Building construction information management method and system based on BIM-GIS fusion

Abstract

This invention provides a method and system for building construction information management based on BIM-GIS integration. It unifies the access of multi-source heterogeneous data to a middle layer, incorporates a built-in BIM-GIS bidirectional coordinate conversion and adaptation module, and configures standardized access protocols and protocol parsing components for different data sources such as IoT time-series data, UAV 3D point cloud data, and construction text form data. Using the geospatial data of the construction site as a benchmark, it integrates various data and binds them through spatiotemporal association primary keys to construct an integrated spatiotemporal data foundation. Edge nodes collect real-time data from nearby locations, which is then streamed via a lightweight message queue, with high-risk monitoring data being prioritized. A bidirectional association is established between BIM-GIS and real-time data, automatically locating the corresponding spatial points when data changes.

Description

Technical Field

[0001] This invention relates to the field of information management technology for building construction, and more specifically, to a method and system for information management of building construction based on BIM-GIS integration. Background Technology

[0002] The construction industry is rapidly developing towards informatization and intelligentization. Construction sites generate heterogeneous data from multiple sources, including BIM models, GIS geographic information, IoT time-series monitoring, UAV 3D point clouds, and construction documents. However, the existing management model lacks a standardized and unified access mechanism. Various data spatial benchmarks are inconsistent, coordinate systems are biased, and effective spatiotemporal association and integration rules for the entire domain and lifecycle have not been established, which easily leads to data silos. At the same time, there is no targeted edge deployment solution for real-time data acquisition, and the transmission process does not prioritize data. High-risk monitoring data is prone to problems such as high latency and loss. The edge end also lacks lightweight preprocessing capabilities, making it difficult to meet the core needs of efficient real-time data management at construction sites.

[0003] Existing BIM-GIS integrated management platforms have significant shortcomings in visualization and data linkage capabilities. They cannot adaptively adjust the model display accuracy according to the performance of terminal devices, construction scenarios, and user interaction needs, which can easily lead to wasted system resources or rendering lag. Furthermore, real-time data and BIM-GIS spatial elements lack a precise two-way correlation mechanism, and data changes cannot automatically locate corresponding spatial points. In addition, there is a lack of cross-validation of multi-source data and automatic anomaly re-collection methods. Risk warnings are mostly in a passive response mode, lacking an integrated mechanism of hierarchical push and model linkage highlighting, and have not established dynamic adjustment rules to adapt to changes in the construction stage. As a result, it is difficult to achieve accurate monitoring, intelligent early warning, and efficient visual management of the building construction process. Summary of the Invention

[0004] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a method for information management of building construction based on BIM-GIS integration, the method comprising: Multi-source heterogeneous data is uniformly accessed to the middle layer, with a built-in BIM and GIS bidirectional coordinate transformation and adaptation module. Standardized access protocols and protocol parsing components are configured for different types of data sources, such as IoT time-series data, UAV 3D point cloud data, and construction text form data. Based on the geospatial data of the construction site, various types of data are integrated and bound together through a spatiotemporal correlation primary key to construct an integrated spatiotemporal data foundation. Real-time data is collected from nearby edge nodes and transmitted via a lightweight message queue, with high-risk monitoring data being pushed first. Establish a two-way link between BIM-GIS and real-time data, and automatically locate the corresponding spatial points when the data changes; Data and models are updated incrementally by component / location / region, and storage and display are refreshed synchronously. Cross-validation of multi-source data identifies anomalies through spatial matching and logical verification, automatically removes them, and triggers re-collection. The system adaptively adjusts the model display accuracy, overlays real-time data visualization labels, and identifies risks, highlights the model, and pushes warnings when thresholds are exceeded.

[0005] Furthermore, embodiments of the present invention also provide a building construction information management system based on BIM-GIS integration, characterized in that it includes: A processor; a machine-readable storage medium for storing machine-executable instructions of the processor; wherein the processor is configured to execute the above-described BIM-GIS integrated building construction information management method by executing the machine-executable instructions.

