Building digital information security management method and system based on block chain

Abstract

The invention discloses a building digital information security management method and system based on a block chain, and the method comprises the steps: collecting building BIM parameters through distributed nodes, constructing an initial information block, extracting feature vectors through a graph convolutional neural network, carrying out the verification through a block chain consensus mechanism, storing the feature vectors in a specific structure, and constructing a security evaluation model in combination with the parameters. And when the threshold value is exceeded, triggering the smart contract to execute operation. The system comprises a parameter acquisition unit, a feature extraction unit, a consensus verification unit, a chain storage unit, a security assessment unit, an intelligent contract execution unit and other associated units, and all the units operate cooperatively. By dynamically adjusting a storage and verification mechanism, building component parameters are utilized to construct a model, the integrity of feature extraction and the storage security are enhanced, the problems of insufficient feature characterization and lack of dynamic management in the prior art are solved, and the method is suitable for building full-life-cycle information security management.

Description

Technical Field

[0001] This invention relates to the field of building information management, and in particular to a blockchain-based method and system for building digital information security management. Background Technology

[0002] As the construction industry advances its digitalization process, Building Information Modeling (BIM) is increasingly being used throughout the entire project lifecycle. The parameters it encompasses, such as geometry, materials, component connections, and construction progress, require efficient and secure management. Graph convolutional neural networks (BNNs) demonstrate advantages in processing the associated features of building components, extracting complex relationships between parameters. However, with the increasing scale of digital building parameters, security issues in information transmission and storage become prominent. Traditional management methods struggle to address risks such as information tampering and access control confusion in a distributed environment. Blockchain-based management methods and systems have emerged as an important direction for exploring solutions to these problems.

[0003] Existing technologies for building digital information security management have two significant drawbacks. First, when combining building BIM information models with graph convolutional neural networks, feature extraction is limited to single component parameters, failing to fully integrate the spatial topological relationships and dynamically changing parameters of building components. This results in extracted feature vectors that do not adequately represent the overall structural information of the building, affecting the accuracy of subsequent security assessments. Second, the information storage and verification mechanisms lack dynamic adjustment capabilities. The storage structure does not consider changes in parameters such as building construction progress and component lifespan, and the verification difficulty is not linked to the data volume and security assessment results, leading to low storage efficiency and verification delays, making it difficult to meet the information security management needs throughout the entire building project lifecycle. Summary of the Invention

[0004] In order to overcome the shortcomings and deficiencies of existing technologies, this invention provides a blockchain-based method and system for building digital information security management.

[0005] The technical solution adopted in this invention is a blockchain-based method for digital information security management of buildings, comprising the following steps:

[0006] Step S1: Distributed node collection is performed on the geometric parameters, material property parameters, component connection parameters and construction progress parameters included in the building BIM information model. Different parameters are converted into fixed-length information digests through a hash algorithm, and initial information blocks are constructed according to the spatial topological relationship of building components.

[0007] Step S2: Use graph convolutional neural networks to extract features from the building digitization parameters in the initial information block. Capture the correlation features between different building component parameters through multi-layer convolution operations to generate a feature vector set. The setting of the convolution kernel is matched with the type and parameter dimension of the building component.

[0008] Step S3: Import the feature vector set into the blockchain consensus mechanism module, and use the improved proof-of-work algorithm to verify the feature vectors submitted by different nodes. During the verification process, the construction progress parameter is introduced as a dynamic adjustment factor to adjust the verification difficulty coefficient.

[0009] Step S4: Store the building digitization parameters corresponding to the verified feature vectors in a chain. The storage structure adopts a Merkle tree structure. Each leaf node corresponds to a specific parameter of a building component, and non-leaf nodes are combinations of hash values ​​of their child nodes. A verification block is generated every preset number of blocks.

[0010] Step S5: Construct a building digital information security assessment model based on the output of the graph convolutional neural network. The model input is the difference between the real-time collected building BIM parameters and the historical parameters stored in the blockchain. The security assessment value is output through a multilayer perceptron. The range of the assessment value is related to the importance of the building components.

[0011] Step S6: When the security assessment value exceeds the preset threshold, the blockchain's smart contract module is triggered. The smart contract automatically performs the corresponding information isolation or permission adjustment operations according to the preset building component priority parameters. The operation record is written into a new block in real time and broadcast to all nodes.

[0012] Furthermore, in step S2, the feature extraction process of the graph convolutional neural network adopts the following model formula:

[0013]

[0014] in, This represents the output feature value of the i-th building component in the j-th feature dimension; For activation functions; The weight parameters are those of the k-th convolutional kernel. This represents the k-th original digital parameter value of the i-th building component, including geometric dimensions and material strength. is the spatial association matrix element between building components i and j, with a value of 1 indicating a direct association and 0 indicating no direct association; b is the bias term; n is the total number of digital building parameters.

[0015] In the consensus mechanism of step S3, the dynamic adjustment process of the verification difficulty coefficient is combined with the dimension parameter of the feature vector. When the dimension of the feature vector increases, the verification difficulty coefficient increases linearly.

[0016] Furthermore, during the chained storage process in step S4, a dynamic update model for the building's digital parameters is constructed, and the model formula is:

[0017]

[0018] in, for The current state of the block storage; The block storage state at time t; A historical state retention coefficient, the value of which is related to the service life parameters of building components; The digital parameters of the building collected at time t and The difference in time parameters, including temperature changes and dynamic parameters of structural deformation; The hash function value for storing the state at time t;

[0019] In the safety assessment model of step S5, the matching degree between the dynamically updated storage state and the real-time acquired parameters is used as one of the assessment indicators. The calculation of the matching degree involves the cross-validation of the position parameters and time parameters of the components in the building BIM model.

[0020] Furthermore, in the process of constructing the security assessment model in step S5, the following security assessment value calculation model is constructed by combining the feature vector output by the graph convolutional neural network and the historical parameters stored in the blockchain:

[0021]

[0022] Where E is the safety assessment value; The weighting coefficient for the m-th digital building parameter is determined based on the safety importance level of the component. This refers to the m-th parameter value collected in real time. The historical average value of the m-th parameter stored in the blockchain; is the deviation influence coefficient of the m-th parameter, which is related to the seismic resistance grade parameters of the building structure; p is the total number of digital building parameters involved in the evaluation;

[0023] During the smart contract execution process in step S6, the preset threshold setting and the weighting coefficients in the evaluation value calculation model are... Establish a connection.

