Metal mine data knowledge base management system and method based on multi-party participation

Abstract

The application discloses a metal mine data knowledge base management system and method based on multi-party participation, relates to the technical field of mine engineering informatization management, and comprises a standardized multi-party engineering participant data set, an engineering graph structure constructed through a graph neural network, and a standardized engineering knowledge subgraph set divided; a quantum register is initialized through the standardized engineering knowledge subgraph set, a quantum fingerprint is acquired, a difference value of a building information model quantum fingerprint and a field construction record quantum fingerprint is acquired, and a verification safety engineering knowledge subgraph set carrying a timestamp is acquired; a four-dimensional engineering tensor is constructed according to the verification safety engineering knowledge subgraph set carrying the timestamp, a joint conflict score is calculated, a conflict coordinate set carrying a WBS code and a coordinate index is acquired, multi-source confidence weight is calculated based on the conflict coordinate set carrying the WBS code and the coordinate index, and a fusion scheme set is generated. The application realizes automatic identification and resolution of conflicts and improves data credibility.

Description

Technical Field

[0001] This invention relates to the field of information management technology for mining engineering, and in particular to a data knowledge base management system and method for metal mines based on multi-party participation. Background Technology

[0002] With the development of large-scale and intelligent mining in metal mines, the lifecycle management of mining engineering projects has gradually expanded from simple production scheduling to a comprehensive collaboration encompassing multiple stages such as investment planning, design schemes, construction organization, cost management, quality inspection, and supervision and evaluation. In recent years, with the introduction of advanced information technologies such as Building Information Modeling (BIM), 3D point cloud mapping, knowledge graphs, and graph neural networks, mining engineering data has gradually moved from a single dimension to multi-source heterogeneous integration, forming a data-driven engineering knowledge management model. Simultaneously, emerging technologies such as blockchain, quantum computing, and multimodal deep learning have shown potential in the reliable storage of engineering data, complex conflict detection, and multi-source decision support, especially in improving data consistency verification, security assurance, and conflict resolution efficiency, demonstrating strong application value.

[0003] Current metal mine engineering data management primarily relies on traditional project management software or one-way database archiving, which has two main shortcomings: First, data consistency verification methods are simplistic, generally employing static comparison methods based on hashing or signatures. This fails to effectively address dynamic differences between field-measured data and design models, easily leading to insufficient data reliability and broken traceability chains. Second, conflict detection and resolution mechanisms are relatively outdated, relying heavily on manual judgment or localized detection based on geometric models. This lack of comprehensive consideration of multi-dimensional factors such as schedule, cost, quality, and spatial location results in a high false alarm rate and low processing efficiency. These shortcomings not only affect the real-time performance and precision of mine engineering management but also restrict the efficiency and transparency of collaborative decision-making involving multiple parties. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a multi-party participation-based method for managing metal mine data knowledge bases to address the problems of insufficient data consistency verification and incomplete conflict detection and resolution in existing metal mine engineering data knowledge base management technologies.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: This invention provides a method for managing a metal mine data knowledge base based on multi-party participation, which includes: Collect datasets from multiple project participants, preprocess them, and generate standardized datasets from these participants. Based on a standardized dataset of multiple engineering participants, an engineering graph structure is constructed using a graph neural network and divided into a set of standardized engineering knowledge subgraphs. The quantum register is initialized by a standardized engineering knowledge subgraph set, the quantum fingerprint is obtained, the difference between the quantum fingerprint of the building information model and the quantum fingerprint of the on-site construction record is obtained, and the verification safety engineering knowledge subgraph set with timestamp is obtained. Based on the verification security engineering knowledge subgraph with timestamps, a four-dimensional engineering tensor is constructed, a joint conflict score is calculated, and a set of conflict coordinates with WBS encoding and coordinate index is obtained. Based on the set of conflicting coordinates with WBS encoding and coordinate index, multi-source confidence weights are calculated to generate a set of fusion schemes; Based on the fusion scheme set, multimodal spatiotemporal convolution kernels are used to perform restricted registration and local reconstruction of building information models and field measured point cloud data at conflict coordinates, generating a spatiotemporal unified engineering knowledge graph that eliminates conflicts and carries timestamps and WBS encoded indexes.

[0007] As a preferred embodiment of the multi-party participation-based metal mine data knowledge base management method of the present invention, the step of constructing an engineering graph structure based on a standardized multi-party engineering participant dataset using a graph neural network is as follows: Extract node elements from a standardized dataset of multiple project participants according to the dimension of the project object; Define the relationships between node elements and generate edge relationships based on the engineering logic; The parameters in the multi-party engineering participant dataset are written into the attributes of node elements and edge relationships to generate a graph-structured multi-party engineering participant dataset. The node elements, edge relationships, and attributes in the graph-structured multi-party engineering participant dataset are used as input to a graph neural network to perform message passing and embedding computation, thereby generating an engineering graph structure.

[0008] As a preferred embodiment of the multi-party participation-based metal mine data knowledge base management method of the present invention, the specific steps of dividing the data into standardized engineering knowledge subgraph sets are as follows: The engineering drawing structure is divided according to the work breakdown structure code to generate engineering knowledge sub-graphs; Write the work breakdown structure code, timestamp identifier, and digital signature into all engineering knowledge subgraphs to generate a standardized set of engineering knowledge subgraphs.

