Blockchain-based project quality inspection batch traceability management method

Abstract

The application discloses a construction quality inspection batch traceability management method based on a blockchain storage record, and belongs to the technical field of construction quality traceability management. The method comprises the following steps: acquiring construction process multi-source monitoring data of a current inspection batch of a target construction project, and constructing an initial digital twin mirror image. By performing deviation analysis on the mirror image and a standard construction specification, a key quality feature node that needs to be stored is accurately positioned. The node feature data is encapsulated into traceability data blocks in timestamp order, the content hash value of the traceability data blocks is calculated, and the content hash value is associated with the block header hash value of a previous inspection batch to calculate the block header hash value of the current inspection batch, thereby forming an unalterable blockchain storage record data chain. Quality traceability is inquired based on the data chain, and a verification result is output. The application combines digital twin visualization mapping and the anti-tampering characteristics of a blockchain, realizes accurate positioning, credible storage and efficient tracing of construction quality, and enhances the transparency of construction quality management and the data credibility.

Description

Technical Field

[0001] This invention relates to the field of engineering quality traceability management technology, specifically to a method for traceability management of engineering quality inspection batches based on blockchain-based evidence storage. Background Technology

[0002] In the management of engineering quality inspection batches, the authenticity, completeness, and traceability of construction quality data are fundamental to ensuring the overall quality of the project. Existing engineering quality traceability management technologies typically use a centralized database to record construction testing data. Quality inspectors manually fill in the values ​​from on-site sampling onto paper forms or enter them into an electronic system, creating independent inspection batch record files. These files are stored on a single server indexed by batch numbers, and during quality traceability, the corresponding records are retrieved and compared by searching the batch number.

[0003] Existing technical solutions have shortcomings. In a centralized storage model, inspection batch data is stored on a server controlled by a single organization, making it vulnerable to subsequent tampering or loss. Furthermore, it's difficult to prove the originality of data in the event of quality disputes. Data exists as isolated files across multiple inspection batches, failing to form a tamper-proof chain. Replacing data in one batch does not affect others, rendering the starting point of the traceability chain unreliable. At the traceability verification level, existing methods only compare the values ​​of the inspection batch under investigation with standard specifications, failing to verify the completeness of the batch's historical evolution throughout the construction process or detect whether data was partially replaced during storage or transmission.

[0004] In the application of combining digital monitoring with blockchain evidence storage, two key issues need to be addressed. First, the volume of multi-source monitoring data generated by sensors at construction sites is enormous. If all monitoring data is stored on the blockchain indiscriminately, it will consume significant storage resources and consensus time. Relying solely on manual annotation of evidence nodes is extremely inefficient and dependent on personal experience. How to automatically identify key feature nodes that are crucial to quality assessment within a digital twin mirror for selective evidence storage is an unresolved issue in this field. Second, in a chain-based evidence storage structure, the strength of the association between blocks in each inspection batch determines the difficulty of tampering. The conventional method of directly embedding the hash value of the previous block header into the current block header has collision resistance limited by a single hash algorithm. If this algorithm is compromised in the future, the tamper-proof nature of the entire chain will collapse. How to improve the cryptographic strength of the association between blocks without significantly increasing computational overhead is another technical challenge for enhancing traceability credibility. Summary of the Invention

[0005] The purpose of this invention is to provide a blockchain-based method for traceability management of engineering quality inspection batches, which solves the problems of automatically locating evidence storage nodes in a digital twin mirror and enhancing the cryptographic strength of inter-block associations, thereby achieving reliable traceability management of construction quality data.

[0006] To achieve the above objectives, this invention provides the following technical solution: This invention provides a method for traceability management of engineering quality inspection batches based on blockchain evidence storage, comprising: acquiring multi-source monitoring data of the construction process of the current inspection batch in the target project, and constructing an initial digital twin image of the current inspection batch based on the multi-source monitoring data of the construction process; locating key quality feature nodes to be stored in the initial digital twin image according to the deviation analysis results between the initial digital twin image and standard construction specification data; encapsulating the feature data corresponding to the key quality feature nodes into traceability data blocks in timestamp order, and calculating the content hash value of the traceability data blocks; associating the content hash value with the block header hash value of the previous inspection batch to generate the block header hash value of the current inspection batch, forming a blockchain evidence storage data chain; performing quality traceability query on the current inspection batch based on the block header hash value of the current inspection batch in the blockchain evidence storage data chain, and outputting the traceability verification result of the current inspection batch. This method combines digital twins with blockchain evidence storage, making construction quality deviation data tamper-proof after being uploaded to the chain, achieving full-chain traceability.

[0007] As a preferred technical solution of the present invention, acquiring multi-source monitoring data of the construction process and constructing an initial digital twin image specifically includes: using multiple sensors deployed at the construction site to collect structural geometric dimension data, measured material strength data, environmental temperature and humidity data, and construction machinery operation parameter data of the current inspection batch; after time synchronization calibration, mapping these multi-source monitoring data to the corresponding spatial coordinate positions of a preset three-dimensional building information model; in the three-dimensional building information model, setting attribute parameter binding interfaces for each component in advance, and establishing a one-to-one mapping relationship between the interface and the component's geometric dimension attributes, material strength attributes, and construction process attributes; assigning the mapped data to the corresponding attributes through the interface, driving the corresponding components in the three-dimensional building information model to undergo three-dimensional geometric deformation updates and material attribute updates, generating the initial digital twin image, thereby accurately reproducing the actual construction state.

[0008] The implementation method for deviation analysis and key quality feature node location optimization is as follows: The standard component size range, standard material strength lower limit, and standard construction process parameter interval values ​​corresponding to the current inspection batch are extracted from a pre-set standard construction specification database and used as standard construction specification data. The actual component size, actual material strength, and actual construction process parameter values ​​of each component node in the initial digital twin mirror are compared item by item with the above standard data to obtain the size deviation, strength deviation, and process deviation values ​​of each component node. Each individual deviation value is normalized and weighted according to its corresponding weight coefficient to obtain a comprehensive deviation score for each component node. When the comprehensive deviation score exceeds a pre-set global deviation threshold, the corresponding component node is marked as a key quality feature node to be certified, and its spatial location identifier and deviation type identifier are recorded. Preferably, the weight coefficients corresponding to each deviation value are dynamically determined using the entropy weight method, objectively allocating weights based on the dispersion of each attribute data in the current inspection batch, making the identification of key quality feature nodes more consistent with actual quality fluctuations.