[0006] In another aspect, embodiments of the present invention also provide a computer program product, the computer program product including machine-executable instructions, the machine-executable instructions being stored in a computer-readable storage medium, the processor of a computer device reading the machine-executable instructions from the computer-readable storage medium, the processor executing the machine-executable instructions, causing the computer device to execute the above-described building construction information management method based on BIM-GIS integration.

[0007] Based on the above, by constructing a unified access system for multi-source heterogeneous data, standardized access and spatial benchmark normalization of various types of data such as BIM, GIS, IoT, and UAV point clouds have been achieved. Relying on a spatiotemporal dual-layer hierarchical architecture and a unique spatiotemporal association primary key, an integrated spatiotemporal data foundation has been created, completely breaking down data silos. The hierarchical transmission strategy of edge node grid-based proximity collection and lightweight message queues, combined with edge preprocessing, multi-source data cross-validation, and automatic data replenishment mechanisms, effectively eliminates abnormal data and achieves data closed-loop correction. At the same time, through dynamic adaptation mechanisms, spatial consistency verification, and distributed caching optimization, the accuracy, effectiveness, and high-speed query of data association are ensured, significantly improving the standardization, integration, and real-time level of construction data management.

[0008] This invention, through an adaptive rendering engine integrated into a BIM-GIS fusion platform, achieves dynamic adjustment of model display accuracy based on terminal performance, construction scenarios, and interactive needs, balancing visualization effects with system operating efficiency. Relying on the bidirectional correlation mechanism between BIM-GIS and real-time data, it enables automatic mutual determination of data changes and spatial locations. Combined with hierarchical incremental update technology, it significantly reduces the resource consumption of data and model updates. Simultaneously, through standardized real-time data visualization labels, automatic risk level determination, and hierarchical early warning push, it achieves precise location of construction risks, model highlighting, and multi-channel closed-loop early warning, upgrading construction information management from passive data recording to proactive intelligent monitoring and early warning. This significantly improves the refinement and intelligence of construction management, providing precise spatiotemporal data support for on-site safety control and scientific decision-making, effectively reducing construction safety risks and improving overall construction management efficiency. Attached Figure Description

[0009] Figure 1 This is a schematic diagram of the execution flow of the building construction information management method based on BIM-GIS integration provided in the embodiments of the present invention.

[0010] Figure 2 This is a schematic diagram of exemplary hardware and software components of the building construction information management system based on BIM-GIS integration provided in an embodiment of the present invention. Detailed Implementation

[0011] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating a construction information management method based on BIM-GIS integration, provided in one embodiment of the present invention. The following is a detailed description of this construction information management method based on BIM-GIS integration.

[0012] Step S110: Unified access of multi-source heterogeneous data to the middle layer, with a built-in BIM and GIS bidirectional coordinate transformation and adaptation module, and standardized access protocols and protocol parsing components configured for different types of data sources such as IoT time series data, UAV 3D point cloud data, and construction text form data. Taking high-rise residential building construction projects as an example, heterogeneous data from multiple sources, such as BIM model data, GIS geographic data, IoT monitoring data, UAV point cloud data, and construction form data, are uniformly accessed through a data middleware layer. The middleware layer has a built-in BIM and GIS bidirectional coordinate conversion and adaptation module, an MQTT protocol parsing component for IoT settlement and tilt sensor time-series data, a LAS format parsing component for UAV 3D point cloud data, and a JSON protocol parsing component for construction text form data. Each component performs data format stripping, field extraction, and standardized conversion, achieving unified access and preprocessing of different types of data sources, providing a standardized foundation for subsequent data integration.

[0013] Step S120: Based on the geospatial data of the construction site, integrate various types of data and bind them through spatiotemporal association primary keys to build an integrated spatiotemporal data foundation; Based on the site geospatial of the aforementioned residential construction project, the BIM component data, GIS terrain data, IoT real-time data, UAV real-scene modeling data, and construction business form data, which have been preprocessed by the intermediate layer, are bound one by one through a unified spatiotemporal association primary key. This integrates and collects the scattered multi-source data, constructs an integrated spatiotemporal data foundation that combines storage, association, and scheduling, breaks down the data silos between various data sources, and realizes integrated management of multi-source data.