[0009] As a preferred embodiment of the multi-party participation-based metal mine data knowledge base management method described in this invention, the steps include: initializing the quantum register through a standardized engineering knowledge subgraph set, obtaining the quantum fingerprint, obtaining the difference value between the quantum fingerprint of the building information model and the quantum fingerprint of the on-site construction record, and obtaining a verification safety engineering knowledge subgraph set carrying a timestamp. The specific steps are as follows: Each engineering knowledge subgraph in the standardized engineering knowledge subgraph set is encoded into a quantum state sequence, generating a quantum register loaded with the standardized engineering knowledge subgraph set; A quantum logic gate transformation is applied to the sequence of qubits in the quantum register to obtain the encoded quantum state; Quantum fingerprints are generated by performing measurements and hash operations on quantum states; The difference between the quantum fingerprint of the building information model and the quantum fingerprint of the on-site construction record is obtained. Combined with the difference threshold, a knowledge sub-graphet of verification safety engineering with timestamp is obtained.

[0010] As a preferred embodiment of the multi-party participation-based metal mine data knowledge base management method of the present invention, the specific steps of constructing a four-dimensional engineering tensor based on a timestamped verification safety engineering knowledge sub-graphite, calculating a joint conflict score, and obtaining a set of conflict coordinates with WBS encoding and coordinate index are as follows. Extract the spatial location coordinates, planned and actual project duration, cost and quality indicators, work breakdown structure code, and participant tags from each engineering knowledge subgraph in the verification safety engineering knowledge subgraph set with timestamps; Based on four dimensions—spatial location coordinates, planned and actual construction period, cost and quality indicators, and participant labels—a four-dimensional engineering tensor is generated by traversing all engineering knowledge subgraphs. Based on the four-dimensional engineering tensor, the joint conflict score is calculated one by one; Points with a joint conflict score greater than the joint threshold are marked as conflict points, and a set of conflict coordinates with WBS coding and coordinate index is generated by combining the corresponding work breakdown structure coding.

[0011] As a preferred embodiment of the multi-party participation-based metal mine data knowledge base management method described in this invention, the specific steps for calculating multi-source confidence weights and generating a fusion scheme based on a set of conflicting coordinates with WBS encoding and coordinate indexing are as follows: For conflicting coordinate points in a conflicting coordinate set with WBS encoding and coordinate index, extract data slices provided by multiple parties to generate a conflicting coordinate set with data slices from participating parties. Calculate the multi-source confidence weights based on the set of conflict coordinates of data slices with participating parties. Dynamic voting is performed based on the multi-source confidence weights to generate a fusion scheme.

[0012] As a preferred embodiment of the multi-party participation-based metal mine data knowledge base management method described in this invention, the following steps are taken: According to the fusion scheme, a multimodal spatiotemporal convolutional kernel is used to perform restricted registration and local reconstruction of the building information model and the field measured point cloud data at conflict coordinates, generating a spatiotemporal unified engineering knowledge graph that eliminates conflicts and carries timestamps and WBS encoded indexes. The specific steps are as follows: Read the work breakdown structure code, conflict coordinate index, priority of data sources to be accepted, geometric correction range, schedule correction range and cost correction range from the fusion scheme, and generate a set of correction parameters; Based on the work breakdown structure coding and conflict coordinate index of the fusion scheme, extract the slice set of building information model and on-site measured point cloud data; Based on the modified parameter set, and combined with the building information model and the on-site measured point cloud data slice set, a multimodal spatiotemporal convolution kernel parameter set is set; Restricted registration and local reconstruction are performed on the multimodal spatiotemporal convolution kernel parameter set and the modified parameter set to generate a spatiotemporal unified engineering knowledge graph that eliminates conflicts and carries timestamps and WBS encoded indexes.

[0013] As a preferred embodiment of the metal mine data knowledge base management method based on multi-party participation described in this invention, the multi-party participation includes the owner, designer, general contractor, cost estimator, construction contractor, and supervisor. The multi-party project participant dataset includes investment plans and acceptance standards, building information models and drawings, overall schedule plans, budget breakdown lists, on-site quality inspection data, and sub-item evaluation reports.

[0014] As a preferred embodiment of the metal mine data knowledge base management method based on multi-party participation described in this invention, the preprocessing includes unified coordinate system calibration, unified time reference alignment, unified encoding mapping, and integrity verification.

