Data security multi-party cooperative protection method and device in power internet of things environment

Abstract

The application discloses a kind of data security multi-party collaborative protection method and device under the environment of electric power internet of things, belong to electric power internet of things and data security technical field.The method comprises: the multidimensional state data of node is collected in real time;Security trust threshold value and resource demand parameter corresponding to security collaborative task request are parsed, and multiple participating nodes that meet the condition are selected from multiple collaborative computing nodes;The real-time process computing difficulty characteristic value of multiple participating nodes is obtained, and the inter-computer difference between any two participating nodes is calculated based on real-time process computing difficulty characteristic value;According to the inter-computer difference and the preset dynamic clustering threshold value, all participating nodes are dynamically divided into multiple non-overlapping computer sets, and each computer set is controlled to execute the differentiated security multi-party computing strategy matched with the security collaborative task request.The application can improve the execution efficiency and response speed of multi-party collaborative computing on the basis of ensuring the privacy and security of electric power data.

Description

Technical Field

[0001] This invention relates to the field of power Internet of Things (IoT) and data security technology, and in particular to a multi-party collaborative protection method and device for data security in a power IoT environment. Background Technology

[0002] With the construction of new power systems, the number of sensing terminals and edge devices in the power Internet of Things (IoT) is growing explosively, and massive amounts of power data need to be processed collaboratively in the cloud or at the edge. Since power data involves user privacy and grid operation security, ensuring secure computation while keeping data available but not visible has become a critical issue that urgently needs to be addressed in multi-party collaborative computing scenarios. Secure multi-party computation technology, as an effective privacy protection method, is widely used to solve the problem of collaboratively computing agreed-upon functions without disclosing each other's input data.

[0003] Existing data security collaboration solutions for the power Internet of Things (IoT) typically focus on static trust assessments or node selection based on fixed rules. Specifically, nodes participating in computation are selected based on their historical reputation scores, hardware configuration parameters, or fixed network topology. After selecting nodes, the system usually distributes the computation tasks to each node using random allocation, round-robin allocation, or simple load balancing strategies, thereby executing the corresponding encrypted computation protocols.

[0004] However, related technologies mainly rely on the static attributes or historical data of nodes for task scheduling, failing to fully consider the dynamic fluctuations in the actual performance of nodes when processing specific computing tasks. This static or coarse-grained scheduling method can easily lead to inconsistent processing paces among nodes during collaborative computing, resulting in low overall computing efficiency and failing to meet the real-time data processing requirements of power services. Summary of the Invention

[0005] This invention provides a multi-party collaborative protection method and device for data security in a power Internet of Things (IoT) environment, which can solve the problem of low computational efficiency in related technologies. The technical solution is as follows: On the one hand, a multi-party collaborative protection method for data security in a power Internet of Things environment is provided, the method comprising: Real-time acquisition of multi-dimensional status data from multiple collaborative computing nodes in the power Internet of Things; The system receives a security collaboration task request, parses the security trust threshold and resource requirement parameters corresponding to the security collaboration task request, and selects multiple participating nodes that meet the security trust threshold and resource requirement parameters from the multiple collaborative computing nodes based on the multi-dimensional state data. Obtain the real-time process computation difficulty feature value of the multiple participating nodes in response to the security collaboration task request, and calculate the inter-computer difference degree between any two participating nodes based on the real-time process computation difficulty feature value. Based on the differences between the computers and a preset dynamic clustering threshold, all participating nodes are dynamically divided into multiple non-overlapping computer sets, and each computer set is controlled to execute a differentiated secure multi-party computation strategy that matches the secure collaborative task request.

[0006] On the other hand, a data security multi-party collaborative protection device for a power Internet of Things environment is provided, the device comprising: The data acquisition unit is used to collect multi-dimensional status data of multiple collaborative computing nodes in the power Internet of Things in real time. The parsing and filtering unit is used to receive a security collaboration task request, parse the security trust threshold and resource requirement parameters corresponding to the security collaboration task request, and filter out multiple participating nodes that meet the security trust threshold and resource requirement parameters from the multiple collaborative computing nodes based on the multi-dimensional state data. The computing unit is used to acquire the real-time process computing difficulty feature value of the multiple participating nodes in response to the security collaboration task request, and to calculate the inter-computer difference between any two participating nodes based on the real-time process computing difficulty feature value. The collaborative computing unit is used to dynamically divide all participating nodes into multiple non-overlapping computer sets based on the differences between the computers and a preset dynamic clustering threshold, and to control each computer set to execute a differentiated secure multi-party computing strategy that matches the secure collaborative task request.