[0024] Furthermore, during the initial information block construction in step S1, construction stage parameters of the building BIM model are introduced as the basis for block grouping. The digital parameters of the building at different construction stages are divided into different sub-blocks, and the different sub-blocks are linked through the following model formula:

[0025]

[0026] in, The strength of the association between sub-blockchain a and sub-blockchain b; Let c be the standardized value of the c-th building digitization parameter in sub-blockchain a; Let c be the standardized value of the building digitization parameter in sub-blockchain b; This is an XOR operation; q represents the time overlap coefficient of the construction stages corresponding to sub-blockchains a and b; q represents the number of parameters used for association calculation.

[0027] In the feature extraction process of step S2, different convolution kernels of different sizes are used for the parameters of different sub-blockchains. The size of the convolution kernel is proportional to the number of building components contained in the sub-blockchain.

[0028] Furthermore, in the smart contract module of step S6, the execution strength of the permission adjustment operation is determined by the following model formula:

[0029]

[0030] Where P is the execution strength parameter for permission adjustment, and its value range is... is a proportionality coefficient, which is related to the safety level parameters of the building project; E is the safety assessment value obtained in step S5. This is the preset maximum safety assessment value; The importance score is assigned to the d-th affected building component, and the score is determined based on the component's functional and structural parameters; r is the total number of affected building components;

[0031] In the consensus verification of step S3, this execution strength parameter is used as an additional reference indicator for node verification. When the execution strength parameter is greater than 0.6, an additional node secondary verification process is initiated.

[0032] Furthermore, step S3 includes the following sub-steps:

[0033] Step S3.1: Collect feature vectors processed by graph convolutional neural networks from different nodes, perform preliminary verification of the dimension and data type of the feature vectors, and remove feature vectors with mismatched dimensions or incorrect data types.

[0034] Step S3.2: Determine the current verification difficulty benchmark value based on the construction progress parameters. The larger the progress parameters, the higher the benchmark value. Then, dynamically adjust the verification difficulty coefficient in combination with the number of feature vectors to make the verification speed match the amount of data.

[0035] Step S3.3: The improved proof-of-work algorithm is used to perform hash operation on the feature vector. The operation result is required to meet the preset leading zero number condition. The leading zero number is determined by the adjusted verification difficulty coefficient. Each node performs the operation independently and submits the result.

[0036] Step S3.4: Summarize and compare the calculation results submitted by different nodes, select the node that first meets the conditions and has the correct result as the accounting node, write the node's identification information and verification result into a temporary block, and wait for subsequent connection with other blockchains.

[0037] Furthermore, step S4 includes the following sub-steps:

[0038] Step S4.1: Classify the building digitization parameters corresponding to the feature vector confirmed by the accounting node according to the spatial location of the building components. Component parameters in the same spatial area are grouped into the same data group for easy subsequent query and management. The classification is based on the three-dimensional coordinate parameters in the building BIM model.

[0039] Step S4.2: Perform hash calculation on the parameters in each data group to generate a unique hash value for each parameter. Then, perform a second hash operation on the hash values ​​of all parameters in the same data group to obtain the Merkle leaf node values ​​of that data group.

[0040] Step S4.3: According to the construction rules of the Merkle tree, the leaf node values ​​are hashed upwards layer by layer to generate non-leaf node values ​​until the root node hash value is obtained. The root node hash value is stored in the block header as the unique identifier of the data group, simplifying the process of verifying the association between blocks.

[0041] Step S4.4: Combine the information including the root node hash value, data group parameters, the previous block hash value, and timestamp into a new block, and broadcast it to all nodes through the peer-to-peer network. After receiving the block, different nodes verify the validity of the block information. If the verification is successful, they add it to the end of their local blockchain.

[0042] Furthermore, step S5 includes the following sub-steps:

[0043] Step S5.1: Extract the digital parameters of historical buildings related to the current assessment object from the blockchain, calculate the average value and fluctuation range of different parameters, and establish a historical parameter benchmark library. The update cycle of the benchmark library is consistent with the maintenance cycle parameters of the building components.

[0044] Step S5.2: Collect the current parameters of the building BIM model in real time, extract the feature vector of the current parameters through graph convolutional neural network, calculate the difference between the feature vector of the corresponding parameter in the historical parameter benchmark library, and obtain the parameter change matrix.

[0045] Step S5.3: Input the parameter change matrix into the safety assessment model. The model calculates the safety assessment value based on the weight coefficients and deviation influence coefficients of different parameters. The weight coefficients and deviation influence coefficients are retrieved from the attribute database of the building BIM model.

[0046] Step S5.4: Compare the calculated security assessment value with the preset threshold to generate a comparison result. The comparison result includes the specific parameters and the extent to which the assessment value exceeds or falls below the threshold, providing a basis for subsequent smart contract execution.

[0047] A blockchain-based digital information security management system for buildings includes:

[0048] The distributed acquisition and preprocessing unit for building BIM parameters has its input end connected to multiple building information acquisition terminals and its output end connected to the input end of the graph convolutional neural network feature extraction unit. It is used to acquire and preliminarily process different parameters of the building BIM information model.

[0049] The graph convolutional neural network feature extraction unit has its output connected to the input of the blockchain consensus verification unit and the security assessment model construction unit, respectively, and is used to extract features from building digitization parameters and generate feature vectors.

[0050] The blockchain consensus verification unit, whose output is connected to the input of the chain storage unit, is used to verify the feature vector and determine the accounting node.

[0051] The chained storage unit, with its bidirectional connection to the security assessment model construction unit, is used to store the digital parameters of the building structure and related verification information in a preset structure.

[0052] The security assessment model building unit connects its output to the input of the smart contract execution unit. It is used to build a model based on parameters stored on the blockchain and real-time parameters and output an assessment value.

[0053] The smart contract execution unit, whose output is connected to the input of the blockchain consensus verification unit, is used to perform corresponding operations based on the evaluation value and feed the operation record back to the blockchain system.