[0009] When generating the traceability data block, the node number, node spatial coordinates, deviation type identifier, deviation value, and corresponding collection timestamp of the key quality feature nodes in the initial digital twin image are extracted as feature data. According to the order of collection timestamps, the node number, node spatial coordinates, deviation type identifier, and deviation value are arranged into a data field sequence, and start and end markers are added to the beginning and end of the sequence, respectively, to form the traceability data block. All binary data of the traceability data block are concatenated and spliced, and a secure hash algorithm is applied to calculate the content hash value of the spliced ​​binary data string to obtain a fixed-length content hash value, thereby ensuring the integrity of the feature data.

[0010] When forming a blockchain-based evidence storage data chain, the block header hash value of the previous inspection batch is read from the local storage node of the blockchain. If the current inspection batch is the starting inspection batch, the preset initial hash root value is used as the block header hash value of the previous inspection batch. The binary representation sequence of the content hash value is concatenated with the binary representation sequence of the block header hash value of the previous inspection batch according to a preset alternating interleaving rule to obtain a concatenated hash value. The concatenated hash value is subjected to a first hash function operation to obtain a first intermediate hash value, and then subjected to a second hash function operation to obtain a second intermediate hash value. Different hash algorithms are used for the two operations. The first half of the bit string of the first intermediate hash value and the second half of the bit string of the second intermediate hash value are extracted and cross-merged to generate the block header hash value of the current inspection batch. The block header hash value of the current inspection batch, the content hash value, and the block header hash value of the previous inspection batch are stored in the blockchain distributed ledger in order of block height to form a blockchain-based evidence storage data chain. By employing double hashing and cross-segmentation operations, the hash collision resistance is enhanced, effectively preventing the forgery of inspection batch data through collision attacks.

[0011] When performing quality traceability queries, the system receives the header hash value of the current inspection batch as input from an external source. It then locates the matching target block in the blockchain distributed ledger, extracts the content hash value of the current inspection batch and the block header hash value of the previous inspection batch from the target block, and recursively traces forward along the blockchain's evidence storage data chain based on the previous block header hash value. This process sequentially obtains the historical block header hash values ​​and historical content hash values ​​of all inspection batches prior to the current batch, forming a complete historical hash traceability path. The system calculates the binary Hamming distance between the k-th level historical block header hash value and the corresponding block header hash value stored in the target block. If the Hamming distance at all levels is less than a preset tolerance threshold, the traceability verification result for the current inspection batch is output, and a timestamp sequence for each inspection batch is generated as a traceability time chain. If the Hamming distance at any level exceeds the tolerance threshold, subsequent comparisons are stopped, an abnormal traceability verification result is output, and the corresponding inspection batch number is marked as an abnormal traceability point. By using Hamming distance for hierarchical comparison, the location of content tampering can be accurately identified while tolerating a small number of bit flips, ensuring the reliability and accuracy of traceability verification.

[0012] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0013] An initial digital twin image is constructed by acquiring multi-source monitoring data of the current inspection batch of the target project. Based on the deviation analysis results between the initial digital twin image and standard construction specification data, key quality feature nodes to be certified are located. The feature data corresponding to the key quality feature nodes are encapsulated into traceability data blocks, and their content hash values ​​are calculated. This method automatically calculates the comprehensive deviation score of each component node by comparing the standard construction specification data with the actual monitoring data in the digital twin image item by item. It then uses a deviation threshold to filter out key nodes that have a real impact on quality assessment, replacing the manual process of judging the objects to be certified. In scenarios where multi-source sensors continuously generate massive amounts of monitoring data, only nodes with decisive differences in quality judgment and their deviation characteristics are included in the certification scope. This avoids the storage expansion and consensus delay caused by indiscriminately uploading all monitoring data to the blockchain. Simultaneously, the uploaded data directly points to the specific component location and deviation type that does not conform to the standard specifications, reducing the interference of redundant information on traceability queries.

[0014] When calculating the association between the content hash value and the block header hash value of the previous checksum, the concatenated hash value is processed twice using different hash functions, and the intersection segment is taken to generate the block header hash value of the current checksum. Block association no longer relies on a single hash algorithm; instead, two different hash algorithms are used sequentially, and the final block header hash value is obtained by cross-merging the first half of the bit string from the previous operation with the second half of the bit string from the subsequent operation. If an attacker attempts to forge a checksum block while maintaining consistency in the subsequent block header hash values, they must simultaneously find collision bit strings for both different hash algorithms. Furthermore, the combination position of these collision bit strings is forcibly restricted to a specific intersection segment between the first and second halves, drastically reducing the collision search space and exponentially increasing the attack difficulty. This method, without increasing additional block storage fields or consensus burden, enhances the cryptographic binding strength between adjacent checksum blocks, ensuring that any tampering with historical checksum feature data will expose abnormal bit string differences during step-by-step verification. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0016] Figure 1 This is a flowchart of a blockchain-based method for traceability management of engineering quality inspection batches.

[0017] Figure 2 This is a flowchart of the multi-source monitoring data processing and digital twin image generation process during construction;

[0018] Figure 3It is a flowchart for construction quality deviation detection and key node marking based on digital twins and standard construction specifications;

[0019] Figure 4 This is a flowchart of the blockchain-based inspection batch data traceability and verification process;

[0020] Figure 5 It is a diagram showing the distribution of comprehensive deviation scores for component nodes and the identification of key quality characteristic nodes. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] See Figure 1 This invention provides a blockchain-based method for traceability management of engineering quality inspection batches, comprising: acquiring multi-source monitoring data of the construction process of the current inspection batch in the target project, and constructing an initial digital twin image of the current inspection batch based on the multi-source monitoring data of the construction process; locating key quality feature nodes to be stored in the initial digital twin image based on the deviation analysis results between the initial digital twin image of the current inspection batch and the standard construction specification data; encapsulating the feature data corresponding to the key quality feature nodes into traceability data blocks in timestamp order, and calculating the content hash value of the traceability data blocks; associating the content hash value with the block header hash value of the previous inspection batch to generate the block header hash value of the current inspection batch, forming a blockchain evidence storage data chain; performing quality traceability query on the current inspection batch based on the block header hash value of the current inspection batch in the blockchain evidence storage data chain, and outputting the traceability verification result of the current inspection batch.