[0015] This invention provides a multi-party participatory metal mine data knowledge base management system, including, The data acquisition module is used to collect datasets from multiple project participants, and after preprocessing, generate standardized datasets from these participants. The subgraph construction module is used to construct an engineering graph structure based on a standardized multi-party engineering participant dataset, and divide it into a standardized set of engineering knowledge subgraphs. The fingerprint verification module is used to initialize the quantum register through a standardized engineering knowledge subgraph set, obtain the quantum fingerprint, obtain the difference value between the quantum fingerprint of the building information model and the quantum fingerprint of the on-site construction record, and obtain the verification security engineering knowledge subgraph set with timestamps. The conflict scoring module is used to construct a four-dimensional engineering tensor based on the verification security engineering knowledge subgraph with timestamps, calculate the joint conflict score, and obtain a set of conflict coordinates with WBS encoding and coordinate index. The scheme fusion module is used to calculate the multi-source confidence weights and generate a fusion scheme set based on the conflict coordinate set with WBS encoding and coordinate index. The registration and reconstruction module is used to perform restricted registration and local reconstruction of the building information model and the measured point cloud data at conflict coordinates based on the fusion scheme set and using multimodal spatiotemporal convolution kernels. This generates a spatiotemporal unified engineering knowledge graph that eliminates conflicts and carries timestamps and WBS encoded indexes.

[0016] The beneficial effects of this invention are as follows: By collecting and standardizing datasets from multiple engineering participants, constructing an engineering knowledge subgraph using a graph neural network, and then using a quantum register to generate a quantum fingerprint to verify the consistency between design data and field data, the security and reliability of the data are ensured. Furthermore, by constructing a four-dimensional engineering tensor containing spatial, period, cost, and participant dimensions for joint conflict scoring, and combining multi-source confidence weights and multimodal spatiotemporal convolution kernels to complete restricted registration and local reconstruction at conflict coordinates, automatic conflict identification and resolution are achieved. This results in improving the reliability of mining engineering data, reducing false alarms and rework rates, and enhancing the efficiency of multi-party collaborative decision-making. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a data knowledge base management method for metal mines based on multi-party participation.

[0019] Figure 2 This is a schematic diagram of a metal mine data knowledge base management system based on multi-party participation.

[0020] Figure 3 This is a flowchart of the construction of an engineering graph structure and the partitioning of engineering knowledge subgraphs based on graph neural networks.

[0021] Figure 4 A flowchart for acquiring the knowledge sub-graphet for quantum fingerprint generation and verification security engineering. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] Reference Figures 1-4 This is one embodiment of the present invention, which provides a method for managing a metal mine data knowledge base based on multi-party participation, including the following steps: S1: Collect data from multiple project participants, preprocess the data, and generate a standardized dataset of multiple project participants. S1.1: Collect datasets from multiple project participants; Furthermore, a dataset of data from multiple project participants is collected, including the investment plan and acceptance standards provided by the owner, the building information model and drawings uploaded by the designer, the overall schedule entered by the general contractor, the cost estimate breakdown list synchronized by the cost estimator, the on-site quality inspection data collected by the construction party, and the sub-item evaluation reports submitted by the supervisor. These are then compiled to form a dataset of data from multiple project participants.

[0026] S1.2: Perform unified coordinate system calibration, unified time reference alignment, unified encoding mapping, and integrity verification on the multi-party engineering participant dataset to generate a standardized multi-party engineering participant dataset; Furthermore, the datasets of multiple project participants are preprocessed, including unifying the spatial coordinates of building information model components and the location coordinates of on-site quality inspection data, unifying the time bases of investment plans, overall schedule plans, on-site quality inspection data and sub-item evaluation reports, unifying investment item codes, contract item codes, building information model component codes and work breakdown structure codes, and performing integrity verification to output a standardized dataset of multiple project participants.

[0027] S2: Based on a standardized multi-party engineering participant dataset, an engineering graph structure is constructed using a graph neural network and divided into a standardized set of engineering knowledge subgraphs; S2.1: Extract node elements from a standardized multi-party project participant dataset according to the dimension of the project object; Furthermore, node elements are extracted from a standardized multi-party project participant dataset according to the project object dimension. The project object dimension includes building information model components, investment plan items, contract items, schedule tasks, on-site quality inspection data, and sub-item evaluation reports. Each project object is mapped to an independent node element, and the corresponding spatial location, numerical parameters, and text parameters are retained for subsequent relationship construction and graph neural network processing.

[0028] S2.2: Define the relationships between node elements and generate edge relationships based on the engineering logic; Furthermore, the relationships between node elements are defined according to the engineering logic, and edge relationships are generated. The engineering logic includes spatial adjacency relationships, schedule dependency relationships, responsibility attribution relationships, and supervision and evaluation relationships. Each type of relationship is established as an edge between node elements, and the relationship type and relationship weight are recorded in the edge to generate a set of edge relationships that can represent the dependency relationships of engineering objects.

[0029] S2.3: Write the parameters from the multi-party engineering participant dataset into the attributes of node elements and edge relationships to generate a graph-structured multi-party engineering participant dataset; Furthermore, the parameters in the multi-party engineering participant dataset are written into the attributes of node elements and edge relationships. The attributes of node elements include spatial location, geometric parameters, project duration information, monetary indicators and detection results, and the attributes of edge relationships include relationship weight, temporal order and supervision results. After writing, a graph-structured multi-party engineering participant dataset is obtained.

