Method and system for automatically generating security policy of power internet of things based on business intention
Through the sentence state long and short memory network model, the business intention-strategy model is generated, and the problems of low efficiency and poor interpretability in the existing technology are solved, and automated and accurate security strategy generation is realized to ensure the strategy Consistent with business needs, it improves the security and adaptability of the power Internet of Things system.
Patent Information
- Application Number
- CN202411231337.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-04
- Publication Date
- 2024-11-12
AI Technical Summary
The existing power IoT security strategy generation method relies on manual analysis or machine learning, which is inefficient, error-prone and poorly interpretable, making it difficult to adapt to rapidly changing business needs and security threats, and the ambiguity of business intentions leads to the strategy and Inconsistent demand.
The sentence state long and short memory network model is used to extract the characteristics of power IoT business needs, convert it into business intention, and generate security policies through the business intention-strategy model to ensure that the policy is closely related to business needs, and use the SDN controller to issue and continuously verify security The effectiveness of the strategy.
The automated generation of security policies is realized, the accuracy and effectiveness of policies are improved, manual intervention and errors are reduced, the complete consistency between policies and business needs is ensured, and the security and adaptability of power IoT systems are improved.
Smart Images

Figure CN118944965A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of power Internet of Things security technology, and in particular to a method and system for automatically generating power Internet of Things security policies based on business intent. Background Art
[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.
[0003] With the rapid development of science and technology, the power Internet of Things, as the core application of Internet of Things technology in smart grids, is gradually becoming a key force in promoting the digital transformation of the power industry. By integrating advanced communications, information technology and power system infrastructure, the power Internet of Things has realized the interconnection and intelligent interaction of all aspects of the power system, such as power generation, transmission, transformation, distribution, and power consumption, greatly improving the efficiency, reliability and sustainability of the power system. However, with the widespread application of the power Internet of Things, its security issues have become increasingly prominent. How to efficiently and accurately formulate security strategies to ensure the safe and stable operation of the power Internet of Things has become an important research topic for those skilled in the art. At the same time, the inventors found that there are still some technical problems in the existing research on the generation of security strategies and attack defense of the power Internet of Things, such as: At present, the formulation and generation of power Internet of Things security policies are still mainly based on manual analysis and configuration. This method requires security experts to have a deep understanding of the power Internet of Things system and accurately judge security threats. It has high requirements for technical personnel, is inefficient, prone to errors, and is difficult to adapt to the rapidly changing business needs and security threats of the power Internet of Things. Alternatively, the security policy can be generated by analyzing and learning the historical data of the power Internet of Things system using machine learning algorithms, that is, dynamically adjusting and optimizing the security policy according to the learning results to improve the system's adaptability and security; however, the accuracy of this method depends on the quality and quantity of training data, and the generated security policy has poor interpretability. Another method is to automatically generate corresponding security policies based on business intent by combining the business needs and business logic of the power Internet of Things. However, in actual applications, how to accurately extract and represent business intent is still a challenge. The ambiguity or ambiguity of business intent can easily lead to the generated security policy being inconsistent with business needs. Summary of the invention
[0004] In order to overcome the shortcomings of the above-mentioned prior art, the present invention provides a method and system for automatically generating security policies for the power Internet of Things based on business intent, which can accurately issue corresponding security policies based on business intent and provide support for the safe operation of the power Internet of Things.
[0005] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions: The first aspect of the present invention provides a method for automatically generating security policies for the power Internet of Things based on business intent.
[0006] The automatic generation method of power Internet of Things security strategy based on business intent includes: Obtain business requirements for the power Internet of Things; The sentence state long short-term memory network model is used to extract features of the business requirements and convert the business requirements of the power Internet of Things into business intents; Decomposing the business intent, establishing a connection between the business intent and the device instruction, generating a business intent-policy model, and generating a security policy according to a mapping relationship between the business intent-policy model; The SDN controller obtains the security policy and sends it to relevant network devices; continuously verifies the effectiveness of the security policy, and generates a final security policy based on business intent through continuous verification.