[0023] Example 1

[0024] In specific implementation, please refer to Figure 2Various types of sensors deployed at the target project construction site collect data on the structural geometry, measured material strength, ambient temperature and humidity, and construction machinery operating parameters of the current inspection batch during construction. Specifically, structural geometry data is collected using laser rangefinders and total stations; measured material strength data is collected using rebound hammers and ultrasonic testing instruments; ambient temperature and humidity data is collected using temperature and humidity sensors; and construction machinery operating parameter data is collected using vibration sensors and load sensors installed on the construction machinery. The collected structural geometry, material strength, ambient temperature and humidity, and construction machinery operating parameter data are combined to form multi-source monitoring data for the construction process.

[0025] Time synchronization calibration is performed on the multi-source monitoring data during the construction process. In practice, the clock module built into each sensor is calibrated with a standard clock reference before data acquisition begins to obtain a fixed offset for each sensor. For each data point in the multi-source monitoring data during the construction process, its original acquisition timestamp is recorded as follows: subscript This is the index of the data point in the multi-source monitoring data during the construction process. The fixed offset corresponding to the sensor to which this data point belongs is denoted as... , It is the difference obtained by comparing the sensor clock with the NTP time server. The synchronized timestamp is calculated using the following formula. :

[0026]

[0027] in, Indicates the first The timestamps of the multi-source monitoring data points during the construction process after time synchronization and calibration. Indicates the first The original collection timestamps of multi-source monitoring data points during the construction process This represents a fixed offset between the sensor clock from which the data point originates and the standard clock reference. The value of this fixed offset is determined by the sensor clock deviation, typically within ±100 milliseconds. A synchronized timestamp is used. Replace the original acquisition timestamp to complete time synchronization calibration.

[0028] After time synchronization calibration, the multi-source monitoring data of the construction process is mapped to corresponding spatial coordinates in a pre-defined 3D building information model. In practice, the pre-defined 3D building information model contains digital representations of all components in the target project, with each component having a unique spatial coordinate range. For each set of multi-source monitoring data of the construction process, the installation coordinates of the acquisition source sensors in the 3D building information model are pre-calibrated as follows: The coordinates match the spatial coordinate range of the corresponding component. After completing the time synchronization calibration, the multi-source monitoring data of the construction process is written into the corresponding spatial coordinate nodes of the 3D building information model according to the calibrated installation position coordinates, realizing the spatial association between the data and the model components.

[0029] In the 3D Building Information Model (BIM), attribute parameter binding interfaces are pre-defined for each component. These interfaces are a set of data writing channels that establish one-to-one mappings with the corresponding component's geometric dimensions, material strength, and construction process attributes. Geometric dimensions include the component's length, width, height, and thickness; material strength includes the component's concrete compressive strength and steel yield strength; and construction process attributes include the vibration frequency, travel speed, and number of compaction passes of the construction machinery.

[0030] Structural geometric dimension data mapped to the 3D Building Information Model (BIM) is assigned to the geometric dimension attributes of the corresponding components via the attribute parameter binding interface. Measured material strength data mapped to the 3D BIM is assigned to the material strength attributes of the corresponding components via the attribute parameter binding interface. Construction machinery operation parameter data mapped to the 3D BIM is assigned to the construction process attributes of the corresponding components via the attribute parameter binding interface.

[0031] Based on the assigned geometric dimensions, material strength, and construction process attributes, the system drives deformation and material attribute updates in the corresponding 3D geometric model of the building information model (BIM). In practice, each component in the BIM is defined by a parametric geometry. The numerical values ​​of the geometric dimensions directly drive the recalculation of the vertex coordinates of the geometry, resulting in deformation updates. Material attribute updates adjust the color mapping of the component's surface material based on a comparison between the material strength attribute value and the standard lower limit of material strength. For example, when the material strength attribute is lower than the standard lower limit, the component's surface color is adjusted to a warning color. The BIM model after deformation and material attribute updates is used as the initial digital twin image of the current inspection batch.

[0032] Example 2

[0033] See Figure 3The system extracts the standard component size range, standard material strength lower limit, and standard construction process parameter range values ​​corresponding to the current inspection batch from a pre-set standard construction specification database. In practice, the standard construction specification database stores the mapping relationship between different inspection batch types and standard specification data. Based on the type identifier of the current inspection batch, an index matching is performed in the standard construction specification database to read the corresponding standard component size range, standard material strength lower limit, and standard construction process parameter range values. The standard component size range values ​​include the minimum and maximum allowable values ​​for component length, width, and height. The standard material strength lower limit values ​​include the minimum compressive strength of concrete and the minimum yield strength of steel reinforcement. The standard construction process parameter range values ​​include the minimum and maximum allowable values ​​for construction machinery vibration frequency and travel speed. The retrieved standard component size range values, standard material strength lower limit, and standard construction process parameter range values ​​are collectively used as the standard construction specification data.

[0034] The actual component dimensions, measured material strength, and actual construction process parameters of each component node in the initial digital twin image are compared item by item with the standard component size range, standard material strength lower limit, and standard construction process parameter interval, and the individual deviation of each component node is calculated. In specific implementation, for each component node in the initial digital twin image, the actual component dimensions, measured material strength, and actual construction process parameters are extracted from the attribute parameters of the initial digital twin image. The actual component dimensions include the actual length, actual width, and actual height; the measured material strength includes the actual concrete compressive strength and actual steel reinforcement yield strength; and the actual construction process parameters include the actual vibration frequency and actual travel speed.

[0035] The dimensional deviation is calculated as follows: when the actual length exceeds the length range defined by the minimum and maximum allowable length values ​​within the standard component's dimensional range, the absolute difference between the actual length and the nearest endpoint of the range is taken as the length deviation component; when the actual length is within the length range, the length deviation component is zero. The width and height deviation components are calculated in the same way. The sum of the length, width, and height deviation components yields the dimensional deviation value.