[0030] S2.4: Take the node elements, edge relationships and attributes in the graph-structured multi-party engineering participant dataset as input to the graph neural network, perform message passing and embedding computation, and generate the engineering graph structure; Furthermore, the expression for embedded computation is: ; in, Represents a node In the The embedding vector obtained after convolutional layer calculation Represents a non-linear activation function. Indicates neighboring nodes, Represents a node The set of neighboring nodes, Represents a node The degree, Representing neighboring nodes The degree, Indicates the first The weight matrix of the layer, Representing neighboring nodes In the Layer embedding vectors, Represents a node In the Layer embedding vector; By using node elements, edge relationships, and attributes as inputs to a graph neural network, message passing and embedding computation are performed. During message passing, each node element aggregates attribute information from its neighboring nodes. During embedding computation, node attributes and edge attributes are mapped to vector representations and continuously updated, ultimately generating an engineering graph structure that can represent global features.

[0031] S2.5: Divide the engineering drawing structure according to the work breakdown structure code and generate engineering knowledge sub-graphs; Furthermore, the engineering diagram structure is divided according to the work breakdown structure code. The work breakdown structure code serves as the basis for division. The node elements and edge relationships corresponding to different codes are divided into independent engineering knowledge subgraphs. Each engineering knowledge subgraph corresponds to a sub-item engineering task or project, forming a set of engineering knowledge subgraphs.

[0032] S2.6: Write the work breakdown structure code, timestamp identifier and digital signature to all engineering knowledge subgraphs to generate a standardized set of engineering knowledge subgraphs; Furthermore, a work breakdown structure (WBS) code, a timestamp, and a digital signature are written into all engineering knowledge subgraphs. The WBS code records the subgraph's ownership, the timestamp records the generation time, and the digital signature is generated by an encryption algorithm to ensure that the data is tamper-proof. After writing is completed, a standardized set of engineering knowledge subgraphs is obtained.

[0033] S3: Initialize the quantum register through a standardized engineering knowledge subgraph set, obtain the quantum fingerprint, obtain the difference value between the quantum fingerprint of the building information model and the quantum fingerprint of the on-site construction record, and obtain the verification safety engineering knowledge subgraph set with timestamps; S3.1: Encode each engineering knowledge subgraph in the standardized engineering knowledge subgraph set into a quantum state sequence, and generate a quantum register that loads the standardized engineering knowledge subgraph set; Furthermore, the node elements, edge relationships, and attributes in the engineering knowledge subgraph are expanded in a fixed order to form a numerical vector; the numerical vector is normalized to ensure that each component is within the same numerical range and the overall scale is comparable; the normalized components are sequentially mapped as the amplitude coefficients of the qubits to generate a unique corresponding quantum state using amplitude encoding; each engineering knowledge subgraph obtains a set of qubit sequences accordingly, which are then loaded one by one into a quantum register, ultimately forming a quantum register loaded with a set of standardized engineering knowledge subgraphs.

[0034] S3.2: Apply quantum logic gate transformations to the sequence of qubits in the quantum register to obtain the encoded quantum state; Furthermore, quantum logic gate transformations are applied to the qubit sequence in the quantum register. The quantum logic gates used include commonly used logic gates such as Hadamard gates, phase gates, and controlled NOT gates, which are used to introduce superposition states, phase shifts, and controlled flips in the qubit sequence. This enables the qubit sequence to enhance its ability to distinguish engineering knowledge subgraph sets in the quantum state space. After the quantum logic gate transformation, the encoded quantum state is obtained.

[0035] S3.3: Generating quantum fingerprints by performing measurements and hash operations on quantum states; Furthermore, a measurement operation is performed on the quantum state to obtain the measurement result distribution, and the measurement result distribution is hashed using the SHA-256 hash method to output a fixed-length hash value as a quantum fingerprint. The quantum fingerprint is used to compress and characterize the main features of the standardized engineering knowledge subgraph set and has collision resistance capability, finally obtaining the quantum fingerprint corresponding to the standardized engineering knowledge subgraph set.

[0036] S3.4: Obtain the difference between the quantum fingerprint of the building information model and the quantum fingerprint of the on-site construction record, and combine it with the difference threshold to obtain a verification safety engineering knowledge sub-graphet carrying a timestamp; Furthermore, the difference between the quantum fingerprint of the building information model and the quantum fingerprint of the on-site construction record is obtained, and the difference is calculated using Euclidean distance. The difference is compared with a preset difference threshold. When the difference is less than the difference threshold, the subgraph hash value and version timestamp of the standardized engineering knowledge subgraph set are written into the verification chain and marked as trustworthy. When the difference is greater than or equal to the difference threshold, the result is marked as a pending state and written into the anomaly record, finally obtaining a verification security engineering knowledge subgraph set with a timestamp.

[0037] It should be noted that the difference threshold setting needs to be defined in conjunction with the matching accuracy requirements of the Building Information Model (BIM) quantum fingerprint and the on-site construction record quantum fingerprint. The difference threshold is essentially a critical value used to determine whether the two types of data are consistent, and its magnitude directly affects the credibility of verifying the safety engineering knowledge sub-graphet. When the difference value is less than the difference threshold, it indicates that the difference between the BIM quantum fingerprint and the on-site construction record quantum fingerprint is within an acceptable range, and they can be considered consistent. When the difference value is greater than or equal to the difference threshold, it indicates that there is a significant deviation between the two, and they need to be marked as triggering a pending state. The difference threshold is usually determined empirically based on the error distribution in historical project data, referring to the average deviation level of Euclidean distance in engineering scenarios, and appropriately introducing a safety factor to avoid misjudging acceptable small deviations as anomalies. This difference threshold setting ensures the reliability of verification while avoiding oversensitivity that leads to frequent triggering of abnormal records.

