A construction project whole-process management system and method based on BIM3D-6D technology

Abstract

The application discloses a kind of BIM3D-6D technology as pedestal's construction project whole process management system and method, it is related to construction project management technical field, including, through BIM lightweight computing server loads 3D building information model, extracts the space topological relation of geometric component and IFC semantic attribute, and the space topological relation of geometric component and IFC semantic attribute is carried out space topological analysis and semantic attribute mapping, generates three-dimensional data topological network package based on graph structure;Multi-dimensional deduction operation is carried out to three-dimensional data topological network package based on graph structure, obtains the deviation compensation strategy containing optimal cost control path;3D building information model is expanded through dimension and data fusion, generates 5D cost dimension data set.Mathematical transformation is used to convert discrete decision into continuous field model, so that resource flow path has spatial correlation and time continuity, and the beneficial effect of dynamic optimization construction process is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of construction project management, and in particular to a construction project whole-process management system and method based on BIM 3D-6D technology. BACKGROUND

[0002] In the construction project management driven by building information modeling, the conventional method mainly relies on the static association of 3D geometric model and progress, cost data to realize whole-process management and control. The typical process includes: integrating BIM model, construction plan and bill of quantities through BIM platform to form a visual management and control interface; using rule engine to analyze model component attributes to generate resource allocation scheme; combining with Internet of Things device feedback of site state to realize dynamic monitoring of construction progress and cost. Such method significantly improves the construction collaboration efficiency through structured data mapping and visual interaction, and provides a digital foundation for quality and safety management, which has become the core technical support of modern intelligent construction.

[0003] However, the existing method still has two limitations:

[0004] Insufficient multi-dimensional data coupling: the association of cost, progress and spatial topology depends on manual rule configuration, which is difficult to dynamically capture the spatio-temporal constraint relationship of process logic chain, resulting in decoupling risk between resource scheduling strategy and actual site working condition; dynamic response lag: the decision deduction based on static model lacks real-time disturbance correction mechanism, when the site resource consumption deviates from the plan, manual parameter adjustment is required, which is difficult to guarantee the global optimality of the cost control path. SUMMARY

[0005] In view of the above existing problems, the present application is proposed.

[0006] Therefore, the present application provides a construction project whole-process management system and method based on BIM 3D-6D technology, which solves the problems of insufficient dynamic coupling of multi-dimensional data and lag of global optimization of cost control path.

[0007] To solve the above technical problems, the present application provides the following technical solutions:

[0008] In a first aspect, the present application provides a construction project whole-process management method based on BIM 3D-6D technology, which includes:

[0009] loading 3D building information model through BIM lightweight computing server, extracting spatial topology relationship and IFC semantic attributes of geometric components, and performing spatial topology analysis and semantic attribute mapping on the spatial topology relationship and IFC semantic attributes of geometric components to generate a three-dimensional data topology network package based on graph structure;

[0010] The multi-dimensional deduction operation is carried out on the three-dimensional data topology network package based on a graph structure, and a deviation compensation strategy containing an optimal cost control path is obtained.

[0011] The 3D building information model is expanded and fused by dimensions to generate a 5D cost dimension data set.

[0012] The resource adjustment rules in the deviation compensation strategy containing the optimal cost control path are analyzed, a resource allocation optimization model is constructed, and a gradient descent algorithm is used to optimize the resource allocation optimization model, and a construction scheduling instruction set with a minimum cost disturbance factor is generated based on the 5D cost dimension data set.

[0013] The construction scheduling instruction set with the minimum cost disturbance factor is subjected to tensor convolution operation in a cognitive mirror image synchronization mode to generate a reconstructed 4D space-time progress topology structure.

[0014] Based on the reconstructed 4D space-time progress topology structure, the facility state record is updated, and combined with the readings of the Internet of Things sensing device, a 6D full life cycle data record table is generated.

[0015] As a preferred scheme of the construction project whole process management method based on the BIM 3D-6D technology, wherein: the 3D building information model is loaded through the BIM lightweight computing server, the spatial topological relationship and IFC semantic attribute of the geometric component are extracted, and the spatial topological relationship and IFC semantic attribute of the geometric component are subjected to spatial topological analysis and semantic attribute mapping to generate a three-dimensional data topology network package based on a graph structure, and the steps are as follows:

[0016] The 3D building information model is loaded through the BIM lightweight computing server, the geometric deformation degree and the topological relationship matrix of the non-key structure grid are simplified, and the geometric element set is obtained by scanning the 3D building information model based on the geometric deformation degree.

[0017] The IFC semantic attribute in the geometric element set is analyzed, and the topological relationship matrix is converted into a weighted attribute topological graph.

[0018] Based on the spatial coincidence threshold, the adjacent three-dimensional nodes of the weighted attribute topological graph are merged, an R-tree three-dimensional space index is constructed, and the R-tree three-dimensional space index is converted into a three-dimensional topological data network.

[0019] The three-dimensional data topology network package is checked, an architectural structure component connectivity graph and an attribute distribution report are generated, and are packaged to obtain a three-dimensional data topology network package based on a graph structure.

[0020] As a preferred scheme of the construction project whole process management method based on BIM3D-6D technology, wherein: the multi-dimensional deduction operation on the three-dimensional data topology network package based on the graph structure is carried out, and the deviation compensation strategy containing the optimal cost control path is obtained, and the steps are as follows:

[0021] The topology node and topology edge data are extracted from the three-dimensional data topology network package of the graph structure, the spatial coordinates of the building structure component are expanded into 4D space-time parameters based on the time window constraint, the preset resource cost parameter table is superimposed as a dynamic correction factor to generate an enhanced topology network with a disturbance factor;

[0022] Based on the enhanced topology network with the disturbance factor, the disturbance influence is propagated along the topology edge, and the stability disturbance guarantee value is obtained by iteration, and then the multi-dimensional scenario simulation is performed on the stability disturbance guarantee value to generate a three-dimensional diagnostic report;

[0023] Based on the three-dimensional diagnostic report, a multi-objective optimization function is constructed, and then the optimal non-dominated solution set is screened from the multi-objective optimization function to generate a deviation compensation strategy containing an optimal cost control path.

[0024] As a preferred scheme of the construction project whole process management method based on BIM3D-6D technology, wherein: the preset resource cost parameter table contains a component type matching field, a cost parameter value, and structured data related to parameter constraint conditions.

[0025] As a preferred scheme of the construction project whole process management method based on BIM3D-6D technology, wherein: the 3D building information model is expanded in dimension and fused with data to generate a 5D cost dimension data set, and the steps are as follows:

[0026] The geometric element set of the 3D building information model is processed by replacing and simplifying method, and the lightweight geometric data set is generated according to the geometric entity and attribute entity of the 3D building information model;

[0027] The installation time node is bound to the building structure component in the lightweight geometric data set by associating the construction progress plan with the building structure component ID, and the external bill of quantities is parsed to match the building structure component geometry in the lightweight geometric data set to obtain a cost benchmark matrix;

[0028] The 4D space-time data set and the cost benchmark matrix are dynamically cost corrected to generate a 5D cost dimension data set.

[0029] As a preferred scheme of the construction project whole process management method based on BIM3D-6D technology, the external bill of quantities is a structured resource quantization table based on cost specification, and the geometric quantity, resource consumption rule and cost parameter of the building structure component are accurately associated.

[0030] As a preferred scheme of the construction project whole process management method based on BIM3D-6D technology, the resource adjustment rule in the deviation compensation strategy containing the optimal cost control path is analyzed, a resource allocation optimization model is constructed, a gradient descent algorithm is used to optimize the resource allocation optimization model, a construction scheduling instruction set with a minimum cost disturbance factor is generated based on the 5D cost dimension data set, and the steps are as follows:

[0031] The resource type, adjustment direction and adjustment amount in the deviation compensation strategy containing the optimal cost control path are analyzed, the resource type, adjustment direction and adjustment amount are converted into mathematical constraint conditions, and a structured resource adjustment rule table is obtained.