[0014] Step S121: Determine a unified spatial reference system based on the geospatial data of the entire construction site, select the national geodetic coordinate system as the global spatial reference, obtain the conversion parameters between the BIM local engineering coordinate system and the global geodetic coordinate system through actual measurement, complete the unique calibration of the spatial reference of the entire site, and eliminate the spatial reference deviation of multi-source data. This residential construction project selected the CGCS2000 National Geodetic Coordinate System as the global spatial reference. The transformation parameters between the local BIM engineering coordinate system and the global geodetic coordinate system were obtained by using a total station for on-site measurement. The transformation parameters were then entered into the coordinate transformation adaptation module to complete the unique calibration of the spatial reference for the entire site. This eliminated the spatial deviation caused by the different spatial references of BIM, GIS, IoT and other data, and ensured the uniformity of the spatial references of all data.

[0015] Step S122: Perform spatial attribute standardization processing on BIM component information, GIS topographic data, IoT monitoring data, UAV real-scene modeling data and construction business records, match the corresponding spatial points of the construction area for construction records and business forms without spatial coordinates, bind the spatial coordinates of the installation location of IoT sensors, and accurately register the UAV point cloud and BIM model to the global spatial benchmark to achieve spatial normalization of all data. Spatial attribute standardization processing was carried out on BIM beam, column, and slab component information, GIS site topography data, IoT sensor monitoring data, UAV real-scene model data, construction progress, and quality records within the project. This included matching corresponding spatial points in the construction area to construction process records lacking spatial coordinates, binding spatial coordinates to the installation locations of foundation pit and tower crane sensors, and accurately registering UAV point cloud data with the BIM model to the global spatial benchmark, thus completing the spatial normalization processing of all data.

[0016] Step S123: Construct a spatiotemporal dual-layer hierarchical architecture. In the spatial dimension, the architecture is divided into five levels: construction site, construction zone, individual building, component unit, and monitoring point. In the temporal dimension, the architecture is divided into four levels: total construction period, sub-project, sub-process, and real-time data collection time, forming a spatiotemporal hierarchical framework covering the entire domain and the entire lifecycle. A spatiotemporal hierarchical architecture was constructed for this residential project. Spatially, it was divided into five levels: construction site, construction zone, individual residential building, component unit, and monitoring point. Temporally, it was divided into four levels: total construction period, sub-project, sub-process, and real-time data collection time. This formed a spatiotemporal hierarchical framework covering the entire project area and the entire construction cycle, and clearly defined the spatiotemporal affiliation level of each data point.

[0017] Step S1231: Based on the overall construction plan and GIS geographic data, the entire construction site is delineated using vectorization. The construction zone is divided in conjunction with the construction organization design. The boundaries are drawn using GIS to ensure that the zones do not overlap and are fully covered. Based on the overall construction plan and GIS geographic data of the residential project, the entire construction site was delineated using polygon vectorization. Combined with the construction organization design documents, the site was divided into the foundation pit construction area, the main structure area, and the decoration and finishing area according to the construction stage. The boundaries of each zone were drawn using GIS spatial analysis tools to ensure that the zones do not overlap, have no omissions, and cover the entire site space.

[0018] Step S1232: Using the BIM model as the core, extract information of individual buildings and associate them with construction zones, disassemble the model to divide it into component units, assign unique numbers and collect spatial coordinates to complete the spatial binding of components with individual buildings. Using the project's BIM model as the core carrier, the outline boundary and unique identifier ID of each individual residential building are extracted, and each individual building is accurately associated with the corresponding construction zone, establishing the subordinate relationship between the zone and the individual building; the BIM model is disassembled, and component units are divided according to the type of beam, column and slab components and construction sections, and a unique number containing the zone and individual building code is assigned to each component, and the geometric center coordinates of the component are collected to complete the spatial binding between the component and the individual building.

[0019] Step S1233: Combine construction risks and quality control points to set up monitoring points, obtain the actual ground coordinates and associate them with the corresponding components / areas, assign unique identifiers, and complete the spatial five-level structural division and positioning; By combining construction risk points and quality control points such as project foundation pits and tower cranes, monitoring points are set up in the BIM-GIS fusion model. The geodetic coordinates of each monitoring point are obtained through on-site measurements, the corresponding component units or construction areas are identified, and a unique identifier containing component codes is assigned to each monitoring point, thus completing the division and precise spatial positioning of the five-level spatial structure.