[0007] On the other hand, a computer device is provided, the computer device including a memory and a processor, the memory for storing computer programs, and the processor for executing the computer programs stored in the memory to implement the steps of the data security multi-party collaborative protection method in the power Internet of Things environment described above.

[0008] On the other hand, a computer-readable storage medium is provided, wherein a computer program is stored therein, and when the computer program is executed by a processor, the steps of the above-described multi-party collaborative protection method for data security in the power Internet of Things environment are implemented.

[0009] On the other hand, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the above-described multi-party collaborative protection method for data security in the power Internet of Things environment.

[0010] The technical solution provided by this invention can bring at least the following beneficial effects: In this embodiment of the invention, by collecting multi-dimensional state data of collaborative computing nodes and selecting multiple participating nodes from a pool of collaborative computing nodes based on the security trust threshold and resource requirement parameters of the secure collaborative task request, a dynamic indicator, the real-time process computation difficulty feature value, is introduced to quantify the actual computational pressure of nodes when executing specific tasks. By calculating the inter-computer differences among participating nodes and dynamically dividing them into non-overlapping computer sets according to a dynamic clustering threshold, nodes with similar computing capabilities and load states are aggregated into the same set, effectively avoiding collaborative computing blocking problems caused by excessive differences in node performance and significantly reducing synchronization latency within the set. This solution, by controlling each computer set to execute differentiated secure multi-party computation strategies that match the secure collaborative task request, can improve the execution efficiency and response speed of multi-party collaborative computing while ensuring the privacy and security of power data. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a flowchart of a multi-party collaborative protection method for data security in a power Internet of Things environment, provided by an embodiment of the present invention; Figure 2 This is a structural diagram of a data security multi-party collaborative protection device in a power Internet of Things environment provided by an embodiment of the present invention; Figure 3 This is a hardware architecture diagram of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0014] Please refer to Figure 1 This invention provides a multi-party collaborative protection method for data security in a power Internet of Things (IoT) environment, the method comprising: Step 100: Real-time acquisition of multi-dimensional status data from multiple collaborative computing nodes in the power Internet of Things; Step 102: Receive a security collaboration task request, parse the security trust threshold and resource requirement parameters corresponding to the security collaboration task request, and based on the multi-dimensional state data, select multiple participating nodes from the multiple collaborative computing nodes that meet the security trust threshold and resource requirement parameters; Step 104: Obtain the real-time process computation difficulty feature value of the multiple participating nodes in response to the security collaboration task request, and calculate the inter-computer difference between any two participating nodes based on the real-time process computation difficulty feature value. Step 106: Based on the difference between the computers and the preset dynamic clustering threshold, all participating nodes are dynamically divided into multiple non-overlapping computer sets, and each computer set is controlled to execute a differentiated secure multi-party computation strategy that matches the secure collaborative task request.

[0015] In this embodiment of the invention, by collecting multi-dimensional state data of collaborative computing nodes and selecting multiple participating nodes from a pool of collaborative computing nodes based on the security trust threshold and resource requirement parameters of the secure collaborative task request, a dynamic indicator, the real-time process computation difficulty feature value, is introduced to quantify the actual computational pressure of nodes when executing specific tasks. By calculating the inter-computer differences among participating nodes and dynamically dividing them into non-overlapping computer sets according to a dynamic clustering threshold, nodes with similar computing capabilities and load states are aggregated into the same set, effectively avoiding collaborative computing blocking problems caused by excessive differences in node performance and significantly reducing synchronization latency within the set. This solution, by controlling each computer set to execute differentiated secure multi-party computation strategies that match the secure collaborative task request, can improve the execution efficiency and response speed of multi-party collaborative computing while ensuring the privacy and security of power data.

[0016] The following description Figure 1 The execution method for each step is shown.