[0054] Beneficial Effects: This invention proposes a blockchain-based method and system for building digital information security management. Through the deep integration of building BIM information models, graph convolutional neural networks, and building digital parameters, it achieves multiple beneficial effects. At the feature extraction level, by leveraging graph convolutional neural networks to capture the spatial topological relationships and dynamically changing parameters of building components, the generated feature vectors comprehensively represent the overall structural information of the building. This overcomes the shortcomings of existing technologies where feature extraction is limited to single component parameters and insufficient representation of overall structural information, thus improving the correlation between feature vectors and the actual state of the building. Regarding information storage and verification, the blockchain's chain-like storage structure dynamically adjusts the storage status and verification difficulty based on parameters such as construction progress and component lifespan. This ensures that storage efficiency matches data volume, and verification speed matches security requirements, solving the problem of existing storage and verification mechanisms lacking dynamic adjustment capabilities and being difficult to adapt to the entire project lifecycle. Meanwhile, the immutability and distributed nature of blockchain ensure the security of the transmission and storage of digital building parameters. Smart contracts automatically execute operations such as permission adjustment based on security assessment results, further enhancing the timeliness and accuracy of information security management. By integrating the technological advantages of blockchain, BIM models and graph convolutional neural networks, a complete security management chain covering parameter acquisition, feature extraction, verification and storage, security assessment and response processing is formed. Attached Figure Description

[0055] Figure 1 This is a flowchart of the method steps of the present invention;

[0056] Figure 2 This is a diagram showing the system unit composition of the present invention. Detailed Implementation

[0057] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of this application can be combined with each other. The application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0058] like Figure 1 As shown, the blockchain-based digital information security management method for buildings includes the following steps:

[0059] Step S1: Distributed node collection is performed on the geometric parameters, material property parameters, component connection parameters and construction progress parameters included in the building BIM information model. Different parameters are converted into fixed-length information digests through a hash algorithm, and initial information blocks are constructed according to the spatial topology of building components. Each block contains the hash value of the previous block and the timestamp of the currently collected parameters.

[0060] Specifically, in building digital information management, the effective collection and processing of various parameters from the building BIM information model is a fundamental step. The geometric parameters in the building BIM information model reflect the spatial characteristics of components, such as shape and size; material property parameters reflect the physical performance of components; component connection parameters relate to structural stability; and construction progress parameters are linked to the project's timeline. Collecting these parameters through distributed nodes ensures the breadth and dispersion of data sources. Converting them into information digests using hash algorithms enables standardized parameter processing. Constructing initial information blocks based on the spatial topological relationships of building components ensures that the information within each block corresponds to the actual building structure. Each block contains the hash value and timestamp of the previous block, laying the foundation for subsequent chained storage and traceability.

[0061] In practice, distributed nodes are deployed in different areas of the building project. Each node is responsible for collecting parameters of building components within a specific range, including geometric parameters such as component length, width, and height; material properties such as compressive strength and shear strength; connection parameters such as connection method and connection strength; and construction progress parameters such as the percentage of completed work and the current construction stage. After the collected parameters are transmitted to the processing nodes, each parameter is converted into a fixed-length message digest using a hash algorithm. Different types of parameters are converted into strings of the same length. According to the spatial topology of the building components, the message digests corresponding to the parameters of interconnected components within the same area are integrated to construct initial information blocks. Each block contains the message digest of the currently collected parameters and also embeds the hash value of the previous block to form a correlation. At the same time, the specific time of each parameter collection is recorded as a timestamp to ensure the temporal sequence and correlation of information within the block.

[0062] Step S2: Use graph convolutional neural networks to extract features from the building digitization parameters in the initial information block. Capture the correlation features between different building component parameters through multi-layer convolution operations to generate a feature vector set. The setting of the convolution kernel is matched with the type and parameter dimension of the building component.

[0063] Specifically, graph convolutional neural networks (GNNs) have unique advantages in processing data with topological structures, making them suitable for feature extraction from building BIM information model parameters. Complex relationships exist between digital building parameters; parameters of different building components are not isolated but rather mutually influential and interconnected. Through multi-layer convolutional operations, GNNs can deeply uncover the potential correlations between these parameters, generating feature vector sets that centrally reflect the key information of the digital building parameters. Matching the convolutional kernel settings to the building component type and parameter dimensions makes the feature extraction process more targeted, improving the fit between the extracted features and the actual building conditions, and providing high-quality feature data for subsequent information processing and safety assessment.

[0064] In the specific implementation process, the digital building parameters within the initial information block are input into a graph convolutional neural network. The network's input layer receives parameter data, including dimensional data of geometric parameters, performance data of material properties, structural data of component connection parameters, and time data of construction progress parameters. Depending on the type of building component, such as beams, columns, wall panels, and floor slabs, different sizes and numbers of convolutional kernels are set. For components with higher parameter dimensions, more and larger convolutional kernels are used. In the first layer of convolution, the convolutional kernel performs convolution operations with the input parameter data, capturing the correlation features of component parameters within a local range. After multiple layers of convolution, the range of feature capture is gradually expanded to include long-range correlation features between different components. Each layer of convolution outputs a corresponding feature map, which is then integrated after multiple layers of processing to generate a feature vector set. Each feature vector corresponds to a set of comprehensive features of building components, covering correlation information in geometry, materials, connections, and progress.

[0065] Step S3: Import the feature vector set into the blockchain consensus mechanism module, and use the improved proof-of-work algorithm to verify the feature vectors submitted by different nodes. During the verification process, the construction progress parameter is introduced as a dynamic adjustment factor to adjust the verification difficulty coefficient.

[0066] Specifically, the blockchain consensus mechanism is crucial for ensuring data consistency in distributed systems. Importing the feature vector set generated by a graph convolutional neural network into this mechanism module allows for the verification of the feature vector's validity. The improved proof-of-work algorithm, optimized from the original, better adapts to the needs of digital building information management. Introducing construction progress parameters as a dynamic adjustment factor allows the verification difficulty to vary as the project progresses. As construction progresses, the amount and importance of data differ; adjusting the verification difficulty coefficient ensures the efficiency and security of the verification process, avoiding overly cumbersome verification in stages with low data volume or overly lenient verification in stages with high data volume.

