BIM model and construction progress, cost data of heterogeneous fusion modeling method and system

Abstract

The application discloses a BIM model and construction progress, cost data heterogeneous fusion modeling method and system, relates to the building information model technical field, and the heterogeneous fusion modeling method comprises the following steps: analyzing a BIM source model, a construction progress plan and a cost list, extracting geometric, time sequence and cost characteristic vectors, and mapping to a unified characteristic space to generate characteristic points; a coupling degree tensor field containing time, space and cost coupling factors is constructed based on the characteristic points; under the constraint of the coupling degree tensor field, the total energy functional containing geometric deformation energy, progress deviation energy and cost potential energy is minimized, and an initial fusion model is generated; the coupling degree tensor field is updated in real time by accessing Internet of Things data and a cost index, and self-reconstruction is triggered when the energy gradient exceeds a threshold value, so that associated attributes are automatically adjusted. The application realizes deep intrinsic fusion of geometry, progress and cost, has dynamic self-adapting and intelligent early warning capabilities, and can be applied to whole life cycle management of building engineering.

Description

Technical Field

[0001] This invention relates to the field of building information modeling technology, and more specifically, to a heterogeneous fusion modeling method and system for BIM models and construction progress and cost data. Background Technology

[0002] Building Information Modeling (BIM) technology, as a core support for the digital transformation of building engineering, has continuously evolved from 3D geometric modeling to multi-dimensional (nD) applications. Among these advancements, integrating the time dimension (4D) and cost dimension (5D) with the 3D model to achieve dynamic simulation and cost control of the construction process has become an inevitable trend in the industry. Currently, mainstream 4D / 5D BIM applications typically employ an "external" association method based on ID mapping or code matching: the construction schedule is imported into the BIM software, and model components are linked to work breakdown structure (WBS) task items manually or semi-automatically; then, the cost estimate is imported, and components are matched with WBS items using codes. In this model, the 3D geometric model, construction schedule, and cost estimate are essentially independent datasets, with only a loose correspondence established through external key-value pairs.

[0003] However, existing integration methods face many technical bottlenecks in practical applications, which seriously restrict the effectiveness of BIM multidimensional applications.

[0004] First, the data heterogeneity gap leads to insufficient integration depth. BIM models are based on standard formats such as IFC and use object-oriented parametric geometric descriptions, including rich geometric dimensions, spatial occupancy, material properties, and topological relationships. Construction progress data is centered on task sequence networks and includes time attributes such as process logic, duration, and time difference. Cost data is based on the bill of quantities and includes cost attributes such as project code, comprehensive unit price, and resource consumption. These three types of data have vastly different structures, semantic granularity, and update frequencies. The existing "plug-in" association is essentially a weak coupling relationship. Once design changes, schedule adjustments, or price fluctuations occur, this ID- or code-based mapping is easily broken, leading to data inconsistency. It requires manual re-maintenance of the association relationship and cannot achieve automatic synchronization and linkage analysis of multi-dimensional data.

[0005] Secondly, the dynamic adaptability and real-time performance are severely lacking. The construction process is fraught with uncertainty, such as weather changes, material supply delays, equipment failures, and design changes. When actual progress deviates from the plan, or material prices fluctuate, the existing model cannot automatically reconstruct the coupling relationship of "time-space-price". The impact of schedule delays on costs cannot be automatically transmitted within the model; it can only be demonstrated through post-event comparative analysis, lacking real-time early warning capabilities. Conflict detection usually occurs after data entry, which is a post-event remedy rather than a pre-event prediction, making it difficult to provide timely and effective decision support for managers.

[0006] Secondly, decision support capabilities are limited. Existing 4D / 5D models are mostly visualizations of predetermined plans, that is, playing out the model's construction process according to a pre-set schedule, or displaying cumulative cost curves according to a pre-set list. When managers need to assess the impact of "rushing" on costs or "material substitution" on schedule, they cannot conduct accurate quantitative simulations within the model and can only rely on experience-based estimations or external calculations, resulting in decisions lacking a scientific basis. In addition, existing technologies are mostly focused on the construction phase and are difficult to extend to the design and operation and maintenance phases, failing to achieve integrated data management throughout the entire building lifecycle.

[0007] This invention aims to solve the technical problems existing in the prior art. To this end, it proposes a heterogeneous fusion modeling method and system for BIM models and construction progress and cost data. Summary of the Invention

[0008] To address the aforementioned technical issues, this invention provides a heterogeneous fusion modeling method and system for BIM models with construction progress and cost data. The aim is to achieve deep intrinsic fusion of geometry, schedule, and cost, enabling the model to possess dynamic self-adaptation and intelligent early warning capabilities, and making it suitable for the full life cycle management of building projects.