[0020] Step S1234: Determine the total construction period based on the contract and schedule plan, break it down into sub-projects and sub-processes according to specifications, set time connection relationships, unify the collection frequency and ISO8601 timestamp, and complete the four-level time node decomposition and calibration. Based on the project construction contract and schedule, the start and end points of the total construction period are determined. In accordance with the construction quality acceptance specifications for building engineering, the total construction period is broken down into sub-projects such as foundation and main structure, and the time of each sub-project is specified. According to the construction process, it is further refined into sub-processes such as rebar binding and concrete pouring, and the time connection relationship of the process is set. The data collection frequency and ISO8601 timestamp are unified, and the completion time is marked at four levels.

[0021] Step S1235: Formulate spatiotemporal hierarchical coding rules. Spatial coding adopts site, zone, unit, component, and location hierarchical coding. Temporal coding adopts a combination of construction period, division, process, and timestamp. Establish a two-way spatiotemporal mapping relationship through coding. Spatiotemporal hierarchical coding rules were developed for the project. Spatial coding adopts a hierarchical structure of site coding, zone coding, individual unit coding, component coding, and monitoring point coding. Time coding adopts a combination of total project duration coding, sub-project procedure number, sub-item process number, and timestamp. A two-way mapping link between spatial units and time nodes is established through coding association.

[0022] Step S1236: Develop a spatiotemporal hierarchy management module based on the BIMGIS platform, build a visual configuration interface, support hierarchy display and adjustment, and have built-in spatial topology and time logic verification functions; Based on the project's BIM-GIS fusion platform, a spatiotemporal hierarchical management module was developed, and a visual configuration interface was built to realize the spatial five-level structure tree diagram display and the time four-level node timeline display. It supports manual adjustment of hierarchical boundaries and time nodes. The module has built-in spatial topology verification and time logic verification functions to automatically check the rationality of spatial subordination and time connection.

[0023] Step S1237: Establish a spatiotemporal hierarchical dynamic update mechanism. When the construction plan changes, the module automatically updates the hierarchical boundaries, codes and associations, and synchronously links the data layer to achieve rapid association between the hierarchy and multi-source data.

[0024] A spatiotemporal hierarchical dynamic update mechanism is established for the project. When the construction plan is adjusted, resulting in changes to the construction zone, the scope of individual buildings, or the schedule, the spatiotemporal hierarchical management module receives the change instructions, automatically updates the hierarchical boundaries, codes, and relationships, and synchronously links the data access layer and the spatiotemporal data base to achieve rapid association and adaptation between the hierarchy and multi-source data.

[0025] Step S124: Generate a spatiotemporal association primary key based on the combination of spatial grid coding, timestamp and unique identifier of data object, and fuse spatial location coding, data acquisition or effective timestamp, component number, sensor ID and business record coding to make each data unit correspond to a unique spatiotemporal association primary key; Based on the spatial grid code, data acquisition timestamp, component number, sensor ID, and business record code of this project, a spatiotemporal association primary key is generated by combining the spatial location code, timestamp, component number, sensor ID, and business record code to ensure that each data unit corresponds to a unique spatiotemporal association primary key, without duplication or conflict.

[0026] Step S125: Accurately mount the normalized data to the corresponding spatiotemporal nodes according to the preset spatiotemporal hierarchy, and establish a strong binding relationship between BIM component parameters, GIS geographic information, IoT real-time data, real scene model and construction business data through the spatiotemporal association primary key, so as to realize the orderly collection of multi-source data in the same spatiotemporal system; According to the project's pre-set spatiotemporal hierarchy, BIM data, GIS data, IoT data, real-scene model data, and construction business data that have completed spatial normalization are accurately mounted to the corresponding spatiotemporal nodes. By establishing strong binding relationships between various types of data through spatiotemporal association primary keys, the orderly collection and association of multi-source data under the same spatiotemporal system can be realized.