[0032] The resource adjustment rule table is used as a constraint condition, the dynamic weight parameter in the 5D cost dimension data set is used as a decision variable, a resource allocation optimization model with a minimum cost disturbance factor as an objective function is constructed, a gradient descent algorithm is used to solve the objective function and update the dynamic weight parameter, an optimal dynamic weight parameter combination is obtained, and an optimized resource allocation optimization model is generated.

[0033] The optimal dynamic weight parameter combination is matched with the space-time coordinates in the 5D cost dimension data set through the building structure component ID, the optimal dynamic weight parameter combination is injected into the space-time node, then the structured coding task instruction is added with a cost disturbance warning instruction, and the construction scheduling instruction set with a minimum cost disturbance factor is obtained after encapsulation.

[0034] As a preferred scheme of the construction project whole process management method based on BIM3D-6D technology, the construction scheduling instruction set with a minimum cost disturbance factor is subjected to tensor convolution operation in a cognitive mirror synchronization mode, and a reconstructed 4D space-time progress topology is generated, and the steps are as follows:

[0035] The construction scheduling instruction set with a minimum cost disturbance factor is analyzed, the spatial position, time node and optimal dynamic weight parameter combination are extracted, and then the spatial position, time node and optimal dynamic weight parameter combination are mapped into a space-time resource matrix;

[0036] Based on the space-time resource matrix, a three-dimensional convolution kernel is constructed, and then the three-dimensional convolution kernel is subjected to tensor convolution operation, and a convolution-optimized space-time resource matrix is generated.

[0037] Each coordinate node in the spatio-temporal resource matrix after convolution optimization is converted into a topological vertex, the topological vertices are connected based on resource flow paths to form topological edges, and the topological edges are assigned weight values according to resource consumption intensity, and a reconstructed 4D spatio-temporal progress topology structure is generated.

[0038] As a preferred scheme of the construction project whole-process management method based on the BIM 3D-6D technology, the facility state record is updated based on the reconstructed 4D spatio-temporal progress topology structure, and the 6D whole-life cycle data record table is generated in combination with the readings of the Internet of Things sensing device, and the steps are as follows:

[0039] Based on the association of the topological vertices of the 4D spatio-temporal progress topology structure with the field entity facilities based on the building structure component ID, the resource flow paths of the topological vertices are extracted, and then the enhanced topology structure with real-time state is generated by fusing the checked Internet of Things device data to the corresponding topological vertices;

[0040] The construction period data is integrated, the topological vertices of the enhanced topology structure with real-time state are associated with the construction log and the quality detection report, and the 6D whole-life cycle data record table is constructed with the building structure component ID as the core node.

[0041] In the second aspect, the application provides a construction project whole-process management system based on the BIM 3D-6D technology, comprising:

[0042] The topology modeling module loads the 3D building information model through the BIM lightweight computing server, extracts the spatial topological relationship and IFC semantic attribute of the geometric component, and performs spatial topological analysis and semantic attribute mapping on the spatial topological relationship and IFC semantic attribute of the geometric component to generate a three-dimensional data topology network package based on a graph structure;

[0043] The path deduction module performs multi-dimensional deduction operation on the three-dimensional data topology network package based on a graph structure to obtain a deviation compensation strategy containing an optimal cost control path;

[0044] The dimension fusion module generates a 5D cost dimension data set by dimension expansion and data fusion of the 3D building information model;

[0045] The resource optimization module analyzes the resource adjustment rules in the deviation compensation strategy containing the optimal cost control path, constructs a resource allocation optimization model, and optimizes the resource allocation optimization model by using a gradient descent algorithm to generate a construction scheduling instruction set with a minimum cost disturbance factor based on the 5D cost dimension data set;

[0046] The structure reconstruction module performs tensor convolution operation on the construction scheduling instruction set with the minimum cost disturbance factor by the cognitive mirror synchronization mode to generate a reconstructed 4D spatio-temporal progress topology structure;

[0047] The closed-loop feedback module updates the facility state record based on the reconstructed 4D space-time progress topology structure, and generates a 6D full life cycle data record table in combination with the readings of the Internet of Things sensing device.

[0048] The application has the beneficial effects that: the construction scheduling instruction set is subjected to tensor convolution operation in a cognitive mirror image synchronization mode, discrete scheduling instructions are mapped into continuous space-time topology structure, the weight attenuation characteristics of the three-dimensional convolution kernel are utilized to realize space-time smooth diffusion of resource paths, the reconstructed 4D space-time progress topology structure is generated, discrete decisions are converted into continuous field models through mathematical transformation, the resource flow path has spatial correlation and time continuity, and the beneficial effect of dynamically optimizing the construction process is achieved. BRIEF DESCRIPTION OF DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0050] Figure 1 The flow chart of the whole process management method of the construction project based on BIM3D-6D technology.

[0051] Figure 2 The flow chart of the three-dimensional data topology network packet generation.

[0052] Figure 3 The flow chart of the deviation compensation strategy generation.

[0053] Figure 4 The flow chart of the 4D space-time progress topology structure reconstruction. DETAILED DESCRIPTION

[0054] In order to make the above-mentioned purposes, features and advantages of the application more obvious and easy to understand, the specific embodiments of the application will be described in detail below with reference to the drawings of the specification.

[0055] In the following description, many specific details are set forth in order to provide a thorough understanding of the application, but the application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the connotation of the application, therefore the application is not limited to the specific embodiments disclosed below.

[0056] Second, the "one embodiment" or "an embodiment" referred to herein means a specific feature, structure, characteristic, or combination of features and / or characteristics described herein that can be included in at least one implementation of the present application. The various appearances of "in one embodiment" or "an embodiment" in the specification are not necessarily all referring to the same embodiment.

[0057] Referring to Figure 1 For one embodiment of the present application, the embodiment provides a construction project whole-process management method based on BIM 3D-6D technology, comprising the following steps:

[0058] S1: loading a 3D building information model through a BIM lightweight computing server, extracting spatial topological relations and IFC semantic attributes of geometric components, and performing spatial topological analysis and semantic attribute mapping on the spatial topological relations and IFC semantic attributes of the geometric components to generate a three-dimensional data topological network package based on a graph structure.

[0059] Specifically, please refer to Figure 2 , the steps are as follows:

[0060] S1.1: loading a 3D building information model through a BIM lightweight computing server, applying a dynamic detail level selection algorithm to simplify non-critical parts in the structure grid, generating a geometric deformation degree and a topological relation matrix, and using Octree spatial indexing to scan the 3D building information model based on the geometric deformation degree to obtain a geometric element set.

[0061] Specifically, the 3D building information model is loaded through the BIM lightweight computing server, the dynamic detail level selection algorithm is executed to simplify the non-critical structure grid in the 3D building information model, the complete geometric accuracy of the critical structure grid is preserved, the geometric deformation degree of the 3D building information model is obtained, and the topological relation matrix is generated by iterative simplification. Then, using the geometric deformation degree of the 3D building information model as the weight, the 3D building information model is scanned in three-dimensional space using Octree spatial indexing to obtain the contact area and containment level between geometric elements, and the geometric element set is integrated and generated.

[0062] The dynamic detail level selection algorithm formula is as follows:

[0063] ;

[0064] Wherein, represents a newly generated vertex in the 3D building information model, represents an original grid vertex in the 3D building information model, represents the transpose coordinates of the newly generated vertex in the 3D building information model, represents the original grid vertex in the 3D building information model vertex error matrix of the top vertex of the original mesh in the 3D building information model, representing the original mesh top vertex in the 3D building information model vertex error matrix of the top vertex of the original mesh in the 3D building information model, representing the original mesh top vertex in the 3D building information model and merge into a newly generated vertex the degree of geometric deformation introduced when.

[0065] It needs to be explained that the 3D building information model is based on the building industry standard IFC format, using BIM construction software (such as Revit, Archicad), to structure the geometric data (such as the size and position of beams and columns) and semantic attributes (such as material specifications and construction stages) of building structural components, and the core of the 3D building information model is to realize the automatic generation and logical association of building elements through parameterized modeling rules and engineering knowledge base, and finally obtain a digital building twin containing geometric topology and engineering mathematics.