[0009] The specific plan is as follows: The first aspect of this invention provides a heterogeneous fusion modeling method for BIM models and construction progress and cost data, comprising the following steps: S1, Multi-source data parsing and feature space mapping: Parse BIM source model files, construction schedule files and cost list files respectively, extract geometric feature vectors, time series feature vectors and cost feature vectors, construct a unified heterogeneous feature space, and use semantic mapping functions to map feature vectors from different sources to isomorphic feature points in the space; S2, Constructing a coupling tensor field: Based on the isomorphic feature points generated in step S1, construct a coupling tensor field that describes the dynamic correlation strength between geometric components, construction procedures, and cost items; the coupling tensor field includes a time coupling factor, a spatial constraint factor, and a cost transmission factor; S3, Fusion Modeling Based on Variational Principle: Under the constraint of the coupling tensor field, the optimal state of the fusion model is solved by minimizing the total energy functional to generate the initial fusion model; wherein, the total energy functional includes geometric deformation energy, schedule deviation energy, cost potential energy and coupling constraint terms; S4, Dynamic Evolution and Conflict Resolution: Real-time access to IoT data and market cost index at the construction site is used as boundary conditions to update the coupling tensor field. When a state shift at a certain feature point is detected, causing the local energy functional gradient to exceed the threshold, a model self-reconstruction mechanism is triggered to automatically adjust the geometric state, planned time, or budget cost of associated components and output a dynamically updated fusion model.

[0010] Preferably, the semantic mapping function in step S1 is used to establish a mapping from feature vectors to feature space coordinates, and the mapping rules are jointly determined by geometric wiki, temporal wiki and cost wiki; wherein, the geometric feature vector is extracted from the BIM source model file; the temporal feature vector is extracted from the construction schedule file; the cost feature vector is extracted from the cost list file; and the feature points obtained after mapping are composed of geometric wiki coordinates, temporal wiki coordinates and cost wiki coordinates.

[0011] Preferably, the coupling tensor field in step S2 is a third-order tensor, whose elements represent the coupling strength between geometric feature points, temporal feature points, and cost feature points; the temporal coupling factor is calculated based on the logical relationship of the work process, the spatial constraint factor is calculated based on the topological adjacency relationship of the components in three-dimensional space, and the cost transmission factor is calculated based on the pricing rules of the bill of quantities.

[0012] Preferably, the total energy functional in step S3 is composed of a weighted sum of four parts: geometric deformation energy is used to measure the deviation between the actual geometry and the designed geometry of the component; schedule deviation energy is used to measure the deviation between the actual schedule and the planned schedule; cost potential energy is used to measure the deviation between the actual cost and the budgeted cost; the coupling constraint term is determined by the coupling degree tensor field and is used to penalize inconsistencies; the weight coefficients of each term are adjusted according to the characteristics of the project, and the sum is 1.

[0013] Preferably, the model self-reconstruction mechanism in step S4 includes: calculating new energy minimum points based on the updated coupling tensor field, solving for new feature point coordinates through an iterative optimization algorithm, and updating the geometric parameters of BIM components, the duration of construction procedures, or the pricing basis of the cost list based on the new coordinates through reverse mapping.

[0014] A second aspect of this invention provides a heterogeneous fusion modeling system for BIM models and construction progress and cost data, based on the heterogeneous fusion modeling method for BIM models and construction progress and cost data described in the first aspect of this invention, comprising: Multi-source data access module: It is equipped with multiple data interfaces for reading and parsing BIM source model files, construction schedule files and cost list files in different formats; Heterogeneous feature space mapping engine: It has a built-in semantic mapping rule library and is responsible for converting multi-source data into a unified feature point cloud; Coupled field calculation and analysis module: used to construct and update the coupling degree tensor field in real time, and calculate the energy gradient of each feature point; Integrated modeling and dynamic simulation module: Based on the energy functional extremum solution, it generates and displays the integrated model, supporting multi-dimensional dynamic simulation of the construction process; Conflict warning and self-healing decision module: When an energy anomaly or a break in coupling is detected, it issues a warning and provides adjustment suggestions or automatically performs model reconstruction.

[0015] Preferably, the data formats supported by the multi-source data access module include IFC, RVT, DWG, MPP, XLSX, XLS, and XML.

[0016] Preferably, the semantic mapping rule library of the heterogeneous feature space mapping engine is used to define the mapping relationship between similar entities in different data sources and supports user-defined mapping rules.

[0017] Preferably, the conflict early warning and self-healing decision module includes an early warning threshold setting unit, used to set the threshold of the energy functional gradient according to the project risk level.

[0018] A third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the heterogeneous fusion modeling method for BIM models and construction progress and cost data as described in any of the first aspects of the present invention.

[0019] The beneficial effects of this invention are as follows: This invention achieves deep intrinsic fusion: Unlike existing technologies that use IDs or codes for foreign key associations, this invention constructs a heterogeneous feature space and a coupling tensor field to uniformly model schedule and cost data as intrinsic attributes of BIM components, fundamentally eliminating the data heterogeneity gap. In the fusion model, geometry, time, and cost are no longer independent datasets, but different aspects of the same entity. Changes in any one aspect are automatically transmitted to the others through the coupling tensor field, ensuring data consistency and synergy.