[0036] The strength deviation value is calculated as follows: Subtract the actual concrete compressive strength value from the minimum concrete compressive strength limit in the standard material strength lower limit. If the difference is positive, take that positive value as the concrete strength deviation component; otherwise, take zero. Similarly, subtract the actual steel reinforcement yield strength value from the minimum steel reinforcement yield strength limit in the standard material strength lower limit. If the difference is positive, take that positive value as the steel reinforcement strength deviation component; otherwise, take zero. Finally, sum the concrete strength deviation component and the steel reinforcement strength deviation component to obtain the total strength deviation value.

[0037] The process deviation value is calculated as follows: When the actual vibration frequency value exceeds the frequency range defined by the lower and upper allowable limits of the vibration frequency in the standard construction process parameter range, the absolute difference between the actual vibration frequency value and the nearest endpoint of the range is taken as the frequency deviation component; when the actual vibration frequency value is within the frequency range, the frequency deviation component is zero. The speed deviation component is calculated using the same method based on the actual travel speed value and the lower and upper allowable limits of travel speed. The frequency deviation component and the speed deviation component are summed to obtain the process deviation value.

[0038] The dimensional deviation, strength deviation, and process deviation are collectively used as the individual deviation values ​​for the corresponding component nodes.

[0039] The dimensional deviation, strength deviation, and process deviation values ​​are obtained from the individual deviations of each component node. These values ​​are then normalized to obtain normalized dimensional deviation, strength deviation, and process deviation values, respectively. In practice, a set of dimensional deviation values ​​containing all component node dimensional deviation values ​​is constructed. The maximum and minimum values ​​in this set are searched. For each component node, a range normalization method is applied, mapping the dimensional deviation value to the 0-1 interval by subtracting the minimum value and dividing by the difference between the maximum and minimum values, thus obtaining the normalized dimensional deviation value. The strength deviation and process deviation values ​​are normalized using the same method to obtain the normalized strength deviation and normalized process deviation values, respectively.

[0040] The normalized dimensional deviation, normalized strength deviation, and normalized process deviation values ​​are weighted and summed according to preset dimensional weighting coefficients, strength weighting coefficients, and process weighting coefficients to obtain the comprehensive deviation score for each component node. The dimensional weighting coefficient, strength weighting coefficient, and process weighting coefficient are preset according to the construction quality control requirements of the target project; all three weighting coefficients are greater than or equal to 0, and the sum of the three weighting coefficients is 1. The component node is calculated using the following formula. Comprehensive deviation score :

[0041]

[0042] in, Indicates the number is The comprehensive deviation score of the component nodes, It serves as a unique identifier for the component node in the initial digital twin image. This represents the size weighting coefficient, which is set to 0.4 in this embodiment. The reason for this setting is that size deviation is classified as a key control item in the engineering quality acceptance specifications and needs to be given a higher weight. This represents the strength weighting coefficient, which is set to 0.35 in this embodiment. The basis for this setting is that the strength deviation directly affects the structural safety. This represents the process weighting coefficient, which is set to 0.25 in this embodiment. The basis for this setting is that the impact of process deviation on quality is slightly lower than that of dimensional deviation and strength deviation. Indicates the number is The normalized dimensional deviation values ​​of the component nodes. Indicates the number is The normalized strength deviation value of the component nodes, Indicates the number is The normalized process deviation value of the component nodes.

[0043] The comprehensive deviation score of each component node is compared with a preset global deviation threshold. In practice, the preset global deviation threshold is set by construction quality inspectors according to the quality level requirements of the target project; in this embodiment, the global deviation threshold is set to 0.6. When the number is... Comprehensive deviation score of component nodes When the deviation exceeds the global deviation threshold of 0.6, it will be numbered as... The component nodes are marked as key quality feature nodes to be certified, and the node numbers of the key quality feature nodes are assigned. Record it in the list of nodes to be stored. When the number is Comprehensive deviation score of component nodes When the global deviation threshold is less than or equal to 0.6, the number is not assigned. The component nodes are marked as key quality feature nodes to be certified. In the initial digital twin image, the spatial location identifier and deviation type identifier of the key quality feature node are recorded. The spatial location identifier is taken from the center point coordinates of the corresponding spatial coordinate range of the key quality feature node in the 3D building information model. The deviation type identifier is determined based on the largest value among the normalized dimensional deviation value, normalized strength deviation value, and normalized process deviation value. When the largest value is the normalized dimensional deviation value, the deviation type identifier is set to dimensional deviation; when the largest value is the normalized strength deviation value, the deviation type identifier is set to strength deviation; and when the largest value is the normalized process deviation value, the deviation type identifier is set to process deviation.

[0044] See Figure 5 In the figure, the horizontal axis represents the component node number, ranging from 0 to 500, representing the unique identifier of the corresponding component node in the initial digital twin image; the vertical axis represents the comprehensive deviation score of each component node, ranging from 0 to above 1, reflecting the weighted comprehensive deviation degree of the component node based on size, strength, and process deviations. Hollow circles "○" in the figure represent the distribution of comprehensive deviation scores for non-critical quality feature nodes. The scattered points are concentrated between 0 and 0.6 on the vertical axis, generally low and relatively evenly distributed, indicating that the comprehensive deviation scores of most component nodes do not exceed the preset global deviation threshold.

[0045] In the figure, key quality feature nodes are marked with solid crosses "×". The comprehensive deviation scores of these nodes are all significantly higher than the dashed horizontal line of the global deviation threshold of 0.6, and their distribution ranges from approximately 0.6 to 1. Some key nodes have comprehensive deviation scores that reach or approach 1, indicating that their deviation is significant, which meets the marking conditions for key quality feature nodes with comprehensive deviation scores exceeding the threshold in Example 2. The key quality feature nodes are relatively dispersed in terms of node numbering and are not concentrated in a certain interval, indicating that the key nodes are relatively scattered in spatial coordinates, reflecting potential quality hazards in multiple components.

[0046] The dashed horizontal line corresponds to a global deviation threshold of 0.6 on the vertical axis, serving as the boundary between critical quality feature nodes and non-critical nodes. This threshold strictly distinguishes the deviation scores of the two types of nodes. The overall trend shows that as the component node number increases, the comprehensive deviation score does not exhibit significant monotonic change, and the distribution of the deviation score remains relatively stable, reflecting the matching of multi-source monitoring data and standard construction specification data, as well as the uniformity of deviation analysis throughout the inspection batch.