[0066] Among them, the parameterized modeling rule is based on building structure specifications, component size standards and construction process requirements, through analyzing the geometric characteristics and correlation of different types of building structural components, forming a programmable parameter template; the engineering knowledge base is based on historical engineering data, construction experience and material performance information, and the component type, construction stage, resource consumption rule and cost parameter are structured and sorted, and the logical constraints and correlation between components are established.

[0067] It needs to be explained that the determination of the key part and the non-key part of the structure grid is based on the structural importance and functional attributes of the building structural components, specifically, the key structure grid refers to the geometric data that directly affects the load-bearing performance of the building and the building structural components, and the determination basis includes the structural type mark in the IFC semantic attribute, the mechanical analysis parameter and the mandatory requirement in the structure specification; the non-key structure grid refers to the decorative, auxiliary or replaceable building structural components, which are usually automatically identified by building structural component classification code, non-load-bearing attribute mark and LOD (detail level) parameter.

[0068] Among them, the IFC semantic attribute is the digital attribute system defined in the IFC standard, which is used to accurately describe the engineering characteristics, functional classification and physical behavior of building structural components; the IFC semantic attribute is realized by structured data to realize the standardized storage and exchange of component information, the specific content is divided into two categories: IFC entity type and IFC attribute set.

[0069] S1.2: Analyze the IFC semantic attributes in the geometric element set, convert the topological relationship matrix into a weighted attribute topological graph, and accurately bind the IFC semantic attributes in the topological graph nodes corresponding to the weighted attribute topological graph.

[0070] Specifically, the IFC entity type and IFC attribute set of each building structure component in the geometric element set are extracted, and then each non-zero element in the topological relationship matrix is mapped and converted into an edge of the weighted attribute topology graph, the weight value of the topological relationship matrix is taken as the connection strength coefficient, the IFC entity type and IFC attribute set of the geometric element set are bound in the topology graph node corresponding to the weighted attribute topology graph, and the weighted attribute topology graph containing complete IFC semantic annotation is generated.

[0071] It needs to be explained that the IFC entity type refers to the standard defined building component classification, specifically including: structural components, building envelope components, equipment components, and auxiliary components, wherein the structural components include but are not limited to: walls, structural columns, structural beams, and floors, etc.; the building envelope components include but are not limited to: doors, windows, and curtain walls, etc.; the equipment components include but are not limited to: air pipes and cable bridges, etc.; the auxiliary components include but are not limited to: stairs, ramps, and railings, etc. The IFC attribute set refers to the engineering parameter set bound to the IFC entity type, specifically including: material properties, structural properties, construction properties, and custom attribute sets, wherein the material properties include but are not limited to: strength grade, fire rating, and thermal conductivity, etc.; the structural properties include but are not limited to: cross-sectional size, span, and load grade, etc.; the construction properties include but are not limited to: construction stage and prefabrication mark, etc.; the custom attribute set refers to the extended attribute, including but not limited to: unit price and supplier information, etc.

[0072] It needs to be explained that the non-zero element mapping conversion process is to convert the non-zero element in the topological relationship matrix into the directed edge between the connection nodes in the weighted attribute topology graph, and directly assign the data value of the non-zero element as the weight attribute of the edge, while recording the engineering meaning of the original matrix in the weight attribute of the edge, completing the conversion of the weighted attribute topology graph.

[0073] It needs to be explained that the complete IFC semantic annotation refers to the engineering information system carried by each node and edge in the weighted attribute topology graph, which conforms to the IFC standard, specifically including: node annotation and edge annotation, wherein the node annotation represents the engineering characteristics of defining building structure components, determines the functional classification of engineering characteristics in the building process, including but not limited to: IFC entity type, IFC attribute set, and component unique identifier; the edge annotation represents the digital definition of the engineering relationship between components in the weighted attribute topology graph, the core is to accurately describe the interaction between building elements through standardized data structure, including but not limited to: connection type, connection strength coefficient, and engineering basis.

[0074] S1.3: merging adjacent three-dimensional nodes of the weighted attribute topology graph based on a spatial coincidence threshold, constructing an R-tree three-dimensional spatial index, and converting the R-tree three-dimensional spatial index into a GraphML format three-dimensional topology data network.

[0075] Specifically, all three-dimensional nodes of the weighted attribute topology graph are traversed, the spatial bounding box coincidence of each pair of adjacent three-dimensional nodes is calculated, and if the spatial bounding box coincidence of the adjacent three-dimensional nodes exceeds a preset spatial coincidence threshold, the three-dimensional nodes are merged into a new three-dimensional node, wherein the weight of the new three-dimensional node is the average of the weights of the merged adjacent three-dimensional nodes. The topology edge relationship is updated after merging, and then an R-tree three-dimensional spatial index is constructed based on the spatial distribution of the merged three-dimensional nodes. The minimum circumscribed cube of each new three-dimensional node is stored as a leaf node of the R-tree three-dimensional spatial index. Finally, the hierarchical structure and node spatial coordinates of the R-tree three-dimensional spatial index are converted into a GraphML format three-dimensional topology data network.

[0076] The formula for calculating the spatial bounding box coincidence is as follows:

[0077] ;

[0078] wherein, and represent the axis-aligned bounding boxes of the three-dimensional objects of the three-dimensional nodes, represents the spatial bounding box coincidence between the axis-aligned bounding boxes of the two three-dimensional nodes and represents a spatial dimension variable, represents the length direction corresponding to the engineering building, represents the width direction corresponding to the engineering building, represents the vertical direction corresponding to the engineering building, represents the maximum boundary coordinates of the axis-aligned bounding box in the spatial dimension , represents the maximum boundary coordinates of the axis-aligned bounding box in the spatial dimension , represents the minimum boundary coordinates of the axis-aligned bounding box in the spatial dimension , represents the minimum boundary coordinates of the axis-aligned bounding box in the spatial dimension .

[0079] ​It needs to be explained that the preset spatial coincidence threshold is a critical value for judging whether adjacent three-dimensional nodes need to be merged, and the merging operation is triggered when the spatial bounding box coincidence calculation result of two three-dimensional nodes exceeds the preset spatial coincidence threshold; the construction of the preset spatial coincidence threshold is based on the engineering precision requirement, hardware computing resources and industry standard.

[0080] It needs to be explained that the conversion process of the three-dimensional topological data network is to convert the leaf node storage of the R-tree three-dimensional spatial index into the node element of GraphML, the node ID remains unique, the cubic range coordinates are converted into the "bbox" attribute key-value pair, then all the structure nodes of the R-tree three-dimensional spatial index are converted into the nested graph structure element of GraphML, forming a hierarchical relationship, and the spatial inclusion relationship between the R-tree nodes is converted into the edge element of GraphML, wherein the source node and the target node of the edge element of GraphML point to the parent graph structure and the child node respectively; at the same time, all the attributes (such as IFC type, material parameter) of the topological graph node corresponding to the weighted attribute topological graph are converted into the additional data element of GraphML, and finally the GraphML standard header information is added to complete the format conversion of the three-dimensional topological data network.

[0081] S1.4: Check the node connectivity and attribute integrity of the three-dimensional data topological network, generate a building structure component connectivity graph and an attribute distribution report, and perform packaging to obtain a graph structure-based three-dimensional data topological network package.

[0082] Specifically, all three-dimensional nodes of the three-dimensional data topological network are traversed, the number of connection edges of each three-dimensional node is detected, and the attribute field integrity is verified, the three-dimensional nodes and edge data that pass the verification are generated, and based on the three-dimensional nodes and edge data that pass the verification, each three-dimensional node is mapped to a connectivity graph vertex, the node ID and IFC entity type are retained, the three-dimensional node attributes are extracted as vertex additional data, the edge data that pass the verification are converted into undirected edges of the connectivity graph, and a building structure component connectivity graph is constructed, based on the attribute distribution report, finally the building structure component connectivity graph, the attribute distribution report and the three-dimensional topological data network are packaged into a graph structure-based three-dimensional data topological network package conforming to the GraphML specification.