[0020] Possessing dynamic adaptive capabilities: By introducing an energy functional model based on variational principles, the fusion model can automatically seek new equilibrium states under external disturbances, achieving interconnected self-reconfiguration across time, space, and price. This adaptive capability significantly enhances the model's dynamic adaptability to complex construction sites, enabling it to reflect real-time changes and provide accurate data support for project management. For example, when a process is delayed, the model automatically adjusts the timing of subsequent processes and recalculates costs without manual intervention.

[0021] Intelligent conflict early warning: By monitoring abrupt changes in local energy gradients, potential conflict risks can be detected early in the accumulation of deviations, transforming traditional post-event analysis into pre-event warning. Managers can take measures before problems occur to avoid losses. For example, when the energy gradient exceeds a threshold, the system will issue an early warning that a certain area may experience cost overruns due to schedule delays, reminding managers to adjust resource allocation.

[0022] It enhances the scientific rigor of decision-making: Due to the self-consistent mathematical logic within the fusion model, managers can accurately quantify and extrapolate the impact of "rushing work" on costs or "material substitution" on schedule in a virtual environment. By adjusting the parameters in the model, the impact on the overall project can be immediately observed, providing reliable data support for optimizing construction plans and avoiding the risks associated with decision-making based on experience.

[0023] Supporting full lifecycle management: The method of this invention is applicable not only to the construction phase but also to the design and operation and maintenance phases. In the design phase, the impact of different design schemes on construction progress and costs can be simulated to optimize the design; in the operation and maintenance phase, actual operation and maintenance data can be combined to evaluate facility performance and provide a basis for maintenance decisions. Therefore, this invention provides an integrated data model for the full lifecycle management of buildings. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the overall process of the heterogeneous fusion modeling method in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram illustrating the principle of multi-source data parsing and feature space mapping in Embodiment 1 of the present invention; Figure 3 This is a schematic diagram of the fusion modeling and dynamic evolution mechanism based on the variational principle in Embodiment 1 of the present invention; Figure 4 This is a structural block diagram of the heterogeneous fusion modeling system in Embodiment 1 of the present invention. Detailed Implementation

[0025] The technical solutions of the present invention will now be clearly and completely described with reference to the accompanying drawings in the embodiments of the present invention.

[0026] Example 1: Please see Figure 1 This embodiment provides a heterogeneous fusion modeling method for BIM models and construction progress and cost data, including the following steps: S1, Multi-source data parsing and feature space mapping: Parse BIM source model files, construction schedule files and cost list files respectively, extract geometric feature vectors, temporal feature vectors and cost feature vectors, construct a unified heterogeneous feature space, and use semantic mapping functions to map feature vectors from different sources to isomorphic feature points in the space.

[0027] Specifically, please refer to Figure 2 First, the component information in the BIM source model is read through the API or IFC parsing library provided by the BIM software. Geometric features, such as spatial occupancy, volume, surface area, and topological relationships, are extracted for each component to form a geometric feature vector. Simultaneously, task information for each process is read from the construction schedule file, and its time features are extracted, such as planned start time, planned end time, earliest start time (ES), latest start time (LS), total float (TF), free float, and duration, forming a time-series feature vector. Finally, cost information for each item in the cost estimate file is read, and its cost features, such as comprehensive unit price, labor costs, material costs, machinery costs, management fees, profit, quantity of work, and resource consumption, are extracted to form a cost feature vector.

[0028] Then, a multidimensional heterogeneous feature space is defined, with dimensions including geometric, temporal, and cost dimensions, each with its own specific basis. A semantic mapping function is used to transform feature vectors from different sources into coordinate points within this multidimensional heterogeneous feature space.

[0029] The semantic mapping function is used to establish a mapping from feature vectors to feature space coordinates. The mapping rules are jointly determined by the geometric wiki, temporal wiki, and cost wiki. Geometric feature vectors are extracted from the BIM source model file; temporal feature vectors are extracted from the construction schedule file; and cost feature vectors are extracted from the cost estimate file. The resulting feature points consist of geometric wiki coordinates, temporal wiki coordinates, and cost wiki coordinates.

[0030] The design of semantic mapping functions must ensure that similar components are clustered in space, while different types of components are separated, and the semantic information of the original data is preserved. For example, geometric features can be normalized and used as geometric basis coordinates, time features can be transformed into offsets relative to the project start time and used as time basis coordinates, and cost features can be logarithmically transformed and used as cost basis coordinates. In this way, each component has a unique point in the heterogeneous feature space, which encapsulates its geometric, time, and cost attributes, achieving a homogeneous representation of heterogeneous data.