[0007] Converting the business requirements of the power Internet of Things into business intents includes four processes: entity labeling, sentence vectorization, feature code extraction, and business intent conversion.
[0008] The statement vectorization includes: The word features, magnetic label features and position features in business needs are extracted through multiple dimensions to obtain three feature vectors, which are then fused into a complete embedding vector by superposition and fusion.
[0009] The feature code extraction includes: The embedding vector is used as the feature code extraction object, and the sentence state long short-term memory network model is used to iteratively calculate the embedding vector and obtain the feature encoding of the business demand. Specifically: each word in the business demand sentence is mapped to a first-level node, and the sentence state is mapped to a second-level node. In each iterative calculation, the feature state calculation method of the first-level node is jointly determined by the corresponding first-level node of the previous layer and its previous and subsequent first-level nodes and second-level nodes, and the feature state of the second-level node is jointly determined by all the first-level nodes and second-level nodes of the previous layer.
[0010] The business intent conversion includes: The feature code is used as the business intent conversion object, and the sentence state long short-term memory network model is used to decode the feature code to obtain the label vector of the business intent; then, the classifier model is used to perform entity prediction on the label vector to obtain the final required business intent.
[0011] Decompose the business intent, including: The business intent is split according to behavioral actions and operation objects. After the splitting, the relevant behavioral actions and ID attributes are matched based on the inherent attribute characteristics of the operation object, and the connection relationship between the business intent and the device instructions is established, thereby generating a business intent-policy model.
[0012] The effectiveness of the security strategy is verified by calculating the accuracy and recall of the security strategy.
[0013] The second aspect of the present invention provides a system for automatically generating security policies for the power Internet of Things based on business intent.
[0014] The automatic generation system of power IoT security policies based on business intent includes: The business demand acquisition module is configured to: acquire business demands of the power Internet of Things; The business intention conversion module is configured to: extract features of the business requirements using a sentence state long short-term memory network model, and convert the business requirements of the power Internet of Things into business intentions; The security policy generation module is configured to: decompose the business intent, establish a connection between the business intent and the device instruction, generate a business intent-policy model, and generate a security policy according to a mapping relationship between the business intent-policy model; The verification module is configured as follows: the SDN controller obtains the security policy and sends it to relevant network devices; continuously verifies the validity of the security policy, and generates a final security policy based on business intent through continuous verification.
[0015] The third aspect of the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps in the method for automatically generating security policies for the power Internet of Things based on business intent as described in the first aspect of the present invention.
[0016] The fourth aspect of the present invention provides an electronic device, comprising a memory, a processor, and a program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the method for automatically generating security policies for the power Internet of Things based on business intent as described in the first aspect of the present invention are implemented.
[0017] One or more of the above technical solutions have the following beneficial effects: (1) The present invention provides a method for automatically generating security policies for an electric power Internet of Things based on business intent. After obtaining the business needs of the electric power Internet of Things, a sentence state long short-term memory network model is used to extract features of the business needs, and the business needs of the electric power Internet of Things are converted into business intents. This method can achieve accurate extraction of business intent features and avoid the problem of inconsistency between security policies and business needs caused by the ambiguity or ambiguity of the business intent itself.
[0018] (2) The present invention decomposes the business intent, establishes a connection between the business intent and the device instructions, generates a business intent-policy model, and generates a security policy based on the mapping relationship between the business intent-policy model. Therefore, the present invention can ensure that the formulation of security policies is closely centered on business needs, avoiding the problem of the disconnection between policies and business needs in traditional technologies; in addition, by directly associating business intent and device instructions, the security policy is closer to the actual application scenario, improving the effectiveness and practicality of the policy.
[0019] (3) By generating a business intent-policy model, the present invention realizes the automatic generation of security policies, reduces human intervention, and improves the efficiency and accuracy of policy formulation; at the same time, the automation process also helps to reduce human errors and improve the overall quality of security policies.
[0020] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.
[0022] Figure 1 The present invention is a flowchart of a method for automatically generating a security strategy for the power Internet of Things based on business intent in the first embodiment of the present invention.