[0027] Step S126: Construct a spatiotemporal hybrid index based on the spatiotemporal associated primary key, supporting multi-dimensional associated retrieval by spatial location, time node, and data type, ensuring rapid data location and associated retrieval; Based on the project's spatiotemporal associated primary key, a spatiotemporal hybrid index system is constructed to support multi-dimensional associated retrieval by construction area spatial location, process time node, and data type, ensuring rapid data location during retrieval, query, and linkage, and improving the efficiency and response speed of associated retrieval of multi-source data.

[0028] Step S127: Perform spatiotemporal consistency verification on the mounted data, check the accuracy of spatial binding and the rationality of time correspondence, automatically identify and correct spatiotemporal misalignment and primary key conflict anomalies, and ensure the reliability of data association; The system performs spatiotemporal consistency verification on the multi-source data after the project is mounted, checks the accuracy of data space binding and the rationality of time correspondence, automatically identifies abnormal issues such as spatiotemporal misalignment and primary key conflicts, and completes corrections through coordinate calibration, primary key reconstruction and other methods to ensure the reliability and accuracy of multi-source data association.

[0029] Step S128: Integrate and encapsulate the multi-source data that has been bound and verified to form an integrated spatiotemporal data foundation with unified storage, unified association, and unified scheduling, providing standardized data support for subsequent data linkage, updates, visualization, and intelligent early warning.

[0030] The project integrates and encapsulates the multi-source data that has been bound, verified, and corrected, forming an integrated spatiotemporal data foundation with unified storage, association, and scheduling. This provides standardized and regulated data support for subsequent data linkage updates, visualization, and intelligent early warning analysis, ensuring the stable operation of subsequent functional modules.

[0031] Step S130: Edge nodes collect real-time data nearby, and transmit it in a streaming manner through a lightweight message queue, with high-risk monitoring data being pushed first; Taking high-rise residential building construction projects as an example, edge computing nodes are deployed in key monitoring areas such as deep foundation pits, tower cranes, and high formwork. These edge nodes collect real-time monitoring data from nearby sensors and construction data reported by mobile terminals. A streaming transmission channel is built using lightweight message queues such as MQTT and ZeroMQ. High-risk monitoring data such as tower crane tilt angle and foundation pit settlement are given the highest transmission priority and are pushed to the cloud platform first using a preemptive transmission mechanism. Ordinary environmental monitoring and material arrival data are transmitted at the normal rate. This achieves nearby collection, efficient transmission, and priority-based push of real-time data, ensuring low latency and high reliability of high-risk data transmission.

[0032] Step S131: Deploy industrial-grade edge gateways adapted to temperatures ranging from -20℃ to 60℃ in a grid pattern, covering IoT devices and mobile terminals within 50 meters, and accessing the construction IoT private network via 5G and wired dual links; The residential project deploys industrial-grade edge gateways in a grid-like manner according to construction zones. The gateways are adapted to the harsh environment of the construction site, ranging from -20℃ to 60℃. The coverage radius of a single gateway is set at 50 meters, accurately covering the IoT sensor cluster in the foundation pit and the mobile terminal acquisition terminal of the main structure. The gateway adopts a dual-link networking mode of 5G industrial modules and wired Ethernet, and connects to the project's dedicated construction IoT network. The gateways are deployed close to the monitoring equipment to shorten the physical transmission distance, reduce data acquisition latency, and the dual links serve as backups for each other to avoid data transmission interruption due to single-link failure.

[0033] Step S132: The IoT device uses the Modbus / MQTT interface to collect data at 1Hz for high-risk and 10Hz for normal operation; the mobile terminal uploads JSON data via HTTP / HTTPS interface, and the edge terminal parses, normalizes, and adds metadata. In this project, IoT devices such as foundation pit settlement gauges and tower crane tilt sensors are connected to the edge gateway using Modbus RTU / TCP and MQTT standardized interfaces. High-risk monitoring data is collected at a frequency of 1Hz, while ordinary monitoring data is collected at a frequency of 10Hz. On-site management personnel report construction progress, quality acceptance, and other data via mobile devices, which are then uploaded to the edge gateway in JSON format via HTTP / HTTPS interfaces. The edge gateway performs protocol parsing and format normalization on all types of raw data, converting them into a unified standard data format and adding metadata such as device ID, collection timestamp, and spatial code to form standardized data messages.