[0031] Specifically, the semantic mapping function is implemented in the following mathematical form: Let the geometric feature vector corresponding to any building component be . ,in The number of geometric features is represented; the temporal feature vector is... ,in The number of time-series features is represented; the cost feature vector is... ,in This represents the number of cost features. It defines the coordinates of points in the heterogeneous feature space. ,in: ; ; ; in, This is a geometric feature normalization function, which takes the mean value after performing min-max normalization on each component of the geometric feature vector. Indicates the first Geometric eigenvalues, Indicates the first The minimum value of a geometric feature across the entire project scope. Indicates the first The maximum value of a geometric feature across the entire project; This is a time mapping function that maps the planned start time of a specific process. Linear mapping to the total project duration. This is the planned start time for this process (the moment when the project start point is 0). This indicates the earliest start time across the entire project scope. Indicates the latest start time across the entire project scope; For cost mapping function, the comprehensive unit price Perform logarithmic compression normalization. The unit price (yuan / unit of work) is the comprehensive price of the component. This represents the minimum composite unit price for all items in the bill of quantities across the entire project. This represents the maximum composite unit price for all items in the entire project scope. All maximum and minimum values ​​mentioned above are taken from all samples of the corresponding features across the entire project scope to ensure that the mapped coordinates are within a comparable range.

[0032] It should be noted that the above mapping function is only one optional implementation method. Those skilled in the art can also use other forms of linear or nonlinear mapping functions, such as Gaussian kernel mapping, multilayer perceptron mapping, etc., as long as they satisfy the requirement of converting multi-source feature vectors into isomorphic point coordinates in space. S2, Constructing a coupling tensor field: Based on the isomorphic feature points generated in step S1, a coupling tensor field describing the dynamic correlation strength between geometric components, construction procedures, and cost items is constructed; the coupling tensor field includes a time coupling factor, a spatial constraint factor, and a cost transmission factor.

[0033] The coupling tensor field is a third-order tensor, whose elements represent the coupling strength between any geometric feature point, a temporal feature point, and a cost feature point. The calculation of coupling strength comprehensively considers three factors: a temporal coupling factor, calculated based on the logical relationship between processes; for example, if two processes have a successor-successor relationship, the coupling strength between their corresponding temporal feature points and related geometric feature points is high. A spatial constraint factor, calculated based on the topological adjacency relationship of components in three-dimensional space; for example, if two components are directly connected in physical space (such as a beam resting on a column), the coupling strength between their geometric feature points and related processes and cost points is high. A cost transmission factor, calculated based on the pricing rules of the bill of quantities; for example, if two bill of quantities items share the same material, the coupling strength between their corresponding cost feature points and related components and processes is high when the material price fluctuates. By combining these factors, the value of each element in the coupling tensor field can be calculated, forming a coupling network covering the entire project. This network describes how changes in any dimension are transmitted to other dimensions through coupling relationships.

[0034] S3, Fusion Modeling Based on Variational Principle: Under the constraint of the coupling tensor field, the optimal state of the fusion model is solved by minimizing the total energy functional to generate the initial fusion model; wherein, the total energy functional includes geometric deformation energy, schedule deviation energy, cost potential energy and coupling constraint terms.

[0035] Please see Figure 3 The construction process is viewed as a dynamic system, with the system's state determined by the coordinates of all feature points. Any deviation from the planned state introduces an "energy" increment. The total energy functional is defined as a weighted sum of four parts: geometric deformation energy, used to measure the deviation between the actual and designed geometry of a component, such as offsets in component installation positions or dimensional errors; schedule deviation energy, used to measure the deviation between actual and planned progress, such as delays or advances in procedures; cost potential energy, used to measure the deviation between actual and budgeted costs, such as cost overruns due to material price increases; and coupling constraint terms, determined by the coupling degree tensor field, used to penalize incompatible changes, i.e., when a change in one dimension does not cause corresponding adjustments in other related dimensions, the coupling constraint term will increase. The weight coefficients of each item are adjustable according to project characteristics, and the sum is 1. For example, for cost-sensitive projects, the weight of cost potential energy can be increased. By solving for the minimum value of the total energy functional, a new set of feature point coordinates can be obtained. The states corresponding to these coordinates are the optimal equilibrium states after considering various coupling relationships. The system updates the geometric parameters of BIM components, the planned time of construction procedures, or the pricing basis of the cost list based on the new coordinates and reverse mapping, generating an initial fusion model.

[0036] S4, Dynamic Evolution and Conflict Resolution: Real-time access to IoT data and market cost index at the construction site is used as boundary conditions to update the coupling tensor field. When a state shift at a certain feature point is detected, causing the local energy functional gradient to exceed the threshold, a model self-reconstruction mechanism is triggered to automatically adjust the geometric state, planned time, or budget cost of associated components and output a dynamically updated fusion model.

[0037] The model self-reconstruction mechanism includes: calculating new energy minimum points based on the updated coupling tensor field, solving for new feature point coordinates through iterative optimization algorithms, and updating the geometric parameters of BIM components, the duration of construction procedures, or the pricing basis of the cost list based on the new coordinates through reverse mapping.