[0038] S4: Based on the verification security engineering knowledge subgraph with timestamps, construct a four-dimensional engineering tensor, calculate the joint conflict score, and obtain a set of conflict coordinates with WBS encoding and coordinate index. S4.1: Extract the spatial location coordinates, planned and actual project duration, cost and quality indicators, work breakdown structure code, and participant tags from the verification safety engineering knowledge subgraph set with timestamps for each engineering knowledge subgraph; Furthermore, key elements are extracted from each engineering knowledge subgraph in the verification safety engineering knowledge subgraph set with timestamps. The extracted content includes the spatial coordinates of building information model components or construction points, the time interval between the planned and actual construction periods, the numerical parameters of cost and quality indicators, the work breakdown structure code, and the corresponding participant tags. After extraction, engineering knowledge subgraph data entries with multi-dimensional attributes are formed.

[0039] S4.2: Generate a four-dimensional engineering tensor by traversing all engineering knowledge subgraphs according to four dimensions: spatial location coordinates, planned and actual construction period, cost and quality indicators, and participant labels; Furthermore, based on four dimensions—spatial location coordinates, planned and actual construction period, cost and quality indicators, and participant labels—all engineering knowledge subgraphs in the verification safety engineering knowledge subgraph set with timestamps are traversed. The extracted multidimensional attributes are arranged and combined sequentially according to the four dimensions to form a four-dimensional array structure. In the calculation, the four-dimensional array structure serves as a four-dimensional engineering tensor, used to uniformly represent the spatial, temporal, economic, and quality characteristics of engineering objects.

[0040] S4.3: Calculate the joint conflict score one by one based on the four-dimensional engineering tensor; Furthermore, based on the constructed four-dimensional engineering tensor, the joint conflict score is calculated one by one. The calculation method of the joint conflict score can adopt the method of superimposing multi-dimensional indicators, and weighted calculation of geometric conflicts existing in spatial location coordinates, time deviation between planned and actual construction period, difference values ​​of cost and quality indicators, and data divergence corresponding to the labels of participating parties, to obtain the conflict score of each tensor unit, and finally form a joint conflict score result set covering the entire four-dimensional engineering tensor.

[0041] The joint conflict score is calculated using the following expression: ; in, For coding in the work breakdown structure Spatial coordinate index The joint conflict score below, For coding in the work breakdown structure Spatial coordinate index Geometric conflict scoring below For coding in the work breakdown structure Spatial coordinate index The following is the schedule deviation score. For coding in the work breakdown structure Spatial coordinate index Cost deviation score below For coding in the work breakdown structure Spatial coordinate index The quality deviation score below, For coding in the work breakdown structure Spatial coordinate index The following is a score of disagreement among the participants. , These represent the weighting coefficients for geometric conflict scores, schedule deviation scores, cost deviation scores, quality deviation scores, and participant disagreement scores, respectively. It should be noted that in the calculation of the joint conflict score, the geometric conflict score is based on the geometric parameters of the building information model components and the on-site point cloud data. The geometric conflict quantity is calculated through minimum clearance or intersection detection, and then normalized to obtain the score. The schedule deviation score is based on the difference between the planned schedule and the actual schedule, and normalized to obtain the score after standardization. The cost deviation score is based on the non-negative part of the difference between the contract price or estimate and the actual cost, and normalized to obtain the score after standardization. The quality deviation score is based on the difference between the quality inspection results and the acceptance standards, and normalized to obtain the score after standardization within the allowable tolerance range. The participant disagreement score is based on the dispersion of the data submitted by multiple parties, and is calculated by weighted variance or normalized entropy.

[0042] S4.4: Mark the points with joint conflict scores greater than the joint threshold as conflict points, and combine them with the corresponding work breakdown structure code to generate a set of conflict coordinates with WBS code and coordinate index; Furthermore, units in the four-dimensional engineering tensor with joint conflict scores greater than the joint threshold are marked as conflict points, and the spatial coordinates corresponding to the conflict points are bound to the work breakdown structure code to generate a set of conflict coordinates with WBS code and coordinate index that can characterize the conflict location and the sub-project affiliation. The set of conflict coordinates is used as the input for subsequent fusion scheme calculation.

[0043] It should be noted that, based on the statistical analysis of historical projects, geometric conflict scores, schedule deviation scores, cost deviation scores, quality deviation scores, and participant disagreement scores that occurred in previous projects were sampled and statistically analyzed. The distribution range of these scores under normal circumstances was calculated, and the upper quantile of the statistical distribution was selected as the joint threshold.