[0047] Example 3

[0048] In practical implementation, the entropy weight method is used to dynamically determine the size weight coefficient, strength weight coefficient, and process weight coefficient. The dimensional deviation values, strength deviation values, and process deviation values ​​of all component nodes in the current inspection batch in the initial digital twin image are obtained, and a matrix is ​​constructed using these values. lines and The evaluation matrix consists of columns, where The total number of component nodes. The evaluation indicators are, in order, dimensional deviation, strength deviation, and process deviation. The value is 3. The [number]th [unit] in the evaluation matrix. Line number The elements of a column are denoted as , Indicates the first The component node at the _ ... The original deviation values ​​of each evaluation indicator The value range is 1 to integers, The value of is an integer from 1 to 3. Corresponding dimensional deviation value, Corresponding strength deviation value, Corresponding process deviation value.

[0049] For each element in the evaluation matrix Nonnegation processing is performed because the dimensional deviation, strength deviation, and process deviation values ​​are all nonnegative. The nonnegation process directly converts these values ​​into their corresponding values. Remain unchanged. Perform column-wise normalization on the nonnegative evaluation matrix and calculate the... The component node at the _ ... The proportion of each evaluation indicator The following formula is used:

[0050]

[0051] in, Indicates the first The component node at the _ ... The value of the evaluation index accounts for the percentage of all component nodes in the [number]th [period]. The proportion of the total value under each evaluation indicator. Indicates the first The component node at the _ ... The original deviation values ​​of each evaluation indicator Indicates all The component node at the _ ... Sum the original deviation values ​​on each evaluation indicator. Calculate the... Information entropy value of each evaluation indicator Information entropy value The calculation uses the Shannon entropy formula, which... Substitute and multiply Perform normalization. When At that time, it was stipulated The value of is 0. According to the... Information entropy value of each evaluation indicator Calculate the first Coefficient of difference of each evaluation indicator Coefficient of difference For the coefficient of difference , , Summation is performed, and the difference coefficient is calculated. Dividing by the sum of the difference coefficients yields the first... The entropy weight coefficients of each evaluation indicator. The entropy weight coefficient of time is used as the size weight coefficient. The entropy weight coefficient of time is used as the intensity weight coefficient. The entropy weight coefficient is used as the process weight coefficient. The value of the entropy weight coefficient ranges from 0 to 1, and the sum of the three entropy weight coefficients is 1. Through the above process, the dynamically determined size weight coefficient, strength weight coefficient, and process weight coefficient are obtained.

[0052] The process involves obtaining the node number, spatial coordinates, deviation type identifier, deviation value, and corresponding acquisition timestamp of each key quality feature node in the initial digital twin image. In practice, the node number of each key quality feature node is read from the list of nodes to be certified. Based on this node number, the corresponding spatial coordinates are extracted from the initial digital twin image. These coordinates consist of three spatial dimensions. The deviation type identifier is obtained from the aforementioned marking process; dimensional deviation, strength deviation, and process deviation identifiers are represented by predefined enumerated values. The deviation value includes three components: dimensional deviation, strength deviation, and process deviation. The acquisition timestamp is the synchronized timestamp of the multi-source monitoring data corresponding to the deviation value. The node number, spatial coordinates, deviation type identifier, deviation value, and acquisition timestamp are collectively used as the feature data of the key quality feature node.

[0053] Based on the chronological order of the data collection timestamps, the node number, node spatial coordinates, deviation type identifier, and deviation value in the feature data are sequentially arranged into a data field sequence. In specific implementation, the feature data of all key quality feature nodes are sorted in ascending order of collection timestamps to obtain a sorted feature data list. For each entry in the feature data list, a data field sequence is constructed. The first field of the data field sequence contains the node number, the second field contains the three spatial dimensions of the node's spatial coordinates, the third field contains the deviation type identifier, and the fourth field contains the three components of the deviation value. A start marker is added to the beginning of the data field sequence, using the hexadecimal value "0x5A5A" as a two-byte identifier. An end marker is added to the end of the data field sequence, using the hexadecimal value "0xA5A5" as a two-byte identifier. The data field sequence encapsulated by the start and end markers constitutes a traceability data block, and each key quality feature node corresponds to an independent traceability data block.

[0054] All binary data in the traceability data block are concatenated to obtain a concatenated binary data string. In specific implementation, the start marker, the binary representations of the node number field, the node spatial coordinate field, the deviation type identifier field, the deviation value field, and the end marker in the traceability data block are concatenated sequentially according to their storage order to form a continuous binary data string. A secure hash algorithm, SHA-256, is applied to the concatenated binary data string. The binary data string is used as the input message for the SHA-256 algorithm, which performs message padding, block processing, and iterative compression function operations on the input message to generate a 256-bit fixed-length output value. This output value serves as the content hash value of the corresponding traceability data block.

[0055] Example 4

[0056] The block header hash value of the previous inspection batch is read from the local storage node of the blockchain. In practice, the local storage node of the blockchain maintains a block header hash value index table, which records the mapping relationship between the block number and the corresponding block header hash value of each completed inspection batch. The inspection batch number of the previous inspection batch is determined based on the inspection batch number of the current inspection batch. The inspection batch numbers are encoded as continuously increasing integers according to the construction sequence, and the inspection batch number of the previous inspection batch is the inspection batch number of the current inspection batch minus one. Using the inspection batch number minus one as the lookup key, the corresponding block header hash value is retrieved from the block header hash value index table, and the read block header hash value is the block header hash value of the previous inspection batch. If the current inspection batch is the starting inspection batch, that is, the inspection batch number of the current inspection batch is 1, then there is no record of a block header hash value record corresponding to the inspection batch number minus one. In this case, the preset initial hash root value is used as the block header hash value of the previous inspection batch. The initial hash root value is the 256-bit hash output value obtained by the SHA-256 algorithm from the input string "GENESIS".