[0038] Specifically, during construction, real-time on-site data, such as actual construction progress, material consumption, and equipment status, is collected through IoT devices (e.g., sensors, cameras, RFID, etc.). Simultaneously, real-time cost indices, such as steel and cement prices, are obtained from market information platforms. This real-time data is used as new boundary conditions to update relevant factors in the coupling tensor field. Then, the local energy gradient, i.e., the partial derivative of the total energy functional with respect to the coordinates of each feature point, is recalculated. If the energy gradient of a feature point exceeds a preset threshold, it indicates that the state change at that point has caused a significant increase in system energy, potentially leading to a conflict. At this point, a model self-reconstruction mechanism is triggered: the system re-solves for the minimum value of the total energy functional based on the updated tensor field and boundary conditions, obtaining a new set of feature point coordinates, and automatically adjusts the geometric state, planned time, or budgeted cost of associated components according to the new coordinates. This process is automatic and requires no manual intervention. Finally, a dynamically updated fusion model is output, which always remains consistent with the actual construction state.

[0039] Please see Figure 4 This embodiment also discloses a system based on the heterogeneous fusion modeling method of the above-mentioned BIM model and construction progress and cost data, including: Multi-source data access module: Configured with multiple data interfaces for reading and parsing BIM source model files, construction schedule files, and cost estimate files in different formats, including IFC Reader, MS Project Reader, and ExcelReader. This multi-source data access module supports common BIM formats such as IFC, RVT, and DWG; schedule formats such as MPP and XML; and cost estimate formats such as XLSX, XLS, and XML. The multi-source data access module internally includes parsers for the corresponding formats, capable of converting raw files into standardized internal data structures.

[0040] The heterogeneous feature space mapping engine, with its built-in semantic mapping rule library, is responsible for converting multi-source data into a unified feature point cloud. This library defines the mapping relationships between similar entities from different data sources and supports user-defined mapping rules. Specifically, it includes predefined mapping rules, such as how to extract geometric features from IFC entities, how to extract time features from tasks, and how to extract cost features from bill of quantities items. The rule library also supports user-defined rules to adapt to specific project needs. Based on these rules, the heterogeneous feature space mapping engine maps each component, process, and bill of quantities item to a point in the feature space, forming a point cloud dataset.

[0041] The coupling field calculation and analysis module is used to construct and update the coupling degree tensor field in real time, and calculate the energy gradient of each feature point. This module first calculates the initial tensor field based on the feature point cloud and project logical relationships (process logic, spatial topology, and bill of quantities associations). Then, during construction, it updates the tensor field based on real-time data and calculates the energy gradient of each feature point, providing a basis for conflict early warning.

[0042] The integrated modeling and dynamic simulation module generates and displays a fused model based on energy functional extremum solving, supporting multi-dimensional dynamic simulation of the construction process. This module implements numerical solution algorithms for energy functionals, such as gradient descent and conjugate gradient methods, enabling rapid identification of energy minima. It also provides a visual user interface to display the fused results in 3D model form and supports playback of the construction process along a timeline, dynamically displaying cost changes.

[0043] The conflict early warning and self-healing decision-making module issues an early warning and provides adjustment suggestions or automatically executes model reconstruction when an energy anomaly or broken coupling relationship is detected. This module includes an early warning threshold setting unit, allowing users to set different thresholds based on the project's risk level. When the local energy gradient exceeds the threshold, the module issues an early warning signal and highlights the problem area on the user interface. Simultaneously, the module can automatically generate adjustment plans, such as adjusting subsequent process times, replacing materials, or increasing resources, and evaluate the energy changes of each plan, recommending the optimal plan for the user to choose from, or directly executing self-reconstruction.

[0044] Example 2: This embodiment provides a specific case, applying the heterogeneous fusion modeling method of BIM model with construction progress and cost data from Embodiment 1 to the construction management process of a large commercial complex project, including the following steps: S1: Multi-source data analysis and feature space mapping.

[0045] First, the system acquires three types of source data through the multi-source data access module: BIM Source Model: A structural model of a commercial complex built using Revit software, exported as an IFC 2×3 format file. This model includes two underground floors and five above-ground floors, with a total building area of ​​approximately 100,000 square meters, and contains thousands of components such as columns, beams, slabs, and walls.

[0046] Construction Schedule: The overall schedule, created using Microsoft Project, includes the start and end dates and logical relationships of each sub-project. The planned duration is 24 months, encompassing approximately 500 work processes, such as earthwork excavation, foundation construction, main structure, and interior decoration.

[0047] Cost Bill of Quantities Data: The bill of quantities, compiled using Glodon software, includes project codes, names, unit prices, and quantities. Compiled according to national standard GB50500-2013, the bill of quantities contains approximately 800 items, covering civil engineering, decoration, installation, and other related specialties.