[0044] S5: Based on the set of conflicting coordinates with WBS encoding and coordinate index, calculate the multi-source confidence weights and generate a fusion scheme; S5.1: For conflicting coordinate points in a conflicting coordinate set with WBS encoding and coordinate index, extract data slices provided by multiple parties and generate a conflicting coordinate set with data slices from participating parties. Furthermore, for the conflict coordinate points in the conflict coordinate set with WBS coding and coordinate index, data slices provided by the multiple engineering participants involved in the conflict are extracted one by one. The data slices include information such as investment plan items, building information model component parameters, schedule data, budget list items, quality inspection records and sub-item evaluation conclusions, and are matched according to the conflict coordinate points to finally generate a conflict coordinate set with data slices of the participants.

[0045] S5.2: Calculate the multi-source confidence weights based on the set of conflict coordinates of data slices with participating parties; Furthermore, for each participant's data slice, the update records at the conflict coordinate points are statistically analyzed, and the time difference with the current time is calculated. The time difference is then processed using min-max normalization to obtain a timeliness score. The completeness of the data slice's fields is checked, and the completeness score is obtained based on the proportion of missing fields using min-max normalization. Based on the review results of historical projects, the participant's matching ratio on the same type of task is statistically analyzed, and the historical accuracy score is obtained using min-max normalization. The number of modifications or modification frequency of the statistical data slice during the submission process is statistically analyzed, and the change characteristic score is obtained using monotonically decreasing min-max normalization. The timeliness score, completeness score, historical accuracy score, and change characteristic score are linearly weighted according to a weighted ratio to obtain the participant's confidence weight at the conflict coordinate point. After completing the calculations for all participants, the results are summarized to form a multi-source confidence weight set.

[0046] S5.3: Dynamically vote based on the multi-source confidence weights and generate a fusion scheme; Furthermore, based on the multi-source confidence weight set, weighted voting is performed on the conflict coordinate set of data slices involving participating parties. During the voting process, a voting weight proportional to the confidence weight is assigned to each participating party's data slice. All data slices corresponding to the voting weights are collected and the weighted synthesis result is calculated to determine the data source to be accepted at the conflict coordinate points, as well as the geometric correction range, schedule correction range, and cost correction range. Based on this, a fusion scheme is generated and used as the input basis for subsequent restricted registration and local reconstruction.

[0047] S6: According to the fusion scheme, multimodal spatiotemporal convolution kernels are used to perform restricted registration and local reconstruction of building information model and on-site measured point cloud data at conflict coordinates, generating a spatiotemporal unified engineering knowledge graph that eliminates conflicts and carries timestamps and WBS encoded indexes; S6.1: Read the work breakdown structure code, conflict coordinate index, priority of data sources to be accepted, geometric correction range, schedule correction range and cost correction range from the fusion scheme, and generate a set of correction parameters; Furthermore, the work breakdown structure code, conflict coordinate index, priority of data sources to be accepted, geometric correction range, schedule correction range, and cost correction range in the fusion scheme are read and organized according to the correspondence between engineering tasks and spatial coordinates to form a correction parameter set. The correction parameter set is used as the basis for adjustment in the subsequent registration and local reconstruction process.

[0048] S6.2: Extract the building information model and field measured point cloud data slice set based on the work breakdown structure coding and conflict coordinate index of the fusion scheme; Furthermore, based on the work breakdown structure coding and conflict coordinate index of the fusion scheme, data slices corresponding to the conflict locations are extracted from the building information model and the field measured point cloud data. The building information model data slices include the component geometric parameters and spatial location, while the field measured point cloud data slices include the coordinates of the measurement points and the point cloud density information. After extraction, a set of building information model and field measured point cloud data slices is formed.

[0049] S6.3: Based on the modified parameter set, and combined with the building information model and the actual measured point cloud data slice set, set a multimodal spatiotemporal convolution kernel parameter set; Furthermore, based on the modified parameter set, and combined with the building information model and the field measured point cloud data slice set, the parameters of the multimodal spatiotemporal convolution kernel are set. The set parameters include the spatial dimension convolution kernel size, the temporal dimension convolution stride, and the weight ratio of different modal feature fusion. After the multimodal spatiotemporal convolution kernel parameter set is set, it can support the joint analysis of building information model and field measured point cloud data in the spatial and temporal dimensions.

[0050] S6.4: Perform restricted registration and local reconstruction on the multimodal spatiotemporal convolution kernel parameter set and the modified parameter set to generate a spatiotemporal unified engineering knowledge graph that eliminates conflicts and carries timestamps and WBS encoded indexes; Furthermore, restricted registration and local reconstruction are performed on the multimodal spatiotemporal convolution kernel parameter set and the correction parameter set. During the restricted registration process, the spatial position of the building information model components is adjusted by using the geometric correction magnitude in the correction parameter set. At the same time, the local contours of the components are aligned by combining point cloud data slicing. During the local reconstruction process, the schedule correction magnitude and cost correction magnitude in the correction parameter set are used to correct the schedule and cost attributes. Finally, a spatiotemporal unified engineering knowledge graph with conflict elimination and carrying timestamps and work breakdown structure coding indexes is generated.