[0017] First, for step 100, multi-dimensional status data of multiple collaborative computing nodes in the power Internet of Things are collected in real time.

[0018] In this embodiment of the invention, the collaborative scheduling center can issue status collection instructions to all edge nodes at fixed intervals through a lightweight message queue. The collection agent on each node reads multi-dimensional status data and reports it to the collaborative scheduling center.

[0019] In one embodiment of the present invention, the multidimensional state data includes at least: security trust dimension data and computing resource dimension data; in another embodiment of the present invention, the multidimensional state data, in addition to security trust dimension data and computing resource dimension data, may also include: communication load dimension data and task adaptation dimension data. Wherein: The security trust dimension data includes at least one of the following: blockchain-based node trust value, dynamic trust score based on zero trust framework, historical abnormal behavior record score, and device trustworthiness score. The computing resource dimension data includes at least one of the following: CPU utilization, RAM usage, and HDD usage; The communication load dimension data includes at least one of the following: the current communication link bandwidth utilization of the node, the length of the queue of pending collaborative tasks, and the historical collaborative data interaction latency. The task adaptation dimension data includes: the success rate of the node's historical execution of similar security tasks, the types and strengths of encryption resources that the node can provide, and at least one of the security protocols supported by the node.

[0020] This invention collects multi-dimensional state data of nodes rather than single-dimensional data because single-dimensional data, such as relying solely on CPU utilization or trust scores, cannot fully reflect the true collaborative capabilities of nodes and can easily lead to biased decision-making. For example, a malicious node with sufficient computing power but a low trust score, if selected into a collaborative group, may steal sensitive intermediate results or output incorrect calculation results; conversely, a high-trust node that is in a state of communication congestion or computational overload will cause the entire collaborative task to time out and fail while waiting for that node. By collecting and integrating data from multiple dimensions, the system can simultaneously evaluate whether a node is trustworthy, capable, able to communicate in a timely manner, and proficient in this type of task. This allows for the selection of truly optimal participating nodes while ensuring a minimum level of security, thereby improving the success rate and execution efficiency of collaborative tasks.

[0021] Then, for step 102, a security collaboration task request is received, the security trust threshold and resource requirement parameters corresponding to the security collaboration task request are parsed, and based on the multi-dimensional state data, multiple participating nodes that meet the security trust threshold and resource requirement parameters are selected from the multiple collaborative computing nodes.

[0022] In this embodiment of the invention, the secure collaborative task may include privacy computation and federated learning tasks, threat detection and anomaly identification tasks, or encrypted resource sharing and key management tasks. Different secure collaborative tasks have different security trust thresholds and resource requirement parameters.

[0023] In one implementation, the security trust threshold is obtained by a weighted fusion of the dynamic trust score under the zero-trust architecture and the node trust value maintained by the blockchain. The fusion formula can be: Security trust threshold = α × zero trust score + (1-α) × blockchain trust value The weight α can be dynamically set according to the task's security level. For example, it can be 0.7 for high-level tasks, 0.5 for medium-level tasks, and 0.3 for low-level tasks.

[0024] Resource requirements parameters may include: minimum CPU guarantee rate, minimum memory free time, maximum allowable network latency, and must support homomorphic encryption, etc.

[0025] The scheduling center calculates the security trust dimension value of each collaborative computing node based on the multi-dimensional status data reported by each collaborative computing node, and removes nodes whose security trust dimension value is less than the security trust threshold. The security trust dimension value is obtained by weighted fusion of at least one of the following: blockchain-based node trust value, dynamic trust score based on zero trust framework, historical abnormal behavior record score, and device trustworthiness score.

[0026] Collaborative computing nodes whose security trust dimension value is greater than or equal to the security trust threshold and whose computational resource dimension data meets the resource requirement parameters will be selected as participating nodes.

[0027] Then, the following explanations are given regarding step 104, "obtaining the real-time process computation difficulty feature values ​​of the multiple participating nodes for the security collaboration task request, and calculating the inter-computer difference degree between any two participating nodes based on the real-time process computation difficulty feature values" and step 106, "dynamically dividing all participating nodes into multiple non-overlapping computer sets according to the inter-computer difference degree and a preset dynamic clustering threshold, and controlling each computer set to execute a differentiated security multi-party computation strategy that matches the security collaboration task request".