[0048] Next, the heterogeneous feature space mapping engine analyzes the above data. It extracts the geometric feature vectors of each column and beam in the BIM source model, including spatial coordinates (x, y, z), volume, and surface area. For example, for a first-floor frame column KZ-1, its geometric feature vector includes: bottom center coordinates (10.5, 20.3, 0.0), top center coordinates (10.5, 20.3, 4.5), volume 2.25 m³, surface area 18 m², and a list of IDs of the beams connected to it. It also extracts the temporal feature vectors of the corresponding procedures in the construction schedule, including the earliest start time, latest start time, and total float. For example, the construction procedure for KZ-1 is "first-floor frame column concrete pouring," with a planned start date of day 45, a planned end date of day 47, and a total float of 3 days. Finally, it extracts the cost feature vectors of the corresponding items in the cost estimate, including the comprehensive unit price, labor cost percentage, and material cost percentage. For example, the comprehensive unit price of the item "cast-in-place concrete rectangular column" is 850 yuan / m³, of which labor costs account for 20%, material costs account for 70%, machinery costs account for 5%, and management fees account for 5%.

[0049] Define a three-dimensional heterogeneous feature space, with basis vectors representing the geometric, temporal, and cost dimensional bases. A semantic mapping function maps these three feature vectors to a point within this space. For KZ-1, the mapping process is as follows: normalize the geometric volume and use it as the geometric dimensional base coordinates, for example, the relative position of volume 2.25 m³ among all component volumes in the project; convert the planned start date to a time offset relative to the project's starting point and use it as the temporal dimensional base coordinates, for example, the ratio of day 45 to the total project duration of 720 days; and logarithmically transform the comprehensive unit price and use it as the cost dimensional base coordinates, for example, taking the logarithm of 850. In this way, KZ-1 possesses a unique coordinate point in the heterogeneous feature space, which encapsulates its geometric, temporal, and cost attributes.

[0050] S2, construct the coupling tensor field.

[0051] After obtaining the feature points of all components, a third-order coupling tensor field is constructed. The elements in the coupling tensor field are not only related to the points themselves, but also depend on the logical relationships between them.

[0052] Time coupling factor: If the construction process of KZ-1 is the immediate preceding process of "beam KL-1 formwork", then the coupling degree between the feature points of KZ-1 and the feature points of KL-1 in the time dimension is very high, which is set to 0.8.

[0053] Spatial constraint factor: If KZ-1 and KL-1 are directly adjacent in physical space, their spatial constraint factor is assigned a value of 1; otherwise, it is 0. Here, KZ-1 and KL-1 are connected, so the spatial constraint factor is 1.

[0054] Cost transmission factor: If the concrete strength grade of KZ-1 is C30 and the concrete strength grade of KL-1 is also C30, they share the same material. When the price of C30 concrete fluctuates, the cost of both will be affected. Set the cost transmission factor to 0.6.

[0055] By combining these factors, the coupling strength is calculated using methods such as weighted summation or product. For example, the product of the three factors is 0.8 * 1 * 0.6 = 0.48. The coupling strength is calculated for all possible point pairs, forming a coupling tensor field covering the entire project. This coupling tensor field describes how changes in any dimension will affect other dimensions through the transmission of the "field".

[0056] S3, fusion modeling based on variational principles.

[0057] The core of this step is to solve for the optimal state of the fusion model, that is, to obtain the minimum value of the total energy functional. The construction process is regarded as a dynamic system, and any change that deviates from the plan will introduce an increase in "energy".

[0058] The total energy functional is defined as a weighted sum of geometric deformation energy, schedule deviation energy, cost potential energy, and coupling constraint terms. The weighting coefficients are adjusted according to the project characteristics. In this embodiment, they are set as follows: geometric deformation energy α=0.2, schedule deviation energy β=0.3, cost potential energy γ=0.3, and coupling constraint term λ=0.2.

[0059] Geometric deformation energy measures the deviation between the actual installation position and the design position of a component. For example, if the actual installation position of KZ-1 deviates from the design position by 5 cm, the geometric deformation energy will increase. Schedule variability energy measures the deviation between the current schedule and the planned schedule. For example, if KZ-1 starts 2 days late, its time wiki coordinates change, and the schedule variability energy increases. Cost potential energy measures the deviation between the current actual cost and the budgeted cost. For example, a rise in steel reinforcement prices causes a change in the cost wiki coordinates, and the cost potential energy increases. Coupling constraint terms are determined by the coupling tensor field and the coordinates of each point. If the schedule of KZ-1 is delayed, but according to the tensor field, it has a strong coupling relationship with the cost of KL-1, then the coupling constraint terms will increase unless the cost wiki coordinates of KL-1 are adjusted to seek balance.

[0060] By solving a variational problem—that is, finding a set of feature point coordinates that minimizes the total energy functional—a new set of feature point coordinates is obtained. The states corresponding to these coordinates represent the optimal equilibrium state after considering various coupling relationships. The system then generates a fusion model based on the new coordinates: in the 3D model, KZ-1 may be marked as a warning color, with yellow indicating a slight deviation; in the schedule, its subsequent processes may be automatically postponed by 2 days; in the cost report, related cost items may be frozen and reserved to cope with potential cost increases.

[0061] S4, Dynamic Evolution and Conflict Resolution.