[0051] This embodiment also provides a metal mine data knowledge base management system based on multi-party participation, including: The data acquisition module is used to collect datasets from multiple project participants, and after preprocessing, generate standardized datasets from these participants. The subgraph construction module is used to construct an engineering graph structure based on a standardized multi-party engineering participant dataset, and divide it into a standardized set of engineering knowledge subgraphs. The fingerprint verification module is used to initialize the quantum register through a standardized engineering knowledge subgraph set, obtain the quantum fingerprint, obtain the difference value between the quantum fingerprint of the building information model and the quantum fingerprint of the on-site construction record, and obtain the verification security engineering knowledge subgraph set with timestamps. The conflict scoring module is used to construct a four-dimensional engineering tensor based on the verification security engineering knowledge subgraph with timestamps, calculate the joint conflict score, and obtain a set of conflict coordinates with WBS encoding and coordinate index. The scheme fusion module is used to calculate the multi-source confidence weights and generate a fusion scheme set based on the conflict coordinate set with WBS encoding and coordinate index. The registration and reconstruction module is used to perform restricted registration and local reconstruction of the building information model and the measured point cloud data at conflict coordinates based on the fusion scheme set and using multimodal spatiotemporal convolution kernels. This generates a spatiotemporal unified engineering knowledge graph that eliminates conflicts and carries timestamps and WBS encoded indexes.

[0052] This embodiment also provides a computer device applicable to a multi-party participation-based metal mine data knowledge base management method, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the multi-party participation-based metal mine data knowledge base management method proposed in the above embodiment.

[0053] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0054] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the multi-party participation-based metal mine data knowledge base management method proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0055] In summary, this invention achieves data security and reliability by: collecting and standardizing datasets from multiple engineering participants; constructing an engineering knowledge subgraph using a graph neural network; and generating quantum fingerprints using quantum registers to verify the consistency between design data and field data. Furthermore, it constructs a four-dimensional engineering tensor encompassing spatial, duration, cost, and participant dimensions for joint conflict scoring. By combining multi-source confidence weights and multimodal spatiotemporal convolution kernels to perform constrained registration and local reconstruction at conflict coordinates, it realizes automatic conflict identification and resolution. This results in improved reliability of mining engineering data, reduced false alarms and rework rates, and enhanced efficiency of multi-party collaborative decision-making.

[0056] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for managing a metal mine data knowledge base based on multi-party participation, characterized by: include, Collect datasets from multiple project participants, preprocess them, and generate standardized datasets from these participants. Based on a standardized dataset of multiple engineering participants, an engineering graph structure is constructed using a graph neural network and divided into a set of standardized engineering knowledge subgraphs. The quantum register is initialized by a standardized engineering knowledge subgraph set, the quantum fingerprint is obtained, the difference between the quantum fingerprint of the building information model and the quantum fingerprint of the on-site construction record is obtained, and the verification safety engineering knowledge subgraph set with timestamp is obtained. Based on the verification security engineering knowledge subgraph with timestamps, a four-dimensional engineering tensor is constructed, a joint conflict score is calculated, and a set of conflict coordinates with WBS encoding and coordinate index is obtained. Based on the set of conflicting coordinates with WBS encoding and coordinate index, multi-source confidence weights are calculated to generate a set of fusion schemes; Based on the fusion scheme set, multimodal spatiotemporal convolution kernels are used to perform restricted registration and local reconstruction of building information models and field measured point cloud data at conflict coordinates, generating a spatiotemporal unified engineering knowledge graph that eliminates conflicts and carries timestamps and WBS encoded indexes.

2. The method for managing a metal mine data knowledge base based on multi-party participation as described in claim 1, characterized in that: The standardized multi-party engineering participant dataset is used to construct an engineering graph structure through a graph neural network. The specific steps are as follows. Extract node elements from a standardized dataset of multiple project participants according to the dimension of the project object; Define the relationships between node elements and generate edge relationships based on the engineering logic; The parameters in the multi-party engineering participant dataset are written into the attributes of node elements and edge relationships to generate a graph-structured multi-party engineering participant dataset. The node elements, edge relationships, and attributes in the graph-structured multi-party engineering participant dataset are used as input to a graph neural network to perform message passing and embedding computation, thereby generating an engineering graph structure.

3. The method for managing a metal mine data knowledge base based on multi-party participation as described in claim 2, characterized in that: The specific steps for dividing the knowledge into standardized engineering knowledge subgraph sets are as follows. The engineering drawing structure is divided according to the work breakdown structure code to generate engineering knowledge sub-graphs; All engineering knowledge subgraphs are written into the work breakdown structure code, timestamp identifier, and digital signature to generate a standardized set of engineering knowledge subgraphs.

4. The method for managing a metal mine data knowledge base based on multi-party participation as described in claim 3, characterized in that: The process involves initializing the quantum register using a standardized engineering knowledge subgraph set, obtaining the quantum fingerprint, acquiring the difference between the building information model's quantum fingerprint and the on-site construction record's quantum fingerprint, and obtaining a verification safety engineering knowledge subgraph set with timestamps. The specific steps are as follows: Encode each engineering knowledge subgraph in the standardized engineering knowledge subgraph set into a quantum state sequence to generate a quantum register that loads the standardized engineering knowledge subgraph set; A quantum logic gate transformation is applied to the sequence of qubits in the quantum register to obtain the encoded quantum state; Quantum fingerprints are generated by performing measurements and hash operations on quantum states; The difference between the quantum fingerprint of the building information model and the quantum fingerprint of the on-site construction record is obtained. Combined with the difference threshold, a knowledge sub-graphet of verification safety engineering with timestamp is obtained.