[0083] It needs to be explained that the specific verification of the attribute field integrity is to check whether the attribute field of each three-dimensional node in the three-dimensional data topological network contains the required IFC semantic attribute, and to verify whether the valid engineering parameter interval of each three-dimensional node in the three-dimensional data topological network conforms to the multi-dimensional engineering constraint, and mark the three-dimensional nodes with missing fields or out-of-range engineering parameter intervals as verification failures.

[0084] The multi-dimensional engineering constraints include engineering geometric parameter constraints, engineering structure type constraints, engineering material attribute constraints, engineering construction stage constraints, engineering mechanics and engineering safety constraints.

[0085] It should be explained that the attribute distribution report is a document for statistical analysis of attribute values of all verified nodes in the three-dimensional topological data network, recording the numerical distribution frequency and classification proportion of different IFC semantic attributes; the attribute distribution report contains the completeness rate of mandatory fields, such as the cross-sectional size field completeness rate, and the interval distribution of engineering quantitative parameters with units (such as interface size, connection strength, and load level).

[0086] S2: Perform multi-dimensional reasoning operation on the three-dimensional data topological network package based on the graph structure to obtain a deviation compensation strategy containing an optimal cost control path.

[0087] Specifically, please refer to Figure 3 , the steps are as follows:

[0088] S2.1: Extract topological node and topological edge data from the three-dimensional data topological network package based on the graph structure, then combine the time window constraint, expand the spatial coordinates of the building structure component topological node into 4D space-time parameters, and use the preset resource cost parameter table as a dynamic correction factor to obtain an enhanced topological network with disturbance factors.

[0089] Specifically, the topological node data and topological edge data are parsed from the three-dimensional data topological network package based on the graph structure, the spatial coordinates of the topological node and the IFC semantic attributes are extracted, and the connection relationship and weight of the topological edge are extracted; combined with the time window constraint, the spatial coordinates of the topological node are expanded into 4D space-time parameters, while the connection relationship of the topological edge data is retained, then the preset resource cost parameter table is converted into a dynamic correction factor, matched to the corresponding topological node and topological edge data according to the component type, and through the superposition operation of the space-time parameters and the dynamic correction factor, an enhanced topological network with disturbance factors is generated.

[0090] It should be explained that the time window constraint refers to a discrete time node defined based on the construction progress plan or the engineering stage division, which is used to expand the three-dimensional spatial coordinates into 4D space-time parameters; the time window constraint exists in the form of standard time format or relative duration, which forcibly limits the validity of the spatial state of the building component within a certain time interval.

[0091] It needs to be explained that the preset resource cost parameter table is a structured data table, and the core function is to store resource cost parameter information associated with different building structure component types, to ensure accurate matching in the enhanced topology network generation process with disturbance factors; The preset resource cost parameter table is constructed according to the bill of quantities decomposition items, resource consumption quota standards and engineering contract supplementary clauses, wherein the preset resource cost parameter table contains component type matching field, cost parameter value and parameter constraint condition related structured data.

[0092] S2.2: Based on the enhanced topology network with disturbance factors, propagate the disturbance influence along the topology edge, and iteratively obtain the stability disturbance guarantee value, then perform multi-dimensional scenario simulation on the stability disturbance guarantee value, and generate a three-dimensional diagnostic report.

[0093] Specifically, based on all topology edges in the enhanced topology network with disturbance factors, the starting three-dimensional disturbance factor is transmitted to the target three-dimensional node along the topology edge direction, and the disturbance value of all nodes is iteratively updated by weighted accumulation operation until convergence, to obtain the stability disturbance guarantee value, then the multi-dimensional scenario simulation is performed on the stability disturbance guarantee value, the disturbance response of the topology network under different scenarios is recorded, and a three-dimensional diagnostic report containing node displacement, edge weight change rate and other indicators is generated.

[0094] It needs to be explained that the determination of the target three-dimensional node is defined by the directionality of the topology edge. The starting node and the target node of each topology edge are specified in the enhanced topology network, and all topology edges are traversed. According to the preset direction of each topology edge, the starting node and the target node are extracted, the disturbance factor of the starting node is weighted and accumulated to the current disturbance value of the target node through the edge weight, and the disturbance value of all nodes is updated.

[0095] It needs to be explained that the multi-dimensional scenario simulation operation generates multiple disturbance scenes by adjusting the combination of disturbance parameters based on multi-dimensional engineering constraints and historical engineering data, re-obtains the stability disturbance guarantee value of the topology network under each scene, and records the key response indicators of nodes and edges; The specific operation content is: select the core parameters that affect the disturbance factor (such as material cost fluctuation and progress compression rate, etc.), generate orthogonal experimental combinations, and perform disturbance propagation iteration on each orthogonal experimental combination to obtain new stability disturbance guarantee value, then extract the node displacement and edge weight change rate and other quantitative data in the new stability disturbance guarantee value, integrate the index values according to the scene code, and generate a three-dimensional diagnostic report.

[0096] S2.3: Based on the three-dimensional diagnostic report, construct a multi-objective optimization function, then use the Pareto front algorithm to select the optimal non-dominated solution set from the multi-objective optimization function, and generate a deviation compensation strategy containing the optimal cost control path.

[0097] Specifically, based on the node displacement amount and edge weight change rate and other parameters in the three-dimensional diagnostic report, a multi-objective optimization function including cost control target, structure safety target and progress target is constructed, and the Pareto frontier algorithm is used to traverse the solution space, filter the optimal non-dominated solution set that meets all constraint conditions and is not dominated by each other, and finally extract the cost control path from the optimal non-dominated solution set to generate the deviation compensation strategy containing resource adjustment rules and compensation amount.

[0098] The formula of the Pareto frontier algorithm is as follows:

[0099] ;

[0100] Among them, represents the optimal non-dominated solution set, represents the regulation and control parameter combination of the deviation compensation strategy, represents the candidate solution, represents the solution space defined by the multi-objective optimization function, represents the cost control target function of the compared solution, represents the structure safety target function of the compared solution, represents the progress target function of the compared solution, represents the cost control target function of the regulation and control parameter combination, represents the structure safety target function of the regulation and control parameter combination, represents the progress target function of the regulation and control parameter combination.

[0101] It needs to be explained that the judgment of meeting the constraint condition means that the candidate solution needs to meet the following engineering boundary restrictions: the node displacement amount does not exceed the upper limit allowed by the project, the edge weight change rate is not lower than the safety threshold, and the resource adjustment amount meets the range of the contract bill of quantities; The judgment of mutual non-dominance is that the mutual comparison between candidate solutions, under the premise of meeting the constraint conditions, when there is no other candidate solution that is better than the current candidate solution in all targets, the current candidate solution is taken as a non-dominated solution.

[0102] The range of the contract bill of quantities refers to the structured resource quantization table formed according to the number of building structural components, material consumption and construction process quantity listed in the bidding documents, construction contract and cost specification, which is associated with BIM component ID after structured processing, and accurately mapped to each building structural component in the three-dimensional building information model, thereby limiting the range of each resource adjustment amount.

[0103] S3: Generate 5D cost dimension data set by dimension expansion and data fusion on 3D building information model.

[0104] Specifically, the steps are as follows:

[0105] S3.1: Process the geometric element set of the 3D building information model using the replacement simplification method, and generate a lightweight geometric data set according to the geometric entities and attribute entities of the 3D building information model.

[0106] Specifically, the replacement simplification method based on component type and bounding box is used to process the geometric element set of the 3D building information model, the original geometric precision is retained for key structural components, and the axis-aligned bounding box is extracted from non-key structural components and replaced with the original geometry, while filtering non-standard annotations according to the geometric entities and attribute entities of the 3D building information model, and finally generating a lightweight geometric data set.