[0028] In this embodiment of the invention, after the participating nodes are selected, the scheduling center needs to quantify the computational difficulty of each node in handling this security collaboration task, and then determine which participating nodes can be combined to execute this security collaboration task. The real-time process computational difficulty feature value is used to characterize the expected difficulty of participating nodes in the power Internet of Things (IoT) in handling a specific type of security multi-party computation task at the current moment. This value reflects the node's computing resource availability, current load, communication latency, and task adaptability. A larger value indicates that the node has a greater difficulty in executing this type of task, such as longer execution time and a higher likelihood of overload.

[0029] Based on this, the calculation methods for the real-time process computation difficulty feature value can include: The same probe task is sent to each participating node, enabling each node to complete the secure multi-party computation ciphertext operation indicated by the probe task. Key metrics for each participating node's completion of the probe task are measured and recorded. Based on the measured values ​​of these key metrics, a real-time process computation difficulty characteristic value is calculated. These key metrics include at least the task time.

[0030] For example, the detection task could be: requiring each participating node to independently execute a typical MPC operation, generate a Beaver multiplication triple and complete a ciphertext multiplication.

[0031] In one implementation, this key metric can be task duration, CPU utilization, memory utilization, and task queue length. Therefore, the formula for calculating the difficulty metric of real-time process computation can be: Di = w1 × T + w2 × C1 + w3 × M1 + w4 × Q1 Where T is the normalized value of task time; C1 is the normalized value of CPU utilization; M1 is the normalized value of memory utilization; Q1 is the normalized value of task queue length; and w1, w2, w3, and w4 are weight parameters.

[0032] Furthermore, since the probe task is to measure the computational difficulty of nodes when processing secure collaborative tasks, the same probe task sent to each participating node is determined based on the type of secure collaborative task, and the probe task is generated based on the core computational characteristics corresponding to the secure collaborative task. By pre-generating and storing the probe task corresponding to each type, the probe task is determined as follows when it is necessary to calculate the computational difficulty characteristic value of the real-time process: Parse the secure collaboration task request and identify the task type label; the task type label includes privacy computing and federated learning tasks, threat detection and anomaly identification tasks, or encrypted resource sharing and key management tasks. Based on the task type label, select the corresponding type of probe task from the preset probe task mapping library.

[0033] Among them, the core computational features of privacy computing and federated learning tasks are homomorphic encrypted ciphertext multiplication, Beaver triple generation, and secret sharing. The detection tasks include Beaver triple generation and ciphertext multiplication simulation. Key indicators include: ciphertext multiplication time, number of polynomial operations, and memory bandwidth. The core computational features of threat detection and anomaly identification tasks are traffic feature extraction, rule matching, and lightweight inference. The detection task includes standardized pcap file parsing and Snort rule matching. Key metrics include: feature extraction time, rule matching speed, and peak memory allocation. The core computational features of encrypted resource sharing and key management tasks are key negotiation, key derivation, and group key distribution. Therefore, the probing task includes the entire process of SM2 / ECDH key negotiation, and key indicators include: time consumption of dot product operation, quality of random number generation, and round-trip interaction latency.

[0034] In this way, the calculated real-time process computation difficulty feature value can truly reflect the node's ability to perform specific task types, further improving the accuracy of subsequent node combinations, increasing computational efficiency, and avoiding misjudgments of capabilities due to mismatch between the probe task and the actual task.

[0035] After calculating the real-time process computation difficulty feature value of each participating node, the inter-computer difference between any two participating nodes can be calculated based on the real-time process computation difficulty feature value: for the current two participating nodes, calculate the absolute value of the difference between the real-time process computation difficulty feature values ​​of the two participating nodes, and use the absolute value of the difference as the inter-computer difference between the two participating nodes. The step of dynamically dividing all participating nodes into multiple non-overlapping computer sets based on the inter-computer difference degree and a preset dynamic clustering threshold includes: traversing all participating nodes, merging two participating nodes whose inter-computer difference degree is not greater than the dynamic clustering threshold into the same initial computer set, isolating two participating nodes whose inter-computer difference degree is greater than the dynamic clustering threshold into different initial computer sets, and performing merging and deduplication processing on the merged initial computer sets to obtain multiple computer sets in which the inter-computer difference degree of internal nodes is less than the dynamic clustering threshold and they do not overlap.