[0023] Figure 2 This is a module diagram of a system for automatically generating security policies for the power Internet of Things based on business intent in Embodiment 2 of the present invention. DETAILED DESCRIPTION
[0024] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.
[0025] It should be noted that the terms used herein are for describing specific embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.
[0026] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.
[0027] Embodiment 1 This embodiment discloses a method for automatically generating security policies for the power Internet of Things based on business intent.
[0028] like Figure 1 As shown, the automatic generation method of power Internet of Things security policy based on business intent includes: Step S1, obtaining the business requirements of the power Internet of Things; Step S2: extracting features of the business requirements using a sentence state long short-term memory network model, and converting the business requirements of the power Internet of Things into business intents; Step S3, decomposing the business intent, establishing a connection between the business intent and the device instruction, generating a business intent-policy model, and generating a security policy according to the mapping relationship between the business intent-policy model; Step S4: The SDN controller obtains the security policy and sends it to relevant network devices; continuously verifies the validity of the security policy, and generates a final security policy based on business intent through continuous verification.
[0029] In step S1, the business requirements of the power Internet of Things are obtained, including: In the electric power IoT scenario, data connection is formed with various electric power equipment (such as electric meters) through IoT gateways; various electric power equipment are used to regularly collect the status information of the underlying equipment according to the user's business needs, and the collected status information (business needs) is transmitted and stored to the server through the IoT gateway. Among them, the business needs include but are not limited to security management and protection (data encryption transmission, deployment of firewalls, opening and closing of intrusion detection systems, connection ports, disabled ports, etc.), energy consumption analysis and optimization, fault alarm and diagnosis, remote control and scheduling, and data collection and monitoring.
[0030] In step S2, a sentence state long short-term memory network model is used to extract features of the business needs, and the business needs of the power Internet of Things are converted into business intents, including four processes: entity labeling, sentence vectorization, feature code extraction, and business intent conversion.
[0031] First, entity labeling is performed, including dividing business requirements into four types of labels: entity boundaries, entity categories, relationship categories, and entity locations.
[0032] For entity boundary labels, the BIO mode is used to mark the boundary information of each word in the business requirements. More specifically, B is used to mark the first word in each entity phrase in the business requirements, O is used to mark words that are not defined as entities in the business requirements, and I is used to mark words other than the above two cases in the business requirements. At the same time, the business requirements are set to five entity category labels, namely: Location, Bandwidth, Delay, PacketLoss and Value. For relationship categories, the business requirements are set to six relationship category labels, namely: Connect, Disconnect, Max, Min, More-than and Less-than; in addition, for the stacking phenomenon of entity relationships, "-M" is added after the entity relationship for identification. For entity settings, it is used to identify the position of the business requirement entity in the triple, that is: the first position in the triple is represented by "1" and the second position is represented by "2".
[0033] After the entity labeling is completed, in order to obtain the vector format acceptable to the sentence state long short-term memory network model, the business requirement sentence needs to be vectorized. Specifically: The word features, magnetic annotation features, and position features in business requirements are extracted from multiple dimensions to obtain three feature vectors, which are then fused into a complete embedding vector by superposition. The word features of business requirements are obtained using the word vector acquisition technology Word2Vec; the magnetic annotation features of business requirements are obtained using the Stanford CoreNLP software toolkit; and the position features of business requirements are determined by the distance between the sequence positions of the two entities in the above triples.
[0034] After the sentence vectorization is completed, the feature code extraction process is executed, specifically: The embedding vector is used as the feature code extraction object, and the sentence state long short-term memory network model is used to iteratively calculate the embedding vector to obtain the feature state of the business demand, that is: ; in, Indicates the business needs for t The characteristic state generated by the iteration is Indicates the business needs for t The word feature state and sentence state of a single sentence generated by the iteration, Indicates the business needs for t The global sentence state produced by the iteration.