[0067] In the specific implementation process, the feature vector set, output from the graph convolutional neural network, is transmitted to the blockchain consensus mechanism module. The module first obtains current construction progress parameters, such as the number of days the project has been under construction, the number of completed procedures, and the remaining workload, and determines the initial verification difficulty coefficient based on these parameters. When the construction progress parameters are large, i.e., the project is in a later stage, the initial verification difficulty coefficient is set higher; when the construction progress parameters are small, the initial verification difficulty coefficient is set lower. Next, the improved proof-of-work algorithm processes the feature vectors. Each node performs hash operations on the feature vectors based on its own computing power, and the operation must meet specific conditions determined by the verification difficulty coefficient. As the number of feature vectors increases, the verification difficulty coefficient increases accordingly to ensure the rigor of the verification. After each node completes the operation, it submits the results. The system compares the results, selects those that meet the conditions, and completes the verification of the feature vectors.

[0068] Step S4: Store the building digitization parameters corresponding to the verified feature vectors in a chain. The storage structure adopts a Merkle tree structure. Each leaf node corresponds to a specific parameter of a building component, and non-leaf nodes are combinations of hash values ​​of their child nodes. A verification block is generated every preset number of blocks.

[0069] Specifically, chaining verified digital building parameters is a crucial means of ensuring information traceability and tamper-proofness. The Merkle tree structure, known for its efficient data integrity verification, enables the orderly management and rapid verification of large numbers of parameters through the hierarchical relationship between leaf and non-leaf nodes. Each leaf node corresponds to a specific parameter of a building component, ensuring clear correspondence in parameter storage. Non-leaf nodes are combinations of hash values ​​from their child nodes, allowing for the aggregation and verification of information from lower-level nodes. Furthermore, generating verification blocks at preset intervals further strengthens the supervision and verification of stored information, guaranteeing the accuracy and security of the stored data.

[0070] In the specific implementation process, the building's digital parameters corresponding to the verified feature vectors are collected, including the geometric parameters (such as cross-sectional dimensions and span), material property parameters (such as elastic modulus and density), component connection parameters (such as node stiffness and connection angle), and construction progress parameters (such as process duration and resource input) of each building component. Following the construction rules of the Merkle tree structure, the specific parameters of each building component are used as leaf nodes, and the hash value of each leaf node is calculated. Then, the hash values ​​of adjacent leaf nodes are combined to calculate the hash value of the non-leaf nodes at the next higher level, and so on, calculating upwards layer by layer until the hash value of the root node is generated. The hash value of the root node and the relevant information of each level of nodes are included in blocks, and each block is linked in chronological order of generation, with the hash value of the previous block embedded in the next block. Simultaneously, a preset number of block intervals is set. When this number is reached, a verification block is generated. The verification block contains the root node hash values ​​of all blocks within that interval. By verifying the verification block, the integrity of the stored parameters within that interval can be quickly checked.

[0071] Step S5: Construct a building digital information security assessment model based on the output of the graph convolutional neural network. The model input is the difference between the real-time collected building BIM parameters and the historical parameters stored in the blockchain. The security assessment value is output through a multilayer perceptron. The range of the assessment value is related to the importance of the building components.

[0072] Specifically, constructing a building digital information security assessment model is a core step in determining building information security. The output of a graph convolutional neural network contains the correlation features of building digital parameters, while historical parameters stored on the blockchain reflect the building's state information at different stages. Combining these two allows the security assessment model to more comprehensively consider changes in building information. The model input is the difference between real-time collected parameters and historical parameters, which intuitively reflects the magnitude of parameter changes. The output security assessment value, obtained through a multilayer perceptron, transforms complex parameter changes into quantifiable assessment results. Furthermore, the range of assessment values ​​is correlated with the importance of building components, making the assessment results more closely aligned with the actual safety requirements of the building structure.

[0073] In the specific implementation process, historical parameters of the building BIM information model are extracted from the blockchain. These parameters include geometric parameters, material property parameters, component connection parameters, and construction progress parameters of building components from different periods in the past. Simultaneously, various parameters of the current building BIM information model, also covering the above-mentioned types of parameters, are collected in real time. The difference between the real-time collected parameters and the historical parameters is calculated to obtain a set of parameter change differences. This set of differences is input into a safety assessment model constructed based on the output of a graph convolutional neural network. The model's multilayer perceptron contains multiple hidden layers, with each neuron processing the input difference data and incorporating correlation information from the feature vectors extracted by the graph convolutional neural network during the processing. Different ranges of safety assessment values ​​are set according to the importance of building components, such as distinguishing between load-bearing and non-load-bearing components, with stricter ranges for important components. Through the calculation of the multilayer perceptron, the current safety assessment value is output, reflecting the degree of impact of the difference between real-time and historical parameters on building information security.

[0074] Step S6: When the security assessment value exceeds the preset threshold, the blockchain's smart contract module is triggered. The smart contract automatically performs the corresponding information isolation or permission adjustment operations according to the preset building component priority parameters. The operation record is written into a new block in real time and broadcast to all nodes.

[0075] Specifically, the smart contract module plays a role in automatically responding to and handling security issues in the digital information security management of buildings. When the security assessment value exceeds a preset threshold, it indicates that there is a security risk in the building information. Triggering the smart contract module allows for timely countermeasures to prevent the risk from escalating. The smart contract executes corresponding operations based on preset building component priority parameters, making countermeasures more targeted and prioritizing the handling of information related to components that have a greater impact on the structural safety of the building. Writing operation records to new blocks in real time and broadcasting them to all nodes ensures the openness and traceability of operations. All nodes synchronously update information, maintaining the consistency of the entire system and ensuring the effective implementation of security management measures.

[0076] In the specific implementation process, the preset threshold is determined based on the priority parameters of the building components. These priority parameters are set based on the functional importance and structural safety of the components within the building structure; for example, the priority parameters of foundation components and main load-bearing components are higher than those of decorative components. When the safety assessment value output in step S5 exceeds the preset threshold for the corresponding component, the smart contract module is triggered. The smart contract obtains the priority parameters of the component from the building BIM information model and determines the type of operation to be performed based on the parameter value. If the parameter value is high, an information isolation operation is performed, restricting access permissions for the component's related information to authorized nodes only. If the parameter value is medium, a permission adjustment operation is performed, modifying the operation permissions of different nodes for the component's information. During the operation execution, the specific content of the operation is recorded, including the operation type, the component parameters involved, and the execution time. These records are used to generate a new information block, the hash value of the block is calculated, and the hash value of the previous block is embedded. The new block is then broadcast to all distributed nodes. Each node receives the block and updates its local blockchain, completing the entire operation process.

