle of each conjugate mapping pair, and destroy conjugate mapping pairs that exceed the maximum number of calls as invalid conjugate mapping pairs.

10. The industrial IoT data privacy computing system for equipment maintenance as described in claim 9, characterized in that, Transforming the maintenance data of each device into a corresponding signaling vector includes: converting each behavior type in the maintenance data of each device into a corresponding behavior vector; and weighting and fusing the behavior vectors according to a time window to generate a new behavior vector as a signaling vector.

Images