[0035] During the iterative calculation process, the characteristic state calculation method of the first-level node is as follows: ; During the iterative calculation process, the characteristic state calculation method of the secondary node is as follows: ; in, Represents the concatenation of each node and its preceding and following word feature states, and represents the gating parameters for the influx of information about the state of each cell, represents the output gate parameter of each node, , , and represents the model parameters, Represents the sigmoid function.
[0036] Based on the above-mentioned first-level nodes and second-level nodes, the feature encoding of the business requirements is obtained, that is, each word in the business requirement sentence is mapped to a first-level node, and the sentence state is mapped to a second-level node. In each iterative calculation, the feature state calculation method of the first-level node is jointly determined by the corresponding first-level node of the previous layer and its previous and subsequent first-level nodes and second-level nodes, while the feature state of the second-level node is jointly determined by all first-level nodes and second-level nodes of the previous layer.
[0037] The extracted feature code is used as the business intent conversion object and input into the sentence state long short-term memory network model to perform decoding operation on the feature code to obtain the label vector of the business intent. Specifically: The sentence state long short-term memory network model includes an input gate, an output gate, and a forget gate. i Words The input of the sentence state long short-term memory network model decoding unit is i word feature code, previous word label vector and the previous word hidden state , the decoding unit output is i The label vector of the word , and the current hidden state of the decoding unit loop calculation can be obtained and cell status .
[0038] Get label vector Then, calculate the label vector and the entity tag set k The corresponding probability of a note is: ; ; in, Words that express business intent The score value of the corresponding entity label, Words that express business intent Identified as the k The probability of a note, and Represents the model parameters. After the above two calculations, the business intent sentence word Recognized as a tag y ,Right now: ; A softmax classifier model is used to perform entity prediction on the labels to obtain the final required business intent.
[0039] After step S2, the business requirements of the power Internet of Things are formally converted into business intents. Then, step S3 is executed to decompose the business intents. Specifically: The business intent is split according to the behavior and operation object. After the split, the relevant behavior and ID attributes are matched based on the inherent attribute characteristics of the operation object, and the connection relationship between the business intent and the device instruction is established, thereby generating a business intent-policy model. According to the mapping relationship of the business intent-policy model, a security policy is generated. Among them, the business intent-policy model is shown in Table 1: Table 1 Business Intent-Strategy Model
[0040] Then, step S4 is executed, where the SDN controller obtains the above security policy and sends it to relevant network devices; and the accuracy and recall rate of the security policy are calculated to verify the effectiveness of the security policy, namely: ; in, and They represent the accuracy and recall of the security strategy respectively. Indicates the number of times the true value is the same as the predicted value and is 1; Indicates the number of true values 0 but predicted as 1. It indicates the number of cases where the true value is 1 but is predicted to be 0. The higher the accuracy and recall, the more effective the security policy is. Through continuous verification, the final security policy based on business intent is generated.
[0041] Embodiment 2 This embodiment discloses a system for automatically generating security policies for the power Internet of Things based on business intent.
[0042] like Figure 2 As shown in the figure, the automatic generation system of power Internet of Things security policies based on business intent includes: The business demand acquisition module is configured to: acquire business demands of the power Internet of Things; The business intention conversion module is configured to: extract features of the business requirements using a sentence state long short-term memory network model, and convert the business requirements of the power Internet of Things into business intentions; The security policy generation module is configured to: decompose the business intent, establish a connection between the business intent and the device instruction, generate a business intent-policy model, and generate a security policy according to a mapping relationship between the business intent-policy model; The verification module is configured as follows: the SDN controller obtains the security policy and sends it to relevant network devices; continuously verifies the validity of the security policy, and generates a final security policy based on business intent through continuous verification. Embodiment 3 The purpose of this embodiment is to provide a computer-readable storage medium.
[0043] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps in the method for automatically generating a security strategy for an electric power Internet of Things based on business intent as described in Example 1 of the present disclosure.
[0044] Embodiment 4 The purpose of this embodiment is to provide an electronic device.
[0045] An electronic device comprises a memory, a processor and a program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the method for automatically generating a security strategy for an electric power Internet of Things based on business intent as described in Example 1 of the present disclosure are implemented.