[0107] It needs to be explained that the key determination of structural components is to automatically identify the components that play a decisive role in the overall safety, stability and load-bearing of the building structure, such as load-bearing beams, columns, core wall, etc., by analyzing the structural type mark in the IFC semantic attribute, the mechanical analysis parameters (such as bending moment, shear force, axial force, etc.) and the requirements of building structure specifications for building structural components, and defining the components that play a decisive role as key structural components.

[0108] It needs to be explained that the replacement simplification method based on component type and bounding box is to realize the lightweight of 3D building information model through engineering semantic driven geometric dimension reduction technology, to determine the importance of components according to the IFC entity type of components, to retain the original geometric precision of key structural components, to extract the spatial bounding box of non-key structural components, to replace the original complex geometry with the smallest circumscribed cube, and to filter the annotation information in the geometric entity that does not conform to the IFC standard, and finally to generate a lightweight geometric data set.

[0109] It needs to be explained that the geometric entity refers to the structured data of the spatial topology and geometric form of the 3D building information model, which accurately describes the spatial position, shape and size relationship of the building components, including: basic geometric elements, spatial relationships and precision levels; the attribute entity refers to the engineering semantic data system bound to the geometric entity, which establishes a structured description of component characteristics, functions and behaviors based on IFC standard, including: classification identification, engineering parameter set and constraint rule.

[0110] S3.2: Associate the construction progress plan with the building structure component ID, bind the installation time node to the building structure component in the lightweight geometric data set, and parse the external bill of quantities to match the building structure component geometric quantity in the lightweight geometric data set to obtain the cost benchmark matrix.

[0111] Specifically, by associating the installation time node in the construction progress plan with the building structure component ID, the installation time node is bound to the attribute field of the corresponding building structure component in the lightweight geometric data set. Then, the external bill of quantities is parsed, the geometric quantities of the building structure components in the lightweight geometric data set are matched, and a two-dimensional index matrix is established based on the component ID and the time node. The matched quantities and unit costs are filled into the two-dimensional index matrix to generate the cost benchmark matrix.

[0112] It should be explained that the construction progress plan is a construction task execution scheme structured and arranged based on the time dimension in the engineering project, and the core is to accurately control the installation timing and engineering stage division of the building structure component through discrete time nodes.

[0113] It should be explained that the external bill of quantities is a structured resource quantification table constructed based on the cost specification, which accurately associates the geometric quantities, resource consumption rules and cost parameters of the building structure component. The construction of the external bill of quantities is to integrate the geometric quantities, resource consumption rules and cost parameters into a structured measurement system with component ID as the link, supporting the accurate generation of the cost benchmark matrix.

[0114] S3.3: Adopting gradient descent algorithm to dynamically correct the cost of 4D space-time data set and cost benchmark matrix, generating 5D cost dimension data set.

[0115] Specifically, the gradient descent algorithm is adopted to dynamically correct the cost of the 4D space-time data set and the cost benchmark matrix, the cost correction factor is initialized as the benchmark ratio parameter, and the loss function is defined to quantify the overall difference between the cost benchmark matrix value and the predicted cost of the 4D space-time data set. Then, based on the gradient direction of the loss function, the correction factor value is gradually adjusted combined with the preset learning rate parameter. After each iteration, the loss change is re-evaluated, and finally the converged correction factor is fused with the original 4D space-time data set to expand and form a 5D cost dimension data set containing spatial three-dimensional coordinates, time dimension and dynamic correction cost value.

[0116] The formula of the gradient descent algorithm is as follows:

[0117] ;

[0118] Wherein, represents the iteration index, represents the discrete variable of component traversal, represents the learning rate parameter, represents the total number of building structure components, represents the cost correction factor of the first iteration, represents the cost correction factor of the first iteration, represents the cost correction factor of the first iteration, represents the cost correction factor of the first iteration, represents the cost correction factor of the first iteration, a predicted cost value of the component, a reference cost value of the component. a reference cost value of the component.

[0119] It should be explained that the setting of the preset learning rate parameter is based on the magnitude characteristics of engineering cost data, convergence stability requirements and hardware resource limitations, and is determined by the following core principles: cost data dispersion analysis, convergence stability constraints, calculation real-time resource boundaries and engineering experience rules.

[0120] It should be explained that the 5D cost dimension dataset refers to a 5D structured data matrix formed by dynamically correcting and integrating cost quantitative parameters based on the 4D space-time dataset, dynamically associating static cost benchmarks with space-time progress, and realizing precise space-time mapping of construction project resource costs.

[0121] S4: Analyze the resource adjustment rules in the deviation compensation strategy containing the optimal cost control path, build a resource allocation optimization model, and optimize the resource allocation optimization model using the gradient descent algorithm, and generate a set of construction scheduling instructions that minimize cost disturbance factors based on the 5D cost dimension dataset.

[0122] Specifically, the steps are as follows:

[0123] S4.1: Analyze the resource type, adjustment direction and adjustment amount in the deviation compensation strategy containing the optimal cost control path, convert the resource type, adjustment direction and adjustment amount into mathematical constraint conditions, and obtain a structured resource adjustment rule table.

[0124] Specifically, the resource type, adjustment direction and adjustment amount in the deviation compensation strategy containing the optimal cost control path are analyzed, the resource type name, adjustment direction and adjustment amount value are extracted, and then the resource type name is converted into a mathematical variable through resource type standardization mapping; according to the adjustment direction conversion operator rule, the adjustment direction word is mapped to a mathematical operator; according to the adjustment amount conversion mathematical coefficient rule, the adjustment amount value is converted to a decimal coefficient, and finally integrated to generate mathematical constraint conditions and stored in a structured table form to generate a structured resource adjustment rule table.

[0125] It should be explained that resource type standardization is a standardized process of converting engineering resource names into mathematical variable symbols, and the core of resource type standardization is to establish a unique correspondence between engineering terminology and digital variables. For example: the engineering resource name is concrete, and according to the first letter of the English name of concrete, the digital variable is C.

[0126] It needs to be explained that the adjustment direction conversion operator rule is a standardized conversion that maps natural language instructions to mathematical operators, and the essence of the adjustment direction conversion operator rule is to define the logical equivalence relationship between instructions and operators. For example, the adjustment direction word is "append", and the mathematical operator is defined as "+".

[0127] It needs to be explained that the rule of converting adjustment quantities to mathematical coefficients refers to the quantification process of converting engineering numerical descriptions into mathematical coefficients or constants. The core of the rule of converting adjustment quantities to mathematical coefficients is to unify the mathematical expression of engineering numerical values. For example, if the adjustment quantity is described as a "percentage", it is converted into a "decimal" mathematical coefficient; if the adjustment quantity is described as an "absolute value", it is converted into a "constant" mathematical coefficient.

[0128] S4.2: Using the resource adjustment rule table as a constraint and the dynamic weight parameters in the 5D cost dimension dataset as decision variables, construct a resource allocation optimization model with minimizing the cost disturbance factor as the objective function, and use the gradient descent algorithm to solve the objective function and update the dynamic weight parameters to obtain the optimized resource allocation optimization model.

[0129] Specifically, the mathematical constraints in the resource adjustment rule table are bound to the dynamic weight parameters in the 5D cost dimension dataset to construct an objective function that minimizes the cost disturbance factor. The gradient descent algorithm is used to iteratively solve this objective function. The weight parameters are initialized to baseline values. The direction of change of the objective function is calculated, and the dynamic weight parameters are adjusted accordingly. This calculation and adjustment process is repeated. When the gradient magnitude of the dynamic weight parameters during repeated calculations is less than the convergence threshold, the direction of change of the objective function tends to stabilize (based on data precision requirements and floating-point precision limitations). The optimal combination of dynamic weight parameters that satisfies all constraints is obtained, generating the optimized resource allocation optimization model. The expression for the objective function is:

[0130] ;

[0131] in, The objective function is an instruction that declares that the objective of the entire mathematical expression is to minimize it. Represent decision variables; Indicates the first The constructed predicted cost values ​​are derived from the 5D cost dimension dataset; Indicates the first The baseline cost value for each construction.