[0057] The content hash value is concatenated with the block header hash value of the previous check batch using binary bit strings. In practice, the content hash value is represented as a 256-bit binary sequence, and the block header hash value of the previous check batch is also represented as a 256-bit binary sequence. The default alternation rule is as follows: the first bit of the binary sequence of the content hash value is used as the first bit of the concatenated data string; the first bit of the binary sequence of the block header hash value of the previous check batch is used as the second bit; the second bit of the binary sequence of the content hash value is used as the third bit; the second bit of the binary sequence of the block header hash value of the previous check batch is used as the fourth bit, and so on, alternating the corresponding bit values ​​from the two binary sequences until all 256 bits of each binary sequence have been used, resulting in a concatenated hash value containing 512 bits.

[0058] The concatenated hash value is subjected to a first hash function operation to obtain a first intermediate hash value. This first hash function operation uses the SHA-256 algorithm, taking the 512-bit concatenated hash value as input. After message padding, block processing, and compression iterative operations by the SHA-256 algorithm, a fixed-length 256-bit first intermediate hash value is output. The first intermediate hash value is then subjected to a second hash function operation to obtain a second intermediate hash value. This second hash function operation uses the SM3 cryptographic hash algorithm, taking the 256-bit first intermediate hash value as input. After message padding, block processing, and compression iterative operations by the SM3 cryptographic hash algorithm, a fixed-length 256-bit second intermediate hash value is output.

[0059] Extract the first half of the bit string of the first intermediate hash value and the second half of the bit string of the second intermediate hash value. In specific implementation, the first half of the bit string of the first intermediate hash value is a 128-bit binary sequence starting from the most significant bit, and the second half of the bit string of the second intermediate hash value is a 128-bit binary sequence starting from the 129th bit to the 256th bit. The first half of the bit string and the second half of the second intermediate hash value are cross-merged. The cross-merging is performed as follows: the first bit of the first half of the bit string is used as the first bit of the merged data string, the first bit of the second half of the second intermediate hash value is used as the second bit of the merged data string, the second bit of the first half of the bit string is used as the third bit of the merged data string, the second bit of the second half of the second intermediate hash value is used as the fourth bit of the merged data string, and so on. The corresponding bit values ​​in the two bit strings are used alternately until all 128 bits of the binary bits contained in the two bit strings have been used, resulting in a merged data string containing 256 binary bits. This 256-bit merged data string is the block header hash value of the current check batch.

[0060] The block header hash, content hash, and block header hash of the current inspection batch are associated and stored in the blockchain distributed ledger in ascending order of block height. In practice, the blockchain distributed ledger is maintained by multiple blockchain nodes, with each node running a block storage module. The block storage module allocates a block file on its local persistent storage medium. This block file contains multiple block records arranged in ascending order of block height. Each block record includes a block header and a block body. The block header contains a block height field, a block header hash field for the current inspection batch, and a block header hash field for the previous inspection batch. The block body contains a content hash field. The inspection batch number of the current inspection batch is used as the value of the block height field and written into the block header. The block header hash of the current inspection batch and the block header hash of the previous inspection batch are also written into the block header. The content hash is written into the block body. By using the reference relationship between the block header hash value of the previous check batch and the block header hash value of the current check batch of the previous block recorded in the block header, a blockchain evidence storage data chain is formed between the blocks of the entire blockchain distributed ledger.

[0061] Example 5

[0062] In specific implementation, please refer to Figure 4 It receives the hash value of the block header to be queried for the current inspection batch from external input. The hash value of the block header to be queried is provided by the quality supervision party or the engineering acceptance party through the traceability query terminal. The traceability query terminal establishes a network connection with any blockchain node in the blockchain distributed ledger, and encapsulates the hash value of the block header to be queried into a query request message and sends it to the blockchain node.

[0063] The target block is located in the blockchain distributed ledger based on the hash value of the block header to be queried. In practice, the block storage module of the blockchain node maintains a reverse index table of block header hash values. This table records the mapping relationship between the block header hash value of each stored block in the blockchain distributed ledger and its corresponding block height. Using the received hash value of the block header to be queried as the query key, an exact match is performed in the reverse index table. When a match is found, the block height value corresponding to the match is obtained. Based on this block height value, all data of the corresponding block is read from the block file of the blockchain node; the read block is the target block. When no match is found, a query failure response is returned.

[0064] Extract the content hash value of the current checksum and the block header hash value of the previous checksum from the target block. In specific implementation, the target block's block header contains the block header hash value fields for the current checksum and the previous checksum, while the target block's block body contains the content hash value field. The value of the previous checksum's block header hash value is read from the block header as the previous checksum's block header hash value, and the value of the content hash value field in the block body is read as the content hash value of the current checksum.

[0065] Based on the block header hash value of the previous verification batch stored in the target block, the blockchain evidence storage data chain is traced backward to sequentially obtain the historical block header hash values ​​and historical content hash values ​​of all verification batches preceding the current verification batch, forming a complete historical hash tracing path. In specific implementation, the extracted block header hash value of the previous verification batch is used as the current tracing pointer. The block whose block header hash value equals the tracing pointer is searched in the blockchain distributed ledger; this block is the block corresponding to the previous verification batch. The block header hash value field of the previous verification batch in the block header of this block is read to obtain the block header hash value of the next-next verification batch. Simultaneously, the content hash value field in the block body of this block is read to obtain the historical content hash value of the previous verification batch. The read block header hash value of the previous verification batch and the read content hash value are recorded as historical block header hash values. Then, the tracing pointer is updated to the block header hash value of the next-next verification batch, and the tracing continues forward until the tracing pointer equals the preset initial hash root value, at which point the tracing stops. The historical block header hash values ​​and historical content hash values ​​obtained sequentially during the tracing process are arranged in ascending order of proximity. Simultaneously, the block header hash value and content hash value of the target block's current verification batch are placed at the beginning of the sequence, forming a complete historical hash tracing path. Each entry in the complete historical hash tracing path consists of a tracing level number, verification batch number, historical block header hash value, historical content hash value, and a corresponding collection timestamp. The collection timestamp is extracted from the time information associated with the source data block stored in the block body of the corresponding block.

[0066] The hash values ​​of the k-th level historical block headers are extracted sequentially from the complete historical hash tracing path, where the value of k decreases from the level before the current verification batch to the starting verification batch. Simultaneously, the storage block header hash value corresponding to the k-th level historical block header hash value is extracted from the target block. In specific implementation, the block body of the target block also stores a traceability verification hash list, which is constructed synchronously when generating the block of the current verification batch. The traceability verification hash list records the block header hash values ​​of each verification batch from the starting batch to the current batch in ascending order of verification batch number. The verification batch number corresponding to the k-th level historical block header hash value is the current verification batch number minus k. Using this verification batch number as an index, the corresponding storage block header hash value is located in the traceability verification hash list of the target block. The value read from the traceability verification hash list is the storage block header hash value corresponding to the k-th level historical block header hash value.