[0046] The steps involved in the apparatuses of the above embodiments 2, 3 and 4 correspond to the method embodiment 1, and the specific implementation methods can refer to the relevant description part of embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood to include any medium that can store, encode or carry an instruction set for execution by a processor and enable the processor to execute any method in the present invention.
[0047] Those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.
[0048] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.
Claims
1. A method for automatically generating security policies for power Internet of Things based on business intent, characterized in that: include: Obtain business requirements for the power Internet of Things; The sentence state long short-term memory network model is used to extract features of the business requirements and convert the business requirements of the power Internet of Things into business intents; Decomposing the business intent, establishing a connection between the business intent and the device instruction, generating a business intent-policy model, and generating a security policy according to a mapping relationship between the business intent-policy model; The SDN controller obtains the security policy and sends it to the network device; Continuously verify the effectiveness of security policies and generate the final security policies based on business intent through continuous verification.
2. The method for automatically generating security policies for the power Internet of Things based on business intent according to claim 1 is characterized in that: Convert the business requirements of the power Internet of Things into business intent, including: entity labeling, sentence vectorization, feature code extraction and business intent conversion.
3. The method for automatically generating security policies for the power Internet of Things based on business intent according to claim 2 is characterized in that: The statement vectorization includes: The word features, magnetic label features and position features in business needs are extracted through multiple dimensions to obtain three feature vectors, which are then fused into a complete embedding vector by superposition and fusion.
4. The method for automatically generating security policies for the power Internet of Things based on business intent according to claims 2-3 is characterized in that: The feature code extraction includes: The embedding vector is used as the feature code extraction object, and the sentence state long short-term memory network model is used to iteratively calculate the embedding vector and obtain the feature encoding of the business demand. Specifically: each word in the business demand sentence is mapped to a first-level node, and the sentence state is mapped to a second-level node. In each iterative calculation, the feature state calculation method of the first-level node is jointly determined by the corresponding first-level node of the previous layer and its previous and subsequent first-level nodes and second-level nodes, and the feature state of the second-level node is jointly determined by all the first-level nodes and second-level nodes of the previous layer.
5. The method for automatically generating security policies for the power Internet of Things based on business intent according to claims 2-4 is characterized in that: The business intent conversion includes: The feature code is used as the business intent conversion object, and the sentence state long short-term memory network model is used to decode the feature code to obtain the label vector of the business intent; then, the classifier model is used to perform entity prediction on the label vector to obtain the final required business intent.
6. The method for automatically generating security policies for the power Internet of Things based on business intent according to claim 1 is characterized in that: Decompose the business intent, including: The business intent is split according to behavioral actions and operation objects. After the splitting, the relevant behavioral actions and ID attributes are matched based on the inherent attribute characteristics of the operation object, and the connection relationship between the business intent and the device instructions is established, thereby generating a business intent-policy model.
7. The method for automatically generating security policies for the power Internet of Things based on business intent according to claim 1 is characterized in that: The effectiveness of the security strategy is verified by calculating the accuracy and recall of the security strategy.
8. The automatic generation system of power Internet of Things security strategy based on business intent is characterized by: include: The business demand acquisition module is configured to: acquire business demands of the power Internet of Things; The business intention conversion module is configured to: extract features of the business requirements using a sentence state long short-term memory network model, and convert the business requirements of the power Internet of Things into business intentions; The security policy generation module is configured to: decompose the business intent, establish a connection between the business intent and the device instruction, generate a business intent-policy model, and generate a security policy according to a mapping relationship between the business intent-policy model; The verification module is configured to: the SDN controller obtains the security policy and sends it to the network device; Continuously verify the effectiveness of security policies and generate the final security policies based on business intent through continuous verification.
9. A computer-readable storage medium having a program stored thereon, characterized in that: When the program is executed by a processor, the steps in the method for automatically generating a security strategy for the power Internet of Things based on business intent as described in any one of claims 1 to 7 are implemented.
10. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the method for automatically generating power Internet of Things security policies based on business intent as described in any one of claims 1 to 7 are implemented.