[0077] Preferably, in step S2, the feature extraction process of the graph convolutional neural network adopts the following model formula:

[0078]

[0079] in, This represents the output feature value of the i-th building component in the j-th feature dimension; For activation functions; The weight parameters are those of the k-th convolutional kernel. This represents the k-th original digital parameter value of the i-th building component, including geometric dimensions and material strength. is the element of the spatial association matrix between building components i and j, with a value of 1 indicating a direct association and 0 indicating no direct association; b is the bias term; n is the total number of digital building parameters.

[0080] This model formula uses the spatial relationships between building components as weight factors in the convolution operation, ensuring that the extracted features not only include the parameter information of individual components but also incorporate the influencing factors of adjacent components, thus enhancing the feature vector's ability to represent the overall structural information of the building. In the consensus mechanism of step S3, the dynamic adjustment process of the verification difficulty coefficient is combined with the dimensionality parameter of the feature vector; as the dimension of the feature vector increases, the verification difficulty coefficient increases linearly.

[0081] Specifically, by constructing a specific model formula, the spatial relationships between building components are deeply integrated into the feature extraction process. This model fully utilizes digital parameters such as geometric dimensions and material strength from the building BIM information model, as well as spatial association matrix elements reflecting the connection status between components. The results of multi-layer convolution operations are processed through activation functions to generate feature values ​​that reflect the mutual influence between components. In implementation, the weight parameters of the convolution kernel are first determined according to the type of building component; different types of components (such as beams, columns, and wall panels) correspond to different weight settings. Simultaneously, the values ​​of the association matrix elements are determined based on the spatial layout of the components in the building BIM model. Through multi-layer convolution operations, the feature values ​​of each building component not only contain its own parameter information but also integrate the parameter influence of adjacent components, enhancing the feature vector's ability to represent the overall structural information of the building. In the consensus mechanism of step S3, the dimensionality parameter of the feature vector is used as the basis for adjusting the verification difficulty coefficient. When the feature vector dimension increases due to the increase in building digitization parameters, the verification difficulty coefficient increases linearly to adapt to changes in data volume and ensure the effectiveness and security of the verification process.

[0082] Preferably, during the chained storage process in step S4, a dynamic update model for the building's digital parameters is constructed, and the model formula is:

[0083]

[0084] in, for The current state of the block storage; The block storage state at time t; A historical state retention coefficient, the value of which is related to the service life parameters of building components; The digital parameters of the building collected at time t and The difference in time parameters, including temperature changes and dynamic parameters of structural deformation; The hash function value for storing the state at time t.

[0085] This model introduces dynamic changes in building parameters, enabling the information stored on the blockchain to reflect changes in the building's status in real time. In the security assessment model of step S5, the matching degree between this dynamically updated storage status and the real-time collected parameters is used as one of the evaluation indicators. The calculation of the matching degree involves the cross-validation of the position parameters and time parameters of the components in the building BIM model.

[0086] Specifically, a dynamic update model for building digitization parameters is constructed to enable the information stored on the blockchain to reflect changes in the building's status in real time. This model introduces a historical state retention coefficient, the value of which is related to the service life parameters of building components. Components with longer service lives correspond to higher retention coefficients to ensure the stability of historical state information. Simultaneously, it incorporates dynamic changes in building digitization parameters, such as real-time monitoring data on temperature changes and structural deformation, as well as the hash function value of the current storage state. The updated block storage state is obtained through model calculations. In implementation, the initial value of the historical state retention coefficient is first determined based on the design service life of the components in the building BIM information model. Then, real-time data on changes in building parameters such as temperature and deformation are collected, and the difference between the parameters and the previous time point is calculated. This difference, combined with the hash value of the current storage state, is substituted into the model to calculate the new storage state. In the security assessment in step S5, the matching degree between the dynamically updated storage state and the real-time collected parameters is calculated. The matching degree assessment involves cross-validation of the location parameters (such as coordinate information) and time parameters (such as collection time) of the components in the building BIM model. Through multi-dimensional comparison, the comprehensiveness of the security assessment is improved.

[0087] Preferably, in the process of constructing the security assessment model in step S5, the following security assessment value calculation model is constructed by combining the feature vector output by the graph convolutional neural network and the historical parameters stored in the blockchain:

[0088]

[0089] Where E is the safety assessment value; The weighting coefficient for the m-th digital building parameter is determined based on the safety importance level of the component. This refers to the m-th parameter value collected in real time. The historical average value of the m-th parameter stored in the blockchain; is the deviation influence coefficient of the m-th parameter, which is related to the seismic resistance grade parameter of the building structure; p is the total number of digital building parameters involved in the evaluation.

[0090] During the smart contract execution process in step S6, the preset threshold setting and the weighting coefficients in the evaluation value calculation model are... Correlation: For parameters with larger weighting coefficients, the corresponding threshold range is set more strictly, and the threshold will be adjusted linearly as the building's service life parameter increases.

[0091] Specifically, a safety assessment value calculation model is designed, comprehensively considering the weighting coefficients of building digital parameters, the deviation between real-time and historical parameters, and the deviation impact coefficient. The weighting coefficients are determined based on the safety importance level of components in the building structure, with core load-bearing components having higher weighting coefficients than non-load-bearing components. The deviation impact coefficient is related to the seismic resistance level of the building structure; buildings with higher seismic resistance levels correspond to higher deviation impact coefficients. During implementation, the historical average value of building digital parameters is extracted from the blockchain, while real-time parameters are collected simultaneously. The difference between the two is calculated, and combined with the preset weighting coefficients and deviation impact coefficients, the safety assessment value is obtained through model calculation. In the smart contract execution of step S6, the preset threshold setting is related to the weighting coefficients. Parameters with larger weighting coefficients correspond to stricter threshold ranges. Simultaneously, the thresholds are linearly adjusted as the building's service life increases; the longer the service life, the narrower the threshold range, to adapt to the sensitivity requirements of parameter changes during building aging, ensuring that the smart contract can respond appropriately based on the actual condition of the building.