[0062] During construction, on-site monitoring data is input into the system as new boundary conditions. The system updates the coupling tensor field, for example, updating the time coordinates of KZ-1, and recalculates the local energy gradient. It is found that due to the delay of KZ-1, the energy functional gradients of strongly coupled components such as KL-1 and the second-floor slab exceed the preset threshold. At this point, the self-reconfiguration mechanism is triggered.

[0063] Based on the new boundary conditions and the principle of energy minimization, the system automatically generates an adjustment plan. For example, it suggests increasing the resources of the KL-1 work team by 20% to compress the subsequent construction period, or adjusting the construction sequence of the second-floor slabs, constructing other areas first. These adjustments are not only reflected in the schedule but also highlighted in the BIM source model, automatically calculating the additional costs incurred due to expedited work. Ultimately, it outputs a dynamically updated, self-consistent 5D fusion model that reflects the latest construction status and the adjusted plan.

[0064] Example 3: The difference between this embodiment and Embodiment 2 is that this embodiment focuses on the integrated design-procurement-construction model under the EPC (Engineering, Procurement, and Construction) model, while the rest are basically the same.

[0065] In step S1, additional procurement logistics data is introduced as part of the cost feature vector, such as equipment delivery time and transportation costs. Specifically, for a large piece of equipment, such as a chiller unit, the procurement data includes: supplier, contract price, estimated delivery time, transportation method, transportation cycle, customs clearance time, etc. This data is integrated into the cost feature vector; for example, the delivery time is incorporated into the time wiki, and transportation costs are included in the comprehensive unit price.

[0066] In step S2, when constructing the coupling tensor field, a "procurement delay factor" is introduced. This procurement delay factor is calculated based on procurement logistics data. For example, if the expected arrival time of a piece of equipment is later than the installation start time, the procurement delay factor is positive, indicating a risk of delay. When the arrival time of a large piece of equipment procured overseas is delayed due to logistical reasons, this delay factor will be rapidly transmitted through the tensor field to the progress characteristic points of the civil engineering components related to its installation, as well as the cost characteristic points caused by equipment price increases. For example, the equipment foundation construction process can only be carried out after the equipment arrives. Therefore, equipment delays will cause the time wiki coordinates of the foundation construction progress characteristic point to shift backward, and may also trigger the risk of equipment price increases, while the cost characteristic point may also shift upward.

[0067] In the total energy functional of step S3, a stochastic term related to procurement uncertainty is added, such as considering the probability distribution of equipment delays, so that the total energy functional can reflect the impact of uncertainty on the project. The fusion model can dynamically simulate the worst-case impact of equipment delays on the overall construction schedule and cost, and automatically generate adjustment suggestions, such as adjusting the priority of foundation construction, constructing other areas first, or initiating the evaluation process of alternative suppliers.

[0068] Example 4: The difference between this embodiment and embodiment two is that this embodiment specifies the logic of conflict resolution, while the rest are basically the same.

[0069] When the model self-reconstruction mechanism in step S4 is triggered, the system does not directly modify the fused model, but instead enters a "conflict resolution simulation" stage. Assume that a design change in a certain area leads to a change in the geometric dimensions of components. For example, the cross-sectional dimensions of a beam change from 300×600 mm to 300×800 mm, resulting in an increase in its volume and potentially affecting the stress on connected columns, requiring additional reinforcement. The system first freezes the current model in the virtual environment and generates multiple branch schemes: Option A: Maintain the original schedule, but accept the increased costs due to design changes. For example, the increased beam volume leads to increased concrete and steel reinforcement usage, increasing costs by 100,000 yuan. Additionally, the larger beam dimensions may affect the ceiling height, requiring adjustments to subsequent decoration plans, but the schedule remains unchanged.

[0070] Option B: Compress the time for subsequent processes to maintain the total cost. For example, by increasing the number of workers and working overtime, the time for subsequent processes can be shortened to offset the increased costs, but this may result in expedited work costs of 50,000 yuan, and the schedule deviation may increase.

[0071] Option C: Modify the geometric design of subsequent related components to accommodate the changes, but this may introduce new design costs. For example, adjusting the ceiling elevation or modifying the reinforcement of connected columns would require redesign, incurring a design fee of 20,000 yuan, and may also affect the schedule.

[0072] The system calculates the total energy functional value for each of the three schemes and recommends the scheme with the lowest energy for project managers to make a decision. For example, if scheme A has an energy of 100, scheme B has an energy of 95, and scheme C has an energy of 98, then the system recommends scheme B. Once selected, the system then performs the actual model update, thus ensuring the controllability and optimality of the model evolution.