[0132] It needs to be explained that the satisfaction of the constraint condition is verified synchronously after each update of the dynamic weight parameter in the iteration process of the gradient descent algorithm. The judgment standard is divided into three categories: equality constraint judgment standard, inequality constraint boundary judgment standard and engineering physics constraint judgment standard. Among them, the equality constraint judgment standard: the equality relationship in the resource adjustment rule (for example: the additional proportion of concrete) needs to meet the difference between the left and right values less than the engineering specified error threshold; the inequality constraint boundary judgment standard: the resource boundary limit (for example: the lower limit of steel consumption) needs to meet the adjusted value not less than the constraint boundary threshold; the engineering physics constraint judgment standard: the weight parameter must meet the basic requirements of the engineering feasible region (for example: the resource allocation is negative). When all types of constraints are met at the same time, the dynamic weight parameter combination is accepted as a valid solution.

[0133] It needs to be explained that the constraint boundary threshold refers to the upper and lower limits of material consumption and component quantity in the engineering cost specification, construction technology specification and construction contract, which are determined based on each type of engineering resource, and are corrected in combination with the geometric quantity of the component and the construction feasibility parameter.

[0134] S4.3: Through the matching of the 5D cost dimension data set with the space-time coordinates of the building structure component ID, the optimal dynamic weight parameter combination is injected into the space-time node, and then the structured coding task instruction and the cost disturbance warning instruction are added. After encapsulation, the construction scheduling instruction set with the minimum cost disturbance factor is obtained.

[0135] Specifically, through the precise matching of the 5D cost dimension data set with the space-time coordinates of the building structure component ID, the target space-time node is located, and the optimal dynamic weight parameter combination is injected as a new attribute into the attribute field of the corresponding space-time node. Then, the structured coding rule is used to generate the task instruction, and the cost disturbance warning instruction is added. The cost deviation threshold and the response action are defined. Finally, the target space-time node, the task instruction and the cost disturbance warning instruction are encapsulated to generate the construction scheduling instruction set with the minimum cost disturbance factor.

[0136] It needs to be explained that the cost deviation threshold is determined based on the contract bill of quantities and cost specification, the baseline cost and constraint boundary threshold of each type of engineering resource, and the historical engineering data and market price fluctuation. The typical interval of cost disturbance is calculated and the acceptable risk tolerance range is extracted. Then, according to the characteristics of the engineering project (for example: the tightness of the construction period and the stability of resource supply, etc.), the dynamic monitoring interval is set on the intersection of the baseline cost and the risk tolerance range as the cost deviation threshold.

[0137] It needs to be explained that the structured coding rule is a construction task instruction standardization conversion mechanism based on preset field format. Through the forced separator, the task type code, component ID, space-time coordinate and resource parameter are converted into machine-readable instruction string, and the value domain constraint (such as coordinate reservation three decimal places, timestamp ISO8601 format) and fault tolerance rule (missing field triggers error code, illegal value replaces default value) are added, to realize the lossless analysis of construction instruction and the integration of cost disturbance early warning.

[0138] S5: Through the cognitive mirror image synchronization mode, the construction scheduling instruction set with the minimum cost disturbance factor is subjected to tensor convolution operation, to generate the reconstructed 4D space-time progress topology structure.

[0139] Specifically, please refer to Figure 4 , the steps are as follows:

[0140] S5.1: Analyze the construction scheduling instruction set with the minimum cost disturbance factor, extract the spatial position, time node and optimal dynamic weight parameter combination, and then map the spatial position, time node and optimal dynamic weight parameter combination to the space-time resource matrix.

[0141] Specifically, the construction scheduling instruction set with the minimum cost disturbance factor is analyzed, the spatial position coordinate, time node and optimal dynamic weight parameter combination of each instruction item are extracted, the row and column indexes of the space-time resource matrix are constructed based on the spatial position coordinate and time node, the optimal dynamic weight parameter combination is split into independent vectors according to the resource type, and filled into the corresponding unit cells of the space-time resource matrix, the quantization allocation value is obtained, and finally the space-time resource matrix is generated by using the discretization grid mapping method based on the row and column indexes and the quantization allocation value of the space-time resource matrix.

[0142] It needs to be explained that the discretization grid mapping method is an engineering method of converting the row and column indexes and the quantization allocation value of the space-time resource matrix into discrete row and column indexes, and storing the resource allocation data based on the grid structure; the operation steps of the discretization grid mapping method are that the row and column indexes of the space-time resource matrix are used as the unique row index, then independent two-dimensional tables are created according to each resource type, the quantization allocation value is filled into the row and column intersection cells of the corresponding independent two-dimensional table, and the space-time resource matrix is obtained.

[0143] S5.2: Based on the space-time resource matrix, a three-dimensional convolution kernel is constructed, and then the three-dimensional convolution kernel is subjected to tensor convolution operation, to generate the space-time resource matrix after convolution optimization.

[0144] Specifically, the size of the convolution kernel is determined based on the space-time-hardware three-dimensional coupling, the space three-dimensional coordinates and the time dimension in the space-time resource matrix are covered, the distribution of the space three-dimensional coordinates and the time dimension is distributed according to the central diffusion attenuation, the three-dimensional convolution kernel is constructed, and then the three-dimensional convolution kernel and the space-time resource matrix are obtained through full-dimensional convolution operation to obtain the convolution optimized space-time resource matrix.

[0145] It should be explained that the space-time-hardware three-dimensional coupling includes three factors of engineering structure characteristics, construction stage granularity and computing resource constraints, wherein the engineering structure characteristics are set according to the structure geometric density and resource interaction range, the construction stage granularity is set according to the construction rhythm and process continuity requirement, and the computing resource constraint is set according to the real-time response time and memory upper limit.

[0146] It should be explained that the full-dimensional convolution operation is a sliding window weighting operation on a 4D tensor (space three-dimensional and time dimension), and the core of the full-dimensional convolution operation is to synchronize the sliding of the three-dimensional convolution kernel in all dimensions of the space-time resource matrix, to realize the space-time continuity optimization of resource allocation.

[0147] S5.3: Convert each coordinate node in the convolution optimized space-time resource matrix into a topology vertex, connect the topology vertices based on the resource flow path, form a topology edge, and assign a weight value to the topology edge according to the resource consumption intensity, to generate a reconstructed 4D space-time progress topology structure.

[0148] Specifically, each coordinate node in the convolution optimized space-time resource matrix is directly mapped to a topology vertex, wherein the attributes of the topology vertex include the original space coordinates and the timestamp, and based on the preset resource flow path rule, the topology vertices with resource flow relationship are connected to form a directed topology edge, then the topology edge weight value is calculated according to the resource consumption intensity and the flow distance, and finally the reconstructed 4D space-time progress topology structure is generated through the vertex set, the directed topology edge and the attributes of the topology vertex.

[0149] The formula for calculating the topology edge weight value is as follows:

[0150] ;

[0151] Wherein, represents the start vertex of each coordinate node in the convolution optimized space-time resource matrix, represents the end vertex of each coordinate node in the convolution optimized space-time resource matrix, represents the resource intensity weight coefficient, represents the distance attenuation weight coefficient, represents the topology edge weight from the start vertex to the end vertex ​Represents the starting vertex To the final vertex The flow distance, Indicates the maximum permissible flow distance. Indicates the starting vertex Resource consumption per unit time Represents the endpoint vertex Resource consumption per unit of time.

[0152] It should be explained that the preset resource flow path rules are topological connection criteria based on engineering constraints and process logic. By quantifying spatial reachability, temporal continuity, and process dependencies, the rules determine whether there are resource flow paths between topological vertices and generate directed edges. The preset resource flow path rules determine the resource flow paths between topological vertices through triple constraints of engineering equipment capabilities, construction rhythm, and process specifications.