[0092] Preferably, during the initial information block construction in step S1, construction stage parameters of the building BIM model are introduced as the basis for block grouping, and the digital parameters of the building at different construction stages are divided into different sub-blockchains. The different sub-blockchains are linked through the following model formula:

[0093]

[0094] in, The strength of the association between sub-blockchain a and sub-blockchain b; Let c be the standardized value of the c-th building digitization parameter in sub-blockchain a; Let c be the standardized value of the building digitization parameter in sub-blockchain b; This is an XOR operation; q is the time overlap coefficient of the construction stages corresponding to sub-blockchains a and b; q is the number of parameters used for association calculation.

[0095] In the feature extraction process of step S2, different convolution kernels of different sizes are used for the parameters of different sub-blockchains. The size of the convolution kernel is proportional to the number of building components contained in the sub-blockchain.

[0096] Specifically, a model is constructed to determine the enforcement strength of permission adjustments. This model considers the safety assessment value, the maximum safety assessment value, a proportional coefficient, and the importance score of the affected components. The proportional coefficient is related to the safety level parameters of the building project; the higher the safety level, the larger the proportional coefficient. The importance score of the affected components is determined based on their functional parameters (such as load-bearing capacity) and structural parameters (such as material strength). In implementation, the proportional coefficient is first set according to the safety level of the building project. Then, combining the ratio of the safety assessment value obtained in step S5 to the preset maximum safety assessment value, and the sum of the importance scores of the affected components, the enforcement strength parameter for permission adjustments is calculated through the model. In the consensus verification of step S3, when the enforcement strength parameter exceeds a set threshold (e.g., 0.6), an additional node secondary verification process is initiated, increasing the number of verification nodes and the number of verifications to ensure the reliability of the verification results and reduce the possibility of misoperation even with a large enforcement strength.

[0097] Preferably, in the smart contract module of step S6, the execution strength of the permission adjustment operation is determined by the following model formula:

[0098]

[0099] Where P is the execution strength parameter for permission adjustment, and its value range is... is a proportionality coefficient, which is related to the safety level parameters of the building project; E is the safety assessment value obtained in step S5. This is the preset maximum safety assessment value; The importance score is given to the d-th affected building component, and the score is determined based on the component's functional and structural parameters; r is the total number of affected building components.

[0100] This model formula enables permission adjustment operations to dynamically change according to the severity of the security assessment and the importance of the affected components. In the consensus verification of step S3, this execution strength parameter is used as an additional reference indicator for node verification. When the execution strength parameter is greater than 0.6, an additional node secondary verification process is initiated.

[0101] Specifically, the model construction combines a proportional coefficient, the ratio of the safety assessment value to the maximum safety assessment value, and the importance score of the affected components to determine the enforcement strength of permission adjustments. The proportional coefficient is set based on the safety level parameters of the building project, with different coefficient values ​​corresponding to different safety levels. The importance score of the affected components is determined based on their functional parameters (such as whether they participate in major load-bearing) and structural parameters (such as material durability). During implementation, the proportional coefficient is first determined based on the project's safety level parameters in the building BIM information model. Then, the ratio of the safety assessment value obtained in step S5 to the maximum safety assessment value is calculated, and the importance scores of the affected components are summarized. These parameters are then substituted into the model to obtain the enforcement strength parameter. This parameter determines the specific measures for permission adjustments, such as restricting the scope of access and adjusting the level of operational permissions. In the consensus verification of step S3, when the enforcement strength parameter is greater than 0.6, an additional node secondary verification process is initiated, increasing the number of verification nodes and verification steps to ensure the accuracy of permission adjustment operations in high-security scenarios and avoid improper operations due to incorrect judgments by a single node.

[0102] Preferably, step S3 includes the following sub-steps:

[0103] Step S3.1: Collect feature vectors processed by graph convolutional neural networks from different nodes, perform preliminary verification on the dimension and data type of the feature vectors, and remove feature vectors with mismatched dimensions or incorrect data types to ensure that the data format is consistent when entering the consensus verification stage. This process relies on the preset parameter dimension standards in the building BIM model.

[0104] Step S3.2: Determine the current verification difficulty benchmark value based on the construction progress parameters. The larger the progress parameter, the higher the benchmark value. Then, dynamically adjust the verification difficulty coefficient in combination with the number of feature vectors to make the verification speed match the amount of data and avoid verification delays due to excessive data volume.

[0105] Step S3.3: The improved proof-of-work algorithm is used to perform hash operation on the feature vector. The operation result is required to meet the preset leading zero number condition. The leading zero number is determined by the adjusted verification difficulty coefficient. Each node performs the operation independently and submits the result.

[0106] Step S3.4: Summarize and compare the calculation results submitted by different nodes, select the node that first meets the conditions and has the correct result as the accounting node, write the node's identification information and verification result into a temporary block, and wait for subsequent connection with other blockchains.

[0107] Preferably, step S4 includes the following sub-steps:

[0108] Step S4.1: Classify the building digitization parameters corresponding to the feature vector confirmed by the accounting node according to the spatial location of the building components. Component parameters in the same spatial area are grouped into the same data group for easy subsequent query and management. The classification is based on the three-dimensional coordinate parameters in the building BIM model.

[0109] Step S4.2: Perform hash calculation on the parameters in each data group to generate a unique hash value for each parameter. Then, perform a second hash operation on the hash values ​​of all parameters in the same data group to obtain the Merkle leaf node values ​​of the data group, ensuring that the integrity of the parameters is verifiable.

[0110] Step S4.3: According to the construction rules of the Merkle tree, the leaf node values ​​are hashed upwards layer by layer to generate non-leaf node values ​​until the root node hash value is obtained. The root node hash value is stored in the block header as the unique identifier of the data group, simplifying the process of verifying the association between blocks.

[0111] Step S4.4: Combine the information including the root node hash value, data group parameters, the previous block hash value, and timestamp into a new block, and broadcast it to all nodes through the peer-to-peer network. After receiving the block, different nodes verify the validity of the block information. If the verification is successful, they add it to the end of their local blockchain.

[0112] Preferably, step S5 includes the following sub-steps:

[0113] Step S5.1: Extract the digital parameters of historical buildings related to the current assessment object from the blockchain, calculate the average value and fluctuation range of different parameters, and establish a historical parameter benchmark library. The update cycle of the benchmark library is consistent with the maintenance cycle parameters of the building components.

[0114] Step S5.2: Collect the current parameters of the building BIM model in real time, extract the feature vector of the current parameters through graph convolutional neural network, calculate the difference between the feature vector of the corresponding parameter in the historical parameter benchmark library, and obtain the parameter change matrix.