[0034] Step S133: Deploy MQTT / ZeroMQ lightweight message queues at the edge and in the cloud, build a two-level architecture, with a single message ≤1KB, a cache limit of 1 hour for high-risk data, and trigger local persistence if the limit is exceeded; MQTT and ZeroMQ lightweight message queues are deployed on the project's edge gateway and cloud platform respectively to build a two-level transmission architecture of "edge-side local queue - cloud-side core queue". The queue transmission parameters are configured so that the size of a single message does not exceed 1KB, which is suitable for the narrow bandwidth environment of the construction site. The upper limit of the queue cache is set to 1 hour of high-risk monitoring data. When the cached data exceeds the upper limit, the edge-side local persistent storage is automatically triggered to write the overflow data to the local storage medium to avoid the loss of high-risk data and ensure the stability of data transmission.

[0035] Step S134: Set three-level data priority, add identifiers to the message header, and automatically mark the edge end; The project establishes a three-tiered data priority rule, designating high-risk safety data such as tower crane tilt angle and deep foundation pit settlement as top priority, key process progress and large machinery operation data as first priority, and ordinary environmental monitoring and material arrival records as ordinary priority. A fixed-length priority identifier field is added to the data packet header. When the edge gateway collects data, it automatically tags the data packet with the corresponding priority according to the preset priority rule, completing the initial priority labeling and providing a basis for subsequent hierarchical transmission.

[0036] Step S135: Configure multi-priority independent queues, adopt preemptive scheduling, with the highest level latency ≤500ms and the lowest level latency ≤1s. If the threshold is exceeded, an alarm will be triggered and a backup link will be switched. The project configures independent message queue channels for data of different priorities and adopts a preemptive priority scheduling algorithm. The cloud consumer first listens to the top priority queue, and then processes the first-level and ordinary priority queues in turn. A data transmission service level protocol is set, with the latency of top-level data transmission not exceeding 500ms and the latency of first-level data transmission not exceeding 1s. When the data transmission latency exceeds the threshold, the edge gateway immediately triggers a local alarm and automatically switches to the backup transmission link to ensure the timeliness of high-priority data transmission.

[0037] Step S136: Deploy a preprocessing engine at the edge to remove outliers, downsample ordinary data, and proactively push data when special / first-level data exceeds limits or undergoes sudden changes; A lightweight data preprocessing engine is deployed at the project's edge gateway to remove abnormal and sudden values ​​collected by sensors using a sliding window algorithm. Ordinary priority data is downsampled along the time dimension to reduce the amount of invalid data transmission. For high-priority and first-priority data, the edge end calculates the deviation between the monitored values ​​and the safety threshold in real time. The active push mechanism is only triggered when the data exceeds the safety threshold or a sudden change occurs. Regular compliant data is transmitted at the lowest frequency to free up transmission bandwidth and ensure the transmission of high-risk data.

[0038] Step S137: Configure 5G and wired dual links for premium / primary data, persist queues, issue alarms for premium data after 3 failed retransmissions, and encrypt data with TLS1.3 and add a checksum. The project configures a 5G and wired dual-link hot backup mechanism for top-priority and first-priority data, with a link switching latency of no more than 100ms; the message queue enables persistent storage functions for local storage at the edge and distributed storage in the cloud; when top-priority data transmission fails, it performs three retransmissions at intervals of 100ms, 300ms, and 500ms respectively, and if all three retransmissions fail, an alarm is triggered on the management terminal; all transmitted data is encrypted using the TLS1.3 encryption protocol, and a checksum is added to the message to prevent tampering and leakage during data transmission.