[0036] The dynamic partitioning method ensures that the inter-computer difference between any two nodes in a computer set is less than the dynamic clustering threshold, and that multiple computer sets do not overlap.

[0037] During the dynamic partitioning process, an empty set list can be initialized first. Then, for each unvisited participating node, a new set is created. Next, all nodes whose difference from the participating node is less than the dynamic clustering threshold are found and added to the new set. For each newly added node in the new set, the search continues to expand to find other nodes whose difference is less than the dynamic clustering threshold, until no further expansion is possible. This new set is recorded as an initial computer set. The above process is repeated until all nodes have been visited.

[0038] In one embodiment of the present invention, the dynamic clustering threshold can be dynamically generated based on the overall computing power distribution of the currently participating nodes. Specifically, the system calculates the mean and standard deviation of the real-time process computation difficulty feature values ​​of all participating nodes in real time, and uses a preset multiple of the standard deviation as the current dynamic clustering threshold. This ensures that in the power Internet of Things environment, when the overall load of the nodes drifts, the clustering criteria can be adaptively adjusted accordingly, avoiding clustering failure caused by a fixed threshold under extreme operating conditions.

[0039] It should be noted that although the inter-computer difference between any two nodes in each computer set is less than the dynamic clustering threshold, the mean value of the real-time process computation difficulty feature value of the nodes in each computer set is different.

[0040] In this embodiment of the invention, after obtaining a set of non-overlapping computers, each set of computers can be controlled to execute a differentiated secure multi-party computation strategy that matches the secure collaborative task request, specifically including: Calculate the key group prediction separation degree among different computer sets; the separation degree characterizes the heterogeneity of different sets in terms of computing power and communication latency; Based on the comparison results between the separation degree and the preset separation degree threshold, different security levels of key negotiation cycles and key update strategies are configured for different computer sets, so that sets with high separation degree adopt a strongly isolated independent key system, while sets with low separation degree adopt a lightweight group shared key system.

[0041] In calculating the key group prediction separation degree between different computer sets, in this embodiment of the invention, the mean value of the real-time process computation difficulty feature value of the nodes in each computer set can be calculated first, and then the absolute difference between the means of the two computer sets can be used as the separation degree between the two computer sets.

[0042] For example, if the mean of set G1 is 0.39, the mean of set G2 is 0.58, and the mean of set G3 is 0.48, then the separation degrees are respectively: Sep(G1,G2) = |0.39-0.58| = 0.19 Sep(G1,G3) = |0.39-0.48| = 0.09 Sep(G2,G3) = |0.58-0.48| = 0.10 If the separation threshold is 0.1, then G1 and G2 are judged to have high separation, while G1 and G3, as well as G2 and G3, are judged to have low separation.

[0043] In multi-party collaborative computing within the power Internet of Things (IoT), there may be significant differences in computing power and communication latency among different computer sets, i.e., high separation. If a uniform key management strategy is applied to all computer sets, either frequent key negotiations will overwhelm low-performance nodes, or security vulnerabilities will be left on high-performance nodes. Therefore, in this embodiment of the invention, a hierarchical strategy is implemented based on the comparison result between the separation degree and a preset separation degree threshold: For different computer sets with a separation degree higher than the preset separation degree threshold, a strongly isolated independent key system is adopted. Taking G1 and G2 with high separation degree as an example, when a node of G1 needs to send an aggregation gradient to a node of G2, it first needs to prove its identity is legitimate and its current trust score is not less than the security trust threshold. Then, an SM2 public-private key encryption channel is established, and a new temporary key is used for each cross-set communication.