[0073] It should be noted that, in this document, the terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0074] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A heterogeneous fusion modeling method for BIM models and construction progress and cost data, characterized in that, Includes the following steps: S1, Multi-source data parsing and feature space mapping: Parse BIM source model files, construction schedule files and cost list files respectively, extract geometric feature vectors, time series feature vectors and cost feature vectors, construct a unified heterogeneous feature space, and use semantic mapping functions to map feature vectors from different sources to isomorphic feature points in the space; S2, Constructing a coupling tensor field: Based on the isomorphic feature points generated in step S1, construct a coupling tensor field that describes the dynamic correlation strength between geometric components, construction procedures, and cost items; the coupling tensor field includes a time coupling factor, a spatial constraint factor, and a cost transmission factor; S3, Fusion Modeling Based on Variational Principle: Under the constraint of the coupling tensor field, the optimal state of the fusion model is solved by minimizing the total energy functional to generate the initial fusion model; wherein, the total energy functional includes geometric deformation energy, schedule deviation energy, cost potential energy and coupling constraint terms; S4, Dynamic Evolution and Conflict Resolution: Real-time access to IoT data and market cost index at the construction site is used as boundary conditions to update the coupling tensor field. When a state shift at a certain feature point is detected, causing the local energy functional gradient to exceed the threshold, a model self-reconstruction mechanism is triggered to automatically adjust the geometric state, planned time, or budget cost of associated components and output a dynamically updated fusion model.

2. The heterogeneous fusion modeling method for BIM model and construction progress and cost data according to claim 1, characterized in that, The semantic mapping function in step S1 is used to establish a mapping from feature vectors to feature space coordinates. The mapping rules are jointly determined by geometric wiki, temporal wiki, and cost wiki. Among them, geometric feature vectors are extracted from BIM source model files; temporal feature vectors are extracted from construction schedule files; and cost feature vectors are extracted from cost list files. The feature points obtained after mapping are composed of geometric wiki coordinates, temporal wiki coordinates, and cost wiki coordinates.

3. The heterogeneous fusion modeling method for BIM model and construction progress and cost data according to claim 1, characterized in that, The coupling tensor field in step S2 is a third-order tensor, whose elements represent the coupling strength between geometric feature points, temporal feature points, and cost feature points; the temporal coupling factor is calculated based on the process logic relationship, the spatial constraint factor is calculated based on the topological adjacency relationship of components in three-dimensional space, and the cost transmission factor is calculated based on the pricing rules of the bill of quantities.

4. The heterogeneous fusion modeling method for BIM model and construction progress and cost data according to claim 1, characterized in that, The total energy functional in step S3 consists of a weighted sum of four parts: geometric deformation energy, which measures the deviation between the actual geometry and the design geometry of the component; schedule deviation energy, which measures the deviation between the actual schedule and the planned schedule; cost potential energy, which measures the deviation between the actual cost and the budgeted cost; and coupling constraint term, which is determined by the coupling degree tensor field and is used to penalize inconsistencies. The weight coefficients of each term are adjusted according to the characteristics of the project, and the sum is 1.

5. The heterogeneous fusion modeling method for BIM model and construction progress and cost data according to claim 1, characterized in that, The model self-reconstruction mechanism in step S4 includes: calculating new energy minimum points based on the updated coupling tensor field, solving for new feature point coordinates through an iterative optimization algorithm, and updating the geometric parameters of BIM components, the duration of construction procedures, or the pricing basis of the cost list based on the new coordinates through reverse mapping.

6. A heterogeneous fusion modeling system for BIM models and construction progress and cost data, characterized in that, The heterogeneous fusion modeling method based on any one of claims 1 to 5, which integrates BIM models with construction progress and cost data, includes: Multi-source data access module: It is equipped with multiple data interfaces for reading and parsing BIM source model files, construction schedule files and cost list files in different formats; Heterogeneous feature space mapping engine: It has a built-in semantic mapping rule library and is responsible for converting multi-source data into a unified feature point cloud; Coupled field calculation and analysis module: used to construct and update the coupling degree tensor field in real time, and calculate the energy gradient of each feature point; Integrated modeling and dynamic simulation module: Based on the energy functional extremum solution, it generates and displays the integrated model, supporting multi-dimensional dynamic simulation of the construction process; Conflict warning and self-healing decision module: When an energy anomaly or a break in coupling is detected, it issues a warning and provides adjustment suggestions or automatically performs model reconstruction.

7. The heterogeneous fusion modeling system for BIM models and construction progress and cost data according to claim 6, characterized in that, The multi-source data access module supports data formats including IFC, RVT, DWG, MPP, XLSX, XLS, and XML.

8. The heterogeneous fusion modeling system for BIM models and construction progress and cost data according to claim 6, characterized in that, The semantic mapping rule library of the heterogeneous feature space mapping engine is used to define the mapping relationship between similar entities in different data sources and supports user-defined mapping rules.

9. The heterogeneous fusion modeling system for BIM models and construction progress and cost data according to claim 6, characterized in that, The conflict early warning and self-healing decision-making module includes an early warning threshold setting unit, which is used to set the threshold of the energy functional gradient according to the project risk level.

10. A computer-readable storage medium having a computer program stored thereon, the program being executed by a processor to implement a heterogeneous fusion modeling method for BIM models and construction progress and cost data as described in any one of claims 1 to 5.

Images