[0039] Step S138: Data is distributed in the cloud according to priority. The highest priority triggers model linkage and early warning. The first priority updates the construction status. The ordinary batch is put into the database. The transmission indicators are monitored. The highest priority automatically optimizes when the latency exceeds the standard. The cloud-based message queue consumer distributes and processes data according to priority. Top-priority data is directly pushed to the real-time computing engine of the BIM-GIS fusion platform, triggering model linkage and risk warnings. First-level data is pushed to the construction business processing module to update construction progress and machinery status in real time. Ordinary data is entered into the spatiotemporal data base in batches. A transmission monitoring module is deployed to collect real-time statistics on transmission latency, packet loss rate, and other indicators. When top-priority data continuously exceeds the latency threshold, the system automatically optimizes the collection frequency and queue cache parameters to ensure transmission performance.

[0040] Step S139: Monitor queue performance, adjust bandwidth ratio when high-priority queues are piling up, generate daily reports, and optimize edge node deployment and scheduling rules.

[0041] The project monitors message queue performance metrics such as throughput, queue length, and consumption rate in real time. When data accumulates in high-priority queues, the system automatically reduces the collection frequency of ordinary priority data and increases the bandwidth ratio of high-priority data transmission. Daily data transmission performance reports are generated to analyze transmission bottlenecks in various construction zones and data types. Based on the report results, the system iteratively optimizes the deployment location of edge nodes and message queue scheduling rules to continuously improve data transmission efficiency.

[0042] Step S140: Establish a two-way link between BIM-GIS and real-time data, and automatically locate the corresponding spatial points when the data changes; Based on the spatiotemporal correlation primary key, this residential project establishes a two-way correlation mechanism between BIM model, GIS spatial elements and real-time monitoring data. When real-time data such as sensor data and construction records are updated or changed, the system automatically matches the corresponding BIM components and GIS spatial points, quickly locates the spatial location of the data, and realizes real-time linkage between data changes and spatial points, providing accurate spatial positioning support for data visualization and risk warning.

[0043] Step S141: Analyze the relationship dimensions between BIM components, GIS spatial elements and real-time data. Using the spatiotemporal relationship primary key as a unified identification benchmark, establish a three-dimensional relationship index table of BIM component ID, GIS spatial code and real-time data source ID. Clarify the BIM component attributes and GIS spatial range corresponding to each type of real-time data to ensure that the relationship is unique and traceable. The project streamlined the relationships between BIM components, GIS spatial elements, and real-time data. Using the spatiotemporal association primary key as a unified identifier, a three-dimensional association index table was established, consisting of BIM component ID, GIS spatial code, and real-time data source ID. The project clarified the attributes of BIM components such as foundation pit support components and tower crane components, as well as the GIS spatial scope such as construction zones and monitoring points. Each type of real-time data was bound to the corresponding component and spatial point to ensure that the association relationship of each data is unique, traceable, and free from duplication or conflict.

[0044] Step S142: Define a bidirectional mapping rule base, and configure corresponding mapping rules for IoT sensor data, construction record data, and drone real-world data respectively to achieve bidirectional traceability of data and spatial elements; The project defines a two-way mapping rule base. For IoT sensor monitoring data, it configures mapping rules for sensor IDs to BIM component IDs and GIS grid codes. For construction record data, it binds mapping rules for process codes, individual building IDs, and GIS construction zones. For UAV real-scene modeling data, it associates the GIS boundary of the shooting area with the BIM model version number to achieve two-way traceability of real-time data and spatial elements. Users can look up space through data and look up data through space.

[0045] Step S143: Deploy the data change monitoring engine to capture data change events in real time and extract the identifiers and content of the changed data by monitoring the data transmission channels of edge nodes, BIM model update interfaces, and GIS spatial feature editing interfaces. The project deploys a data change monitoring engine to monitor edge node data transmission channels, BIM model update interfaces, and GIS spatial element editing interfaces in real time. When events such as sensor value updates, new construction records, BIM component modifications, and GIS zoning adjustments occur, the engine immediately captures the data change event, extracts core identifiers such as data source ID, component ID, and spatial code, as well as changes in data values ​​and status, providing triggering conditions for subsequent correlation and positioning.

[0046] Step S144: When real-time data changes, the data change monitoring engine triggers a related index query, retrieves the three-dimensional related index table based on the source ID of the changed data, matches the corresponding BIM component ID and GIS spatial code, and obtains the component coordinates, GIS