[0153] It needs to be explained that the 4D spatiotemporal schedule topology refers to a dynamic graph network that integrates spatial three-dimensional coordinates and time dimension, generated by tensor convolution operations. The core of the 4D spatiotemporal schedule topology consists of topological vertices and directed edges. Furthermore, the 4D spatiotemporal schedule topology transforms the discrete parameters of the construction scheduling instruction set into a spatiotemporal resource matrix after convolution optimization. By constraining the spatial proximity and temporal continuity of resource flow paths, it dynamically optimizes the construction process.

[0154] S6: Update facility status records based on the reconstructed 4D spatiotemporal progress topology, and generate a 6D full lifecycle data record table by combining readings from IoT sensing devices.

[0155] Specifically, the steps are as follows:

[0156] S6.1: Based on the building structure component ID, accurately associate the topological vertices in the reconstructed 4D spatiotemporal progress topology with the on-site physical facilities, extract the resource flow paths corresponding to the topological vertices, then receive and verify IoT device data in real time, map the IoT device data to the corresponding topological vertices, and generate an enhanced topology with real-time status.

[0157] Specifically, based on the building structure component ID, the topological vertices in the reconstructed 4D space-time progress topology structure are accurately associated with the field entity facilities, the corresponding relationship between the topological vertices and the field entity facilities is established, the resource flow path associated with the topological vertices is extracted, the spatial coordinate sequence of the resource flow path is recorded, then the Internet of Things device data is received in real time, when the validity of the Internet of Things device data is verified, it is verified that the Internet of Things device data value is within the preset engineering threshold range, and it is verified whether the Internet of Things device data timestamp is within the vertex time dimension interval, and the Internet of Things device data that passes the verification is mapped into the real-time state attribute of the corresponding topological vertex, the vertex attribute field is updated, and finally the enhanced topology structure with real-time state is generated.

[0158] It needs to be explained that the corresponding relationship between the topological vertices and the field entity facilities is the bidirectional mapping between the topological vertices and the physical facilities established by the building structure component ID, which specifically includes the following three-dimensional accurate association: identifier unique binding, space-time state synchronization, and resource flow path entity.

[0159] It needs to be explained that the preset engineering threshold range is set according to the building structure specification, construction process requirements and mathematical constraint conditions, and is corrected in combination with historical engineering data and experience statistical results of the same type of building structure components. Finally, the preset engineering threshold range not only ensures the engineering rationality of the Internet of Things device data, but also serves as a judgment standard for real-time state verification.

[0160] It needs to be explained that the Internet of Things device data refers to the engineering physical quantity and device state data collected in real time by field sensors and controllers, and the core is the real-time information flow connecting physical facilities and digital topology; wherein, the Internet of Things device data includes: resource consumption data, environment and state data, and spatial positioning data.

[0161] S6.2: Integrate construction period data, associate the topological vertices of the enhanced topology structure with real-time state with construction logs and quality detection reports, and build a 6D full life cycle data record table with building structure component ID as the core node.

[0162] Specifically, by integrating construction period data, the topological vertices of the enhanced topology structure with real-time state are accurately associated with construction log entries and quality detection reports through the building structure component ID, a mapping relationship among the component ID, log entry and report number is established, and the topological vertex real-time state attribute, construction log operation record and quality detection report parameter are extracted. Subsequently, the topological vertex real-time state attribute is converted into real-time state data, the construction log operation record is converted into a construction log entry index, and the quality detection report parameter is converted into a quality detection report number, and a 6D full life cycle data record table is built.

[0163] It needs to be explained that the construction log entry is a digital construction operation record unit with the building structure component ID as the unique index, which is created in real time when the process starts and dynamically updated with the construction progress; the construction log entry is a structured record automatically generated at the beginning of the construction operation, which specifically includes spatial coordinates, timestamps, operation types (such as pouring), resource consumption, equipment parameters, and responsible person information, and is bound to the topological vertex corresponding to the component ID and time window through the component ID and time window, and finally converted into an entry index.

[0164] It needs to be explained that the quality detection report is a digital quality inspection result file with the building structure component ID as the unique index, which automatically compares technical specification standards to generate a judgment result by real-time collection of physical parameters through on-site sensors, and binds the construction log entry index and the supervision electronic seal to form a structured record.

[0165] It needs to be explained that the 6D full life cycle data record table is indexed with the building structure component ID as the core, based on the 5D cost dimension data set, combined with historical engineering data and real-time state data, through the fusion of resource consumption records and facility long-term state evolution, the resource flow path is continuously tracked in the spatial correlation and time continuity.

[0166] The embodiment also provides a construction project whole-process management system based on BIM3D-6D technology, comprising:

[0167] The topology modeling module loads the 3D building information model through the BIM lightweight computing server, extracts the spatial topology relationship and IFC semantic attribute of the geometric component, and performs spatial topology analysis and semantic attribute mapping on the spatial topology relationship and IFC semantic attribute of the geometric component, and generates a three-dimensional data topology network package based on a graph structure;

[0168] The path deduction module performs multi-dimensional deduction operation on the three-dimensional data topology network package based on the graph structure to obtain a deviation compensation strategy containing an optimal cost control path;

[0169] The dimension fusion module expands and fuses data of the 3D building information model to generate a 5D cost dimension data set;

[0170] The resource optimization module analyzes the resource adjustment rules in the deviation compensation strategy containing the optimal cost control path, constructs a resource allocation optimization model, and optimizes the resource allocation optimization model using the gradient descent algorithm, and generates a construction scheduling instruction set with a minimum cost disturbance factor based on the 5D cost dimension data set;

[0171] The structure reconstruction module performs tensor convolution operation on the construction scheduling instruction set with the minimum cost disturbance factor through the cognitive mirror synchronization mode to generate a reconstructed 4D space-time progress topology structure;

[0172] The closed-loop feedback module updates the facility state record based on the reconstructed 4D space-time progress topology, and generates a 6D full life cycle data record table in combination with the readings of the Internet of Things sensing device.

[0173] The embodiment also provides a computer device, including 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 realize the construction project whole-process management method based on the BIM 3D-6D technology as a base proposed in the above embodiment.

[0174] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The communication interface of the computer device is used to perform wired or wireless communication with an external terminal. The wireless communication can be achieved through WIFI, an operator network, NFC (Near Field Communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0175] The embodiment also provides a storage medium having a computer program stored thereon. The program is executed by a processor to realize the construction project whole-process management method based on the BIM 3D-6D technology as a base proposed in the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic storage, a flash memory, a magnetic disk or an optical disk.

[0176] To sum up, the application carries out tensor convolution operation on the construction scheduling instruction set through the cognitive mirror synchronization mode, maps the discrete scheduling instruction into a continuous space-time topology structure, realizes space-time smooth diffusion of the resource path by using the weight attenuation characteristics of the three-dimensional convolution kernel, generates a reconstructed 4D space-time progress topology structure, and converts the discrete decision into a continuous field model through mathematical transformation, so that the resource flow path has spatial correlation and time continuity, and the beneficial effect of dynamically optimizing the construction process is achieved.