[0044] For different computer sets with a separation degree not exceeding the preset separation degree threshold, since the inter-computer difference between any two nodes within the computer set is not greater than the dynamic clustering threshold, a strongly isolated independent key system is not mandatory; a lightweight group shared key system can be used within the computer set. Specifically, a slow rotation strategy can be adopted, and Beaver triples can be generated and distributed in batches during the preprocessing stage of secure multi-party computation, allowing participating nodes within the set to reuse the same group session key across multiple computation cycles. For example, within set G3, the real-time process computation difficulty feature value of all nodes is between 0.44 and 0.52, and the inter-computer difference is not greater than the dynamic clustering threshold. The system allocates a group shared key to G3 with a validity period of 30 minutes, and generates 100 Beaver triples in batches during the preprocessing stage and distributes them to each node within the group, so that the subsequent 100 multiplication operations do not require renegotiation of the key.

[0045] In secure multi-party computation and federated learning, linear and nonlinear subtasks have drastically different computational complexities and communication patterns. For example, addition and scalar multiplication are linear subtasks, while ciphertext multiplication, comparison, and activation functions are nonlinear subtasks. Linear operations require simple additions or scalar multiplications between locally shared secret fragments, with low computational cost and no interaction required; while nonlinear operations require complex cryptographic protocols, such as using Beaver multiplication triples for ciphertext multiplication or employing obfuscated circuits for comparison, which not only incurs high computational overhead but also requires multiple rounds of interaction between nodes. If both types of operations are mixed and distributed among all nodes, nodes with poor computational capabilities will become bottlenecks in nonlinear operations, slowing down the entire task.

[0046] Based on this, in one embodiment of the present invention, when controlling each of the computer sets to execute a differentiated secure multi-party computation strategy, the method may further include: Identify the linear and nonlinear operation subtasks contained in the secure collaborative task request; The nonlinear operation subtasks are assigned to a designated set of computers for execution; the linear operation subtasks are distributed to all sets of computers for parallel execution; the designated set of computers is a set of computers whose average real-time process computation difficulty feature value is lower than a preset difficulty benchmark value.

[0047] This invention employs a task identification and separation mechanism to allocate nonlinear operations to a set of computers with lower computational difficulty characteristics (i.e., those with surplus computing power) for real-time processes, while distributing linear operations to all sets for parallel execution. This avoids timeouts or failures caused by low-performance nodes undertaking heavy nonlinear operations, while fully utilizing the computing power of high-performance nodes to accelerate complex calculations. Furthermore, the parallel processing of linear operations further compresses the overall completion time, achieving refined scheduling and maximizing efficiency of computing resources.

[0048] Furthermore, during task execution, based on a preset monitoring period, the fluctuation of the real-time process computation difficulty feature value of each participating node can be monitored. When the fluctuation of the target participating node is detected to cause the difference between it and other nodes in the current computer set to exceed the dynamic clustering threshold, local key renegotiation is triggered, and the target participating node is migrated from the current computer set to the target computer set with matching difference, while the remaining linear operation subtasks are redistributed.

[0049] Furthermore, in practical applications, a trust update mechanism can be introduced, namely: recording and feeding back the behavior of each participating node, the behavior including at least one of calculation accuracy, response time, and protocol regressivity; using the behavior of each participating node to update the zero-trust dynamic score and blockchain trust value of the participating node, so as to use the updated zero-trust dynamic score and blockchain trust value to calculate the security trust threshold for the next round of secure collaborative tasks.

[0050] Please refer to Figure 2 This invention provides a data security multi-party collaborative protection device in a power Internet of Things environment, the device comprising: The acquisition unit 200 is used to acquire multi-dimensional status data of multiple collaborative computing nodes in the power Internet of Things in real time. The parsing and filtering unit 202 is used to receive a security collaboration task request, parse the security trust threshold and resource requirement parameters corresponding to the security collaboration task request, and filter out multiple participating nodes that meet the security trust threshold and resource requirement parameters from the multiple collaborative computing nodes based on the multi-dimensional state data. The acquisition calculation unit 204 is used to acquire the real-time process computation difficulty feature value of the multiple participating nodes in response to the security collaboration task request, and to calculate the inter-computer difference between any two participating nodes based on the real-time process computation difficulty feature value. The collaborative computing unit 206 is used to dynamically divide all participating nodes into multiple non-overlapping computer sets according to the differences between the computers and a preset dynamic clustering threshold, and control each computer set to execute a differentiated secure multi-party computing strategy that matches the secure collaborative task request.