[0115] Step S5.3: Input the parameter change matrix into the safety assessment model. The model calculates the safety assessment value based on the weight coefficients and deviation influence coefficients of different parameters. The weight coefficients and deviation influence coefficients are retrieved from the attribute database of the building BIM model.

[0116] Step S5.4: Compare the calculated security assessment value with the preset threshold to generate a comparison result. The comparison result includes the specific parameters and the extent to which the assessment value exceeds or falls below the threshold, providing a basis for subsequent smart contract execution.

[0117] like Figure 2 As shown, the blockchain-based digital information security management system for buildings includes:

[0118] The distributed acquisition and preprocessing unit for building BIM parameters has its input end connected to multiple building information acquisition terminals and its output end connected to the input end of the graph convolutional neural network feature extraction unit. It is used to acquire and preliminarily process different parameters of the building BIM information model.

[0119] The graph convolutional neural network feature extraction unit has its output connected to the input of the blockchain consensus verification unit and the security assessment model construction unit, respectively, and is used to extract features from building digitization parameters and generate feature vectors.

[0120] The blockchain consensus verification unit, whose output is connected to the input of the chain storage unit, is used to verify the feature vector and determine the accounting node.

[0121] The chained storage unit, with its bidirectional connection to the security assessment model construction unit, is used to store the digital parameters of the building structure and related verification information in a preset structure.

[0122] The security assessment model building unit connects its output to the input of the smart contract execution unit. It is used to build a model based on parameters stored on the blockchain and real-time parameters and output an assessment value.

[0123] The smart contract execution unit, whose output is connected to the input of the blockchain consensus verification unit, is used to perform corresponding operations based on the evaluation value and feed the operation record back to the blockchain system.

[0124] A blockchain-based method and system for building digital information security management addresses this issue. At the feature extraction level, this method and system utilize graph convolutional neural networks to process geometric parameters and material properties in the building's BIM information model. It fully incorporates dynamically changing parameters such as the spatial topological relationships of building components and construction progress, enabling the generated feature vectors to comprehensively reflect the overall structural information of the building. This approach overcomes the shortcomings of previous technologies, which limited feature extraction to single component parameters and failed to adequately represent the overall structural information. It significantly enhances the correlation between feature vectors and the actual state of the building, providing more accurate foundational data for subsequent safety assessments.

[0125] In the information storage and verification phase, a chain-like storage structure built on blockchain technology dynamically adjusts the storage status based on building component lifespan parameters and construction progress parameters. Simultaneously, it links the verification difficulty coefficient with the number of feature vectors and security assessment results. This dynamic adjustment mechanism solves the problems of inflexible storage and verification mechanisms in the background technology, making it difficult to adapt to the full lifecycle management of building projects. It ensures that storage efficiency matches the data volume and that verification speed matches security requirements, guaranteeing the integrity of building digitization parameters throughout their entire lifecycle and the timeliness of verification.

[0126] In terms of overall collaboration, this method and system deeply integrate building BIM information models, graph convolutional neural networks, and blockchain to form a complete link from parameter acquisition to smart contract execution. The immutability of blockchain ensures the security of parameter transmission and storage, while smart contracts automatically execute operations such as permission adjustments based on security assessment values, strengthening the proactive nature of information security management. This collaborative model overcomes the shortcomings of isolated applications of various technologies in the background technologies, comprehensively improving the accuracy, dynamism, and security of building digital information security management, effectively addressing the information security challenges in the complex environment of building projects.

[0127] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," "link," and "fix" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0128] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A blockchain-based method for building digital information security management, characterized in that, include: Step S1: Distributed node collection is performed on the geometric parameters, material property parameters, component connection parameters and construction progress parameters included in the building BIM information model. Different parameters are converted into fixed-length information digests through a hash algorithm, and initial information blocks are constructed according to the spatial topology of building components. Step S2: Use graph convolutional neural networks to extract features from the building digitization parameters in the initial information block. Capture the correlation features between different building component parameters through multi-layer convolution operations to generate a feature vector set. The setting of the convolution kernel is matched with the type and parameter dimension of the building component. Step S3: Import the feature vector set into the blockchain consensus mechanism module, and use the improved proof-of-work algorithm to verify the feature vectors submitted by different nodes. During the verification process, the construction progress parameter is introduced as a dynamic adjustment factor to adjust the verification difficulty coefficient. Step S4: Store the building digitization parameters corresponding to the verified feature vectors in a chain. The storage structure adopts a Merkle tree structure. Each leaf node corresponds to a specific parameter of a building component, and non-leaf nodes are combinations of hash values ​​of their child nodes. A verification block is generated every preset number of blocks. Step S5: Construct a building digital information security assessment model based on the output of the graph convolutional neural network. The model input is the difference between the real-time collected building BIM parameters and the historical parameters stored in the blockchain. The security assessment value is output through a multilayer perceptron. The range of the assessment value is related to the importance of the building components. Step S6: When the security assessment value exceeds the preset threshold, the blockchain's smart contract module is triggered. The smart contract automatically performs the corresponding information isolation or permission adjustment operations according to the preset building component priority parameters. The operation record is written into a new block in real time and broadcast to all nodes.

2. The method according to claim 1, characterized in that, In step S2, the feature extraction process of the graph convolutional neural network adopts the following model formula: ； in, This represents the output feature value of the i-th building component in the j-th feature dimension; For activation functions; The weight parameters are those of the k-th convolutional kernel. This represents the k-th original digital parameter value of the i-th building component, including geometric dimensions and material strength. is the element of the spatial association matrix between building components i and j, with a value of 1 indicating a direct association and 0 indicating no direct association; b is the bias term; n is the total number of digital building parameters.

3. The method according to claim 1, characterized in that, In the chained storage process of step S4, a dynamic update model for the building's digital parameters is constructed, and the model formula is as follows: ； in, for The current state of the block storage; The block storage state at time t; A historical state retention coefficient, the value of which is related to the service life parameters of building components; The digital parameters of the building collected at time t and The difference in time parameters, including temperature changes and dynamic parameters of structural deformation; The hash function value for storing the state at time t.