[0067] Calculate the binary Hamming distance between the hash value of the k-th level historical block header and the hash value of the storage block header. The binary Hamming distance is defined as the total number of positions in two equal-length binary sequences where corresponding bits have different values. In practice, the hash value of the k-th level historical block header is represented as a 256-bit binary sequence, and the hash value of the storage block header is also represented as a 256-bit binary sequence. The values ​​of the corresponding bits in the two sequences are compared bit by bit. When two bits at the same position are different, the difference count is incremented by one. The difference count obtained after traversing all 256 bits is the binary Hamming distance.

[0068] The binary Hamming distance is compared with a preset tolerance threshold. The preset tolerance threshold is set to 0. This is based on the fact that the blockchain evidence storage data chain uses SHA-256 and SM3 cryptographic hash algorithms for hash generation. Both cryptographic hash algorithms have collision resistance; any tampering with the evidence storage data will fundamentally change the generated hash value. Therefore, setting the tolerance threshold to 0 can most strictly guarantee the data integrity verification requirements. When the binary Hamming distance is less than the tolerance threshold (i.e., the binary Hamming distance equals 0), the k-th level comparison result is considered consistent. When the binary Hamming distance is greater than or equal to the tolerance threshold (i.e., the binary Hamming distance is greater than 0), the k-th level comparison result is considered inconsistent, and the inspection batch number corresponding to the current level is recorded. The inspection batch number is the current inspection batch number minus k.

[0069] Following the order of k decreasing from the previous level of the current inspection batch to the starting inspection batch, the above comparison process is sequentially performed on all levels of the complete historical hash tracing path. When the comparison results of all levels are determined to be consistent, the traceability verification pass result for the current inspection batch is generated, and the timestamp sequence of each inspection batch in the complete historical hash tracing path is output as the traceability time chain. The traceability time chain is generated as follows: in ascending order of inspection batch number, the collection timestamp corresponding to each inspection batch is extracted sequentially from the complete historical hash tracing path, these collection timestamps are arranged into a time list, and the time list is appended to the traceability verification pass result as the traceability time chain and output together.

[0070] When any level of comparison results indicates inconsistency, the comparison process at subsequent levels is halted. An anomaly result for the current inspection batch is generated, and the recorded inspection batch number corresponding to the current level is output as the anomaly traceability point. The anomaly traceability point identifies the earliest location of data inconsistency in the blockchain-based evidence storage data chain. Quality regulators can locate the specific construction batch and quality data based on the inspection batch number corresponding to the anomaly traceability point.

[0071] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for traceability management of engineering quality inspection batches based on blockchain-based evidence storage, characterized in that, Includes the following steps: Acquire multi-source monitoring data of the construction process of the current inspection batch in the target project, and construct an initial digital twin image of the current inspection batch based on the multi-source monitoring data of the construction process; Based on the deviation analysis results between the initial digital twin image of the current inspection batch and the standard construction specification data, the key quality feature nodes to be certified are located in the initial digital twin image. The feature data corresponding to the key quality feature nodes are encapsulated into traceability data blocks in timestamp order, and the content hash value of the traceability data blocks is calculated. The content hash value is correlated with the block header hash value of the previous inspection batch to generate the block header hash value of the current inspection batch, thus forming a blockchain evidence storage data chain. Based on the block header hash value of the current inspection batch in the blockchain evidence storage data chain, a quality traceability query is performed on the current inspection batch, and the traceability verification result of the current inspection batch is output.

2. The method for traceability management of engineering quality inspection batches based on blockchain evidence storage according to claim 1, characterized in that, Acquire multi-source monitoring data of the construction process of the current inspection batch in the target project, and construct an initial digital twin image of the current inspection batch based on the multi-source monitoring data of the construction process, specifically including: By deploying various types of sensors at the construction site of the target project, the structural geometric dimensions, measured material strength, ambient temperature and humidity, and construction machinery operating parameters of the current inspection batch are collected during the construction process. The structural geometric dimensions, measured material strength, ambient temperature and humidity, and construction machinery operating parameters are used together as multi-source monitoring data for the construction process. The multi-source monitoring data of the construction process is time-synchronized and calibrated, and the multi-source monitoring data of the construction process after time synchronization calibration is mapped to the corresponding spatial coordinate position in the preset three-dimensional building information model; Based on the multi-source monitoring data of the construction process mapped into the three-dimensional building information model, the attribute parameters of the corresponding components in the three-dimensional building information model are driven to change in a linked manner, generating the initial digital twin image of the current inspection batch.

3. The method for traceability management of engineering quality inspection batches based on blockchain evidence storage according to claim 2, characterized in that, Based on the deviation analysis results between the initial digital twin image of the current inspection batch and the standard construction specification data, the key quality characteristic nodes to be certified are located in the initial digital twin image, specifically including: Extract the standard component size range, standard material strength lower limit, and standard construction process parameter interval values ​​corresponding to the current inspection batch from the preset standard construction specification database, and use the standard component size range, standard material strength lower limit, and standard construction process parameter interval values ​​together as the standard construction specification data; The actual component size value, actual material strength measured value, and actual construction process parameter value of each component node in the initial digital twin mirror are compared with the standard component size range value, the standard material strength lower limit value, and the standard construction process parameter interval value, respectively. The individual deviation of each component node is calculated to obtain the size deviation value, strength deviation value, and process deviation value. Based on the individual deviation of each component node, a comprehensive deviation score for each component node is calculated using a multi-attribute comprehensive evaluation method. Component nodes whose comprehensive deviation scores exceed a preset deviation threshold are marked as key quality feature nodes to be certified. The spatial location identifier and deviation type identifier of the key quality feature nodes are recorded in the initial digital twin image.

4. The method for traceability management of engineering quality inspection batches based on blockchain evidence storage according to claim 3, characterized in that, The multi-attribute comprehensive evaluation method uses the entropy weight method to dynamically determine the weight coefficients corresponding to the dimensional deviation value, the strength deviation value, and the process deviation value.