[0051] In one embodiment of the present invention, the multidimensional state data includes at least: security trust dimension data and computing resource dimension data; wherein, The security trust dimension data includes at least one of the following: blockchain-based node trust value, dynamic trust score based on zero trust framework, historical abnormal behavior record score, and device trustworthiness score. The computing resource dimension data includes at least one of the following: CPU utilization, RAM usage, and HDD usage.

[0052] In one embodiment of the present invention, the step of calculating the inter-computer difference between any two participating nodes based on the real-time process computing difficulty feature value includes: for the current two participating nodes, calculating the absolute value of the difference between the real-time process computing difficulty feature values ​​of the two participating nodes, and using the absolute value of the difference as the inter-computer difference between the two participating nodes; The step of dynamically dividing all participating nodes into multiple non-overlapping computer sets based on the inter-computer difference degree and a preset dynamic clustering threshold includes: traversing all participating nodes, merging two participating nodes whose inter-computer difference degree is not greater than the dynamic clustering threshold into the same initial computer set, isolating two participating nodes whose inter-computer difference degree is greater than the dynamic clustering threshold into different initial computer sets, and performing merging and deduplication processing on the merged initial computer sets to obtain multiple computer sets in which the inter-computer difference degree of internal nodes is less than the dynamic clustering threshold and they do not overlap.

[0053] In one embodiment of the present invention, controlling each of the computer sets to execute a differentiated secure multi-party computation strategy that matches the secure collaborative task request includes: Calculate the key group prediction separation degree among different computer sets; the separation degree is used to characterize the heterogeneity of different computer sets in terms of computing power and communication latency; Based on the comparison results between the separation degree and the preset separation degree threshold, different security levels of key negotiation cycles and key update strategies are configured for different computer sets.

[0054] In one embodiment of the present invention, configuring key negotiation periods and key update strategies with different security levels for different computer sets includes: For different sets of computers with a separation degree higher than the preset separation degree threshold, a strongly isolated independent key system is adopted.

[0055] In one embodiment of the present invention, controlling each of the computer sets to execute a differentiated secure multi-party computation strategy includes: Identify the linear and nonlinear operation subtasks contained in the secure collaborative task request; The nonlinear operation subtasks are assigned to a designated set of computers for execution; the linear operation subtasks are distributed to all sets of computers for parallel execution; the designated set of computers is a set of computers whose average real-time process computation difficulty feature value is lower than a preset difficulty benchmark value.

[0056] It should be noted that the data security multi-party collaborative protection device in the power Internet of Things environment provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the data security multi-party collaborative protection device in the power Internet of Things environment provided in the above embodiments and the data security multi-party collaborative protection method embodiments in the power Internet of Things environment belong to the same concept. The specific implementation process is detailed in the method embodiments and will not be repeated here.

[0057] Embodiments of this application also provide a computer device, please refer to... Figure 3 The computer device includes a processor and a memory, the memory storing at least one instruction, at least one program, code set or instruction set, the at least one instruction, at least one program, code set or instruction set being loaded and executed by the processor to implement the data security multi-party collaborative protection method in the power Internet of Things environment provided in the above method embodiments.

[0058] The embodiments of this application also provide a computer-readable storage medium storing at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement the multi-party collaborative protection method for data security in the power Internet of Things environment provided in the above-described method embodiments.

[0059] Embodiments of this application also provide a computer program product, which includes a computer program. A processor of a computer device reads the computer program from a computer-readable storage medium and executes the computer program, causing the computer device to perform the data security multi-party collaborative protection method in the power Internet of Things environment described in any of the above embodiments.

[0060] For ease of description, the above systems or devices are described separately as various modules or units based on their functions. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware components.

[0061] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0062] Finally, it should be noted that in this document, relational terms such as first, second, third, and fourth are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0063] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A multi-party collaborative protection method for data security in a power Internet of Things environment, characterized in that, The method includes: Real-time acquisition of multi-dimensional status data from multiple collaborative computing nodes in the power Internet of Things; The system receives a security collaboration task request, parses the security trust threshold and resource requirement parameters corresponding to the security collaboration task request, and selects multiple participating nodes that meet the security trust threshold and resource requirement parameters from the multiple collaborative computing nodes based on the multi-dimensional state data. Obtain the real-time process computation difficulty feature value of the multiple participating nodes in response to the security collaboration task request, and calculate the inter-computer difference degree between any two participating nodes based on the real-time process computation difficulty feature value. Based on the differences between the computers and a preset dynamic clustering threshold, all participating nodes are dynamically divided into multiple non-overlapping computer sets, and each computer set is controlled to execute a differentiated secure multi-party computation strategy that matches the secure collaborative task request.