[0177] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. A method for whole-process management of a construction project based on BIM 3D-6D technology, characterized in that: The application relates to a BIM lightweight computing server-based 3D building information model loading method. The method comprises the following steps: loading a 3D building information model through a BIM lightweight computing server, extracting the spatial topological relationship and IFC semantic attribute of a geometric component, and performing spatial topological analysis and semantic attribute mapping on the spatial topological relationship and IFC semantic attribute of the geometric component to generate a three-dimensional data topological network package based on a graph structure; performing multi-dimensional deduction operation on the three-dimensional data topological network package based on the graph structure to obtain a deviation compensation strategy containing an optimal cost control path; expanding the 3D building information model in dimensions and fusing data to generate a 5D cost dimension data set; analyzing the resource adjustment rules in the deviation compensation strategy containing the optimal cost control path, constructing a resource allocation optimization model, and optimizing the resource allocation optimization model by using a gradient descent algorithm to generate a construction scheduling instruction set with a minimum cost disturbance factor based on the 5D cost dimension data set; performing tensor convolution operation on the construction scheduling instruction set with the minimum cost disturbance factor in a cognitive mirror synchronization mode to generate a reconstructed 4D space-time progress topological structure; 2. The construction project whole-process management method based on BIM 3D-6D technology according to claim 1, characterized in that: updating the facility state record based on the reconstructed 4D space-time progress topological structure, and combining the readings of the Internet of Things sensing device to generate a 6D full life cycle data record table. The method for loading a 3D building information model through a BIM lightweight computing server, extracting the spatial topological relationship and IFC semantic attribute of a geometric component, and performing spatial topological analysis and semantic attribute mapping on the spatial topological relationship and IFC semantic attribute of the geometric component to generate a three-dimensional data topological network package based on a graph structure comprises the following steps: loading a 3D building information model through a BIM lightweight computing server, simplifying a non-key structure grid to generate a geometric deformation degree and a topological relationship matrix, and scanning the 3D building information model based on the geometric deformation degree to obtain a geometric element set; analyzing the IFC semantic attribute in the geometric element set, and converting the topological relationship matrix into a weighted attribute topological graph; merging adjacent three-dimensional nodes of the weighted attribute topological graph based on a spatial coincidence degree threshold, constructing an R-tree three-dimensional space index, and converting the R-tree three-dimensional space index into a three-dimensional topological data network; 3. The construction project whole-process management method based on BIM 3D-6D technology according to claim 1, characterized in that: verifying the three-dimensional data topological network to generate a building structure component connectivity graph and an attribute distribution report, and performing encapsulation to obtain a three-dimensional data topological network package based on a graph structure. The method for performing multi-dimensional deduction operation on the three-dimensional data topological network package based on the graph structure to obtain a deviation compensation strategy containing an optimal cost control path comprises the following steps: extracting topological node and topological edge data from the three-dimensional data topological network package based on the graph structure, expanding the spatial coordinates of the building structure component into 4D space-time parameters based on time window constraints, superimposing a preset resource cost parameter table as a dynamic correction factor to generate an enhanced topological network with a disturbance factor; based on the enhanced topological network with the disturbance factor, propagating disturbance influence along the topological edge, and performing iteration to obtain a stability disturbance guarantee value, then performing multi-dimensional scenario simulation on the stability disturbance guarantee value to generate a three-dimensional diagnosis report; based on the three-dimensional diagnosis report, constructing a multi-objective optimization function, then selecting an optimal non-dominated solution set from the multi-objective optimization function to generate a deviation compensation strategy containing an optimal cost control path.

4. The construction project whole-process management method based on BIM 3D-6D technology according to claim 3, characterized in that: The preset resource cost parameter table comprises a component type matching field, a cost parameter value, and structured data related to parameter constraints.

5. The construction project whole-process management method based on BIM 3D-6D technology according to claim 1, characterized in that: The 3D building information model is expanded in dimension and fused with data to generate a 5D cost dimension data set in the following steps: The geometry element set of the 3D building information model is processed by a replacement and simplification method, and a lightweight geometry data set is generated according to the geometry entity and attribute entity of the 3D building information model; The installation time node is bound to the building structure component in the lightweight geometry data set by associating the construction progress plan with the building structure component ID, and the cost benchmark matrix is obtained by parsing the external bill of quantities and matching the building structure component geometry in the lightweight geometry data set; The 4D space-time data set and the cost benchmark matrix are dynamically corrected to generate a 5D cost dimension data set.

6. The construction project whole-process management method based on BIM 3D-6D technology according to claim 5, characterized in that: The external bill of quantities is a structured resource quantization table constructed based on the cost specification, accurately associating the geometry of the building structure component, resource consumption rules, and cost parameters.

7. The construction project whole-process management method based on BIM 3D-6D technology according to claim 1, characterized in that: The resource adjustment rules in the deviation compensation strategy containing the optimal cost control path are analyzed, a resource allocation optimization model is constructed, and the gradient descent algorithm is used to optimize the resource allocation optimization model to generate a construction scheduling instruction set with a minimum cost disturbance factor based on the 5D cost dimension data set in the following steps: The resource type, adjustment direction, and adjustment amount in the deviation compensation strategy containing the optimal cost control path are analyzed, and the resource type, adjustment direction, and adjustment amount are converted into mathematical constraints to obtain a structured resource adjustment rule table; The resource adjustment rule table is used as a constraint condition, the dynamic weight parameter in the 5D cost dimension data set is used as a decision variable, a resource allocation optimization model with a minimum cost disturbance factor as the objective function is constructed, and the gradient descent algorithm is used to solve the objective function and update the dynamic weight parameter to obtain an optimal dynamic weight parameter combination, and an optimized resource allocation optimization model is generated; The optimal dynamic weight parameter combination is injected into the space-time node by matching the space-time coordinates in the 5D cost dimension data set through the building structure component ID, then the structured coding task instruction is added with a cost disturbance warning instruction, and the construction scheduling instruction set with a minimum cost disturbance factor is obtained after encapsulation.

8. The construction project whole-process management method based on BIM 3D-6D technology according to claim 1, characterized in that: The construction scheduling instruction set with a minimum cost disturbance factor is subjected to tensor convolution operation in a cognitive mirror synchronization mode to generate a reconstructed 4D space-time progress topology in the following steps: The construction scheduling instruction set with a minimum cost disturbance factor is analyzed, the spatial position, time node, and optimal dynamic weight parameter combination are extracted, and then the spatial position, time node, and optimal dynamic weight parameter combination are mapped into a space-time resource matrix; Based on the space-time resource matrix, a three-dimensional convolution kernel is constructed, then the three-dimensional convolution kernel is subjected to tensor convolution operation to generate a convolution-optimized space-time resource matrix; Each coordinate node in the convolution-optimized space-time resource matrix is converted into a topology vertex, the topology vertices are connected based on the resource flow path to form topology edges, and the topology edges are assigned weight values according to the resource consumption intensity to generate a reconstructed 4D space-time progress topology.

9. The construction project whole-process management method based on BIM 3D-6D technology according to claim 1, characterized in that: The reconstructed 4D space-time progress topology updates the facility state record, combines the Internet of Things sensing device readings, and generates a 6D full life cycle data record table, with the following steps: Based on the association of the 4D space-time progress topology topological vertex and the field entity facility with the building structure component ID, the topological vertex resource flow path is extracted, and then the verified Internet of Things device data is fused into the corresponding topological vertex to generate an enhanced topological structure with real-time state; Integrate construction period data, associate topological vertices of enhanced topological structure with real-time state with construction log and quality inspection report, and take building structure component ID as core node to build 6D full life cycle data record table.

10. A construction project whole-process management system based on BIM 3D-6D technology, based on the construction project whole-process management method based on BIM 3D-6D technology in any one of claims 1-9, characterized in that: It includes: Topology modeling module, load 3D building information model through BIM lightweight computing server, extract spatial topological relationship and IFC semantic attribute of geometric component, and perform spatial topological analysis and semantic attribute mapping on spatial topological relationship and IFC semantic attribute of geometric component, generate three-dimensional data topology network package based on graph structure; Path inference module, perform multi-dimensional inference operation on three-dimensional data topology network package based on graph structure to obtain deviation compensation strategy containing optimal cost control path; Dimension fusion module, expand and fuse data of 3D building information model by dimension to generate 5D cost dimension data set; Resource optimization module, analyze resource adjustment rules in deviation compensation strategy containing optimal cost control path, construct resource allocation optimization model, and optimize resource allocation optimization model by gradient descent algorithm to generate construction scheduling instruction set with minimum cost disturbance factor based on 5D cost dimension data set; Structure reconstruction module, perform tensor convolution operation on construction scheduling instruction set with minimum cost disturbance factor by cognitive mirror synchronization to generate reconstructed 4D space-time progress topology; Closed-loop feedback module, based on the reconstructed 4D space-time progress topology, update the facility state record, combine the Internet of Things sensing device readings, and generate a 6D full life cycle data record table.

Images