5. The method for traceability management of engineering quality inspection batches based on blockchain evidence storage according to claim 3, characterized in that, The feature data corresponding to the key quality feature nodes are encapsulated into traceability data blocks in timestamp order, and the content hash value of the traceability data blocks is calculated, specifically including: Extract the node number, node spatial coordinates, deviation type identifier, deviation value, and the collection timestamp corresponding to the deviation value of the key quality feature node in the initial digital twin image, and use the node number, node spatial coordinates, deviation type identifier, deviation value, and collection timestamp together as the feature data; According to the order of the collection timestamps, the node number, node spatial coordinates, deviation type identifier and deviation value in the feature data are arranged into a data field sequence, and a start marker and an end marker are added to the beginning and end of the data field sequence respectively to form the traceability data block; All binary data in the traceability data block are concatenated to obtain a concatenated binary data string. A secure hash algorithm is then applied to the concatenated binary data string to generate a fixed-length content hash value.

6. The method for traceability management of engineering quality inspection batches based on blockchain evidence storage according to claim 5, characterized in that, The content hash value is correlated with the block header hash value of the previous inspection batch to generate the block header hash value of the current inspection batch, forming a blockchain evidence storage data chain, specifically including: Read the block header hash value of the previous batch from the local storage node of the blockchain. If the current batch is the starting batch, use the preset initial hash root value as the block header hash value of the previous batch. The content hash value is concatenated with the block header hash value of the previous inspection batch to obtain the concatenated hash value. The intersection segment of the concatenated hash value is then obtained after performing two different hash function operations on the concatenated hash value to generate the block header hash value of the current inspection batch. The block header hash value of the current inspection batch, the content hash value, and the block header hash value of the previous inspection batch are associated and stored in the blockchain distributed ledger in order of block height, forming the blockchain evidence storage data chain.

7. The method for traceability management of engineering quality inspection batches based on blockchain evidence storage according to claim 6, characterized in that, Based on the block header hash value of the current inspection batch in the blockchain evidence storage data chain, a quality traceability query is performed on the current inspection batch, and the traceability verification result of the current inspection batch is output, specifically including: Receive the header hash value of the block to be queried in the current inspection batch from external input, locate the matching target block in the blockchain distributed ledger according to the header hash value of the block to be queried, and extract the content hash value of the current inspection batch and the block header hash value of the previous inspection batch from the target block; Based on the block header hash value of the previous inspection batch stored in the target block, trace back along the blockchain evidence storage data chain to sequentially obtain the historical block header hash value and historical content hash value of all inspection batches before the current inspection batch, forming a complete historical hash tracing path; The hash values ​​of the headers of each historical block in the complete historical hash tracing path are compared step by step with the hash values ​​of the headers of the corresponding positions in the target block. When all the step-by-step comparison results are consistent, the traceability verification result of the current inspection batch is output. When there are inconsistent step-by-step comparison results, the traceability verification result of the current inspection batch is output, and the inspection batch number corresponding to the inconsistent position is marked.

8. The method for traceability management of engineering quality inspection batches based on blockchain evidence storage according to claim 2, characterized in that, Based on the multi-source monitoring data of the construction process mapped into the 3D building information model, the attribute parameters of the corresponding components in the 3D building information model are driven to change in a linked manner, generating an initial digital twin image of the current inspection batch, specifically including: In the three-dimensional building information model, attribute parameter binding interfaces are pre-defined for each component, and a one-to-one mapping relationship is established between the attribute parameter binding interfaces and the corresponding component's geometric dimension attributes, material strength attributes, and construction process attributes. The structural geometric dimension data, the measured material strength data, and the construction machinery operation parameter data mapped to the three-dimensional building information model are respectively assigned to the geometric dimension attribute, the material strength attribute, and the construction process attribute of the corresponding component through the attribute parameter binding interface; Based on the geometric dimension attribute, material strength attribute, and construction process attribute after the assignment is completed, the three-dimensional geometric model of the corresponding component in the three-dimensional building information model is driven to undergo deformation update and material attribute update, and the three-dimensional building information model after deformation update and material attribute update is used as the initial digital twin mirror.

9. The method for traceability management of engineering quality inspection batches based on blockchain evidence storage according to claim 3, characterized in that, Based on the individual deviation of each component node, a comprehensive deviation score is calculated for each component node using a multi-attribute comprehensive evaluation method. Component nodes whose comprehensive deviation scores exceed a preset deviation threshold are marked as key quality feature nodes to be certified, specifically including: Obtain the dimensional deviation value, strength deviation value, and process deviation value from the individual deviation values ​​of each component node, and normalize the dimensional deviation value, strength deviation value, and process deviation value respectively to obtain normalized dimensional deviation value, normalized strength deviation value, and normalized process deviation value. The normalized dimensional deviation value, the normalized strength deviation value, and the normalized process deviation value are weighted and summed according to preset dimensional weight coefficients, strength weight coefficients, and process weight coefficients to obtain the comprehensive deviation score of each component node. The comprehensive deviation score of each component node is compared with a preset global deviation threshold. When the comprehensive deviation score is greater than the global deviation threshold, the corresponding component node is marked as a key quality feature node to be certified, and the node number of the key quality feature node is recorded in the list of nodes to be certified.

10. The method for traceability management of engineering quality inspection batches based on blockchain evidence storage according to claim 6, characterized in that, The content hash value is concatenated with the block header hash value of the previous batch to obtain a concatenated hash value. The concatenated hash value is then subjected to two different hash function operations, and the intersection segment is taken to generate the block header hash value of the current batch. Specifically, this includes: The binary representation sequence of the content hash value and the binary representation sequence of the block header hash value of the previous batch are concatenated according to a preset alternating interleaving rule to obtain the concatenated hash value. The concatenated hash value is subjected to a first hash function operation to obtain a first intermediate hash value, and the first intermediate hash value is subjected to a second hash function operation to obtain a second intermediate hash value, wherein the first hash function operation and the second hash function operation use different hash algorithms; Extract the first half of the bit string of the first intermediate hash value and the second half of the bit string of the second intermediate hash value, and cross-merge the first half of the bit string and the second half of the bit string to generate the block header hash value of the current inspection batch.

Images