4. The method according to claim 1, characterized in that, In the process of constructing the security assessment model in step S5, the following security assessment value calculation model is constructed by combining the feature vector output by the graph convolutional neural network and the historical parameters stored in the blockchain: ； Where E is the safety assessment value; The weighting coefficient for the m-th digital building parameter is determined based on the safety importance level of the component. This refers to the m-th parameter value collected in real time. The historical average value of the m-th parameter stored in the blockchain; is the deviation influence coefficient of the m-th parameter, which is related to the seismic resistance grade parameters of the building structure; p is the total number of digital building parameters involved in the evaluation; During the smart contract execution process in step S6, the preset threshold setting and the weighting coefficients in the evaluation value calculation model are... Establish a connection.

5. The method according to claim 1, characterized in that, During the initial information block construction in step S1, construction stage parameters of the building BIM model are introduced as the basis for block grouping. The digital parameters of the building at different construction stages are divided into different sub-blocks, and the different sub-blocks are linked through the following model formula: ； in, The strength of the association between sub-blockchain a and sub-blockchain b; Let c be the standardized value of the c-th building digitization parameter in sub-blockchain a; Let c be the standardized value of the building digitization parameter in sub-blockchain b; This is an XOR operation; q represents the time overlap coefficient of the construction stages corresponding to sub-blockchains a and b; q represents the number of parameters used for association calculation. In the feature extraction process of step S2, different convolution kernels of different sizes are used for the parameters of different sub-blockchains. The size of the convolution kernel is proportional to the number of building components contained in the sub-blockchain.

6. The method according to claim 1, characterized in that, In the smart contract module of step S6, the execution strength of the permission adjustment operation is determined by the following model formula: ； Where P is the execution strength parameter for permission adjustment, and its value range is... is a proportionality coefficient, which is related to the safety level parameters of the building project; E is the safety assessment value obtained in step S5. This is the preset maximum safety assessment value; The importance score is assigned to the d-th affected building component, and the score is determined based on the component's functional and structural parameters; r is the total number of affected building components; In the consensus verification of step S3, this execution strength parameter is used as an additional reference indicator for node verification. When the execution strength parameter is greater than 0.6, an additional node secondary verification process is initiated.

7. The method according to claim 1, characterized in that, Step S3 includes the following sub-steps: Step S3.1: Collect feature vectors processed by graph convolutional neural networks from different nodes, perform preliminary verification of the dimension and data type of the feature vectors, and remove feature vectors with mismatched dimensions or incorrect data types. Step S3.2: Determine the current verification difficulty benchmark value based on the construction progress parameters. The larger the progress parameters, the higher the benchmark value. Then, dynamically adjust the verification difficulty coefficient in combination with the number of feature vectors to make the verification speed match the amount of data. Step S3.3: The improved proof-of-work algorithm is used to perform hash operation on the feature vector. The operation result is required to meet the preset leading zero number condition. The leading zero number is determined by the adjusted verification difficulty coefficient. Each node performs the operation independently and submits the result. Step S3.4: Summarize and compare the calculation results submitted by different nodes, select the node that first meets the conditions and has the correct result as the accounting node, write the node's identification information and verification result into a temporary block, and wait for subsequent connection with other blockchains.

8. The method according to claim 1, characterized in that, Step S4 includes the following sub-steps: Step S4.1: Classify the building digitization parameters corresponding to the feature vector confirmed by the accounting node according to the spatial location of the building components. Component parameters in the same spatial area are grouped into the same data group for easy subsequent query and management. The classification is based on the three-dimensional coordinate parameters in the building BIM model. Step S4.2: Perform hash calculation on the parameters in each data group to generate a unique hash value for each parameter. Then, perform a second hash operation on the hash values ​​of all parameters in the same data group to obtain the Merkle leaf node values ​​of that data group. Step S4.3: According to the construction rules of the Merkle tree, the leaf node values ​​are hashed upwards layer by layer to generate non-leaf node values ​​until the root node hash value is obtained. The root node hash value is stored in the block header as the unique identifier of the data group, simplifying the process of verifying the association between blocks. Step S4.4: Combine the information including the root node hash value, data group parameters, the previous block hash value, and timestamp into a new block, and broadcast it to all nodes through the peer-to-peer network. After receiving the block, different nodes verify the validity of the block information. If the verification is successful, they add it to the end of their local blockchain.

9. The method according to claim 1, characterized in that, Step S5 includes the following sub-steps: Step S5.1: Extract the digital parameters of historical buildings related to the current assessment object from the blockchain, calculate the average value and fluctuation range of different parameters, establish a historical parameter benchmark library, and keep the update cycle of the benchmark library consistent with the maintenance cycle parameters of building components; Step S5.2: Collect the current parameters of the building BIM model in real time, extract the feature vector of the current parameters through graph convolutional neural network, calculate the difference between the feature vector of the corresponding parameter in the historical parameter benchmark library, and obtain the parameter change matrix. Step S5.3: Input the parameter change matrix into the safety assessment model. The model calculates the safety assessment value based on the weight coefficients and deviation influence coefficients of different parameters. The weight coefficients and deviation influence coefficients are retrieved from the attribute database of the building BIM model. Step S5.4: Compare the calculated security assessment value with the preset threshold to generate a comparison result. The comparison result includes the specific parameters and the extent to which the assessment value exceeds or falls below the threshold, providing a basis for subsequent smart contract execution.

10. A blockchain-based digital information security management system for buildings, characterized in that: include: The distributed acquisition and preprocessing unit for building BIM parameters has its input end connected to multiple building information acquisition terminals and its output end connected to the input end of the graph convolutional neural network feature extraction unit. It is used to acquire and preliminarily process different parameters of the building BIM information model. The graph convolutional neural network feature extraction unit has its output connected to the input of the blockchain consensus verification unit and the security assessment model construction unit, respectively, and is used to extract features from building digitization parameters and generate feature vectors. The blockchain consensus verification unit, whose output is connected to the input of the chain storage unit, is used to verify the feature vector and determine the accounting node. The chained storage unit, with its bidirectional connection to the security assessment model construction unit, is used to store the digital parameters of the building structure and related verification information in a preset structure. The security assessment model building unit connects its output to the input of the smart contract execution unit. It is used to build a model based on parameters stored on the blockchain and real-time parameters and output an assessment value. The smart contract execution unit, whose output is connected to the input of the blockchain consensus verification unit, is used to perform corresponding operations based on the evaluation value and feed the operation record back to the blockchain system.

Images