2. The method according to claim 1, characterized in that, The multidimensional state data includes at least: security trust dimension data and computing resource dimension data; wherein... The security trust dimension data includes at least one of the following: blockchain-based node trust value, dynamic trust score based on zero trust framework, historical abnormal behavior record score, and device trustworthiness score. The computing resource dimension data includes at least one of the following: CPU utilization, RAM usage, and HDD usage.

3. The method according to claim 1, characterized in that, The step of calculating the inter-computer difference between any two participating nodes based on the real-time process computation difficulty feature value includes: for the current two participating nodes, calculating the absolute value of the difference between the real-time process computation difficulty feature values ​​of the two participating nodes, and using the absolute value of the difference as the inter-computer difference between the two participating nodes; The step of dynamically dividing all participating nodes into multiple non-overlapping computer sets based on the inter-computer difference degree and a preset dynamic clustering threshold includes: traversing all participating nodes, merging two participating nodes whose inter-computer difference degree is not greater than the dynamic clustering threshold into the same initial computer set, isolating two participating nodes whose inter-computer difference degree is greater than the dynamic clustering threshold into different initial computer sets, and performing merging and deduplication processing on the merged initial computer sets to obtain multiple computer sets in which the inter-computer difference degree of internal nodes is less than the dynamic clustering threshold and they do not overlap.

4. The method according to claim 3, characterized in that, Controlling each of the aforementioned computer sets to execute a differentiated secure multi-party computation strategy that matches the secure collaborative task request includes: Calculate the key group prediction separation degree among different computer sets; the separation degree is used to characterize the heterogeneity of different computer sets in terms of computing power and communication latency; Based on the comparison result between the separation degree and the preset separation degree threshold, different security levels of key negotiation cycles and key update strategies are configured for different computer sets.

5. The method according to claim 4, characterized in that, The configuration of different security levels for key negotiation cycles and key update strategies for different computer sets includes: For different sets of computers with a separation degree higher than the preset separation degree threshold, a strongly isolated independent key system is adopted.

6. The method according to any one of claims 1-5, characterized in that, The control of each of the computer sets to execute a differentiated secure multi-party computation strategy includes: Identify the linear and nonlinear operation subtasks contained in the secure collaborative task request; The nonlinear operation subtasks are assigned to a designated set of computers for execution; the linear operation subtasks are distributed to all sets of computers for parallel execution; the designated set of computers is a set of computers whose average real-time process computation difficulty feature value is lower than a preset difficulty benchmark value.

7. A data security multi-party collaborative protection device in a power Internet of Things environment, characterized in that, The device includes: The data acquisition unit is used to collect multi-dimensional status data from multiple collaborative computing nodes in the power Internet of Things in real time. The parsing and filtering unit is used to receive a security collaboration task request, parse the security trust threshold and resource requirement parameters corresponding to the security collaboration task request, and filter out multiple participating nodes that meet the security trust threshold and resource requirement parameters from the multiple collaborative computing nodes based on the multi-dimensional state data. The computing unit is used to acquire the real-time process computing difficulty feature value of the multiple participating nodes in response to the security collaboration task request, and to calculate the inter-computer difference between any two participating nodes based on the real-time process computing difficulty feature value. The collaborative computing unit is used to dynamically divide all participating nodes into multiple non-overlapping computer sets based on the differences between the computers and a preset dynamic clustering threshold, and to control each computer set to execute a differentiated secure multi-party computing strategy that matches the secure collaborative task request.

8. A computer device, characterized in that, The computer device includes a memory and a processor. The memory is used to store computer programs, and the processor is used to execute the computer programs stored in the memory to implement the steps of the method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the steps of the method described in any one of claims 1-6.

10. A computer program product, characterized in that, Includes a computer program, which, when executed by a processor, implements the steps of the method according to any one of claims 1-6.

Images