5. The method for managing a metal mine data knowledge base based on multi-party participation as described in claim 4, characterized in that: The steps for constructing a four-dimensional engineering tensor based on a timestamped verification security engineering knowledge sub-graph, calculating a joint conflict score, and obtaining a set of conflict coordinates with WBS encoding and coordinate indices are as follows. Extract the spatial location coordinates, planned and actual project duration, cost and quality indicators, work breakdown structure code, and participant tags from each engineering knowledge subgraph in the verification safety engineering knowledge subgraph set with timestamps; Based on four dimensions—spatial location coordinates, planned and actual construction period, cost and quality indicators, and participant labels—a four-dimensional engineering tensor is generated by traversing all engineering knowledge subgraphs. Based on the four-dimensional engineering tensor, the joint conflict score is calculated one by one; Points with a joint conflict score greater than the joint threshold are marked as conflict points, and a set of conflict coordinates with WBS coding and coordinate index is generated by combining the corresponding work breakdown structure coding.

6. The method for managing a metal mine data knowledge base based on multi-party participation as described in claim 5, characterized in that: The process of calculating multi-source confidence weights and generating a fusion scheme based on a set of conflicting coordinates with WBS encoding and coordinate indexing is as follows: For conflicting coordinate points in a conflicting coordinate set with WBS encoding and coordinate index, extract data slices provided by multiple parties to generate a conflicting coordinate set with data slices from participating parties. Calculate the multi-source confidence weights based on the set of conflict coordinates of data slices with participating parties. Dynamic voting is performed based on the multi-source confidence weights to generate a fusion scheme.

7. The method for managing a metal mine data knowledge base based on multi-party participation as described in claim 6, characterized in that: According to the fusion scheme, multimodal spatiotemporal convolutional kernels are used to perform restricted registration and local reconstruction of the building information model and the measured point cloud data at conflict coordinates, generating a spatiotemporal unified engineering knowledge graph that eliminates conflicts and carries timestamps and WBS encoded indexes. The specific steps are as follows. Read the work breakdown structure code, conflict coordinate index, priority of data sources to be accepted, geometric correction range, schedule correction range and cost correction range from the fusion scheme, and generate a set of correction parameters; Based on the work breakdown structure coding and conflict coordinate index of the fusion scheme, extract the slice set of building information model and on-site measured point cloud data; Based on the modified parameter set, and combined with the building information model and the on-site measured point cloud data slice set, a multimodal spatiotemporal convolution kernel parameter set is set; Restricted registration and local reconstruction are performed on the multimodal spatiotemporal convolution kernel parameter set and the modified parameter set to generate a spatiotemporal unified engineering knowledge graph that eliminates conflicts and carries timestamps and WBS encoded indexes.

8. The method for managing a metal mine data knowledge base based on multi-party participation as described in claim 1, characterized in that: The various parties mentioned include the owner, the designer, the general contractor, the cost estimator, the construction contractor, and the supervisor. The multi-party project participant dataset includes investment plans and acceptance standards, building information models and drawings, overall schedule plans, budget breakdown lists, on-site quality inspection data, and sub-item evaluation reports.

9. The method for managing a metal mine data knowledge base based on multi-party participation as described in claim 1, characterized in that: The preprocessing includes unified coordinate system calibration, unified time reference alignment, unified encoding mapping, and integrity verification.

10. A metal mine data knowledge base management system based on multi-party participation, based on the metal mine data knowledge base management method based on multi-party participation as described in any one of claims 1 to 9, characterized in that: include, The data acquisition module is used to collect datasets from multiple project participants, and after preprocessing, generate standardized datasets from these participants. The subgraph construction module is used to construct an engineering graph structure based on a standardized multi-party engineering participant dataset, and divide it into a standardized set of engineering knowledge subgraphs. The fingerprint verification module is used to initialize the quantum register through a standardized engineering knowledge subgraph set, obtain the quantum fingerprint, obtain the difference value between the quantum fingerprint of the building information model and the quantum fingerprint of the on-site construction record, and obtain the verification security engineering knowledge subgraph set with timestamps. The conflict scoring module is used to construct a four-dimensional engineering tensor based on the verification security engineering knowledge subgraph with timestamps, calculate the joint conflict score, and obtain a set of conflict coordinates with WBS encoding and coordinate index. The scheme fusion module is used to calculate the multi-source confidence weights and generate a fusion scheme set based on the conflict coordinate set with WBS encoding and coordinate index. The registration and reconstruction module is used to perform restricted registration and local reconstruction of the building information model and the measured point cloud data at conflict coordinates based on the fusion scheme set and using multimodal spatiotemporal convolution kernels. This generates a spatiotemporal unified engineering knowledge graph that eliminates conflicts and carries timestamps and WBS encoded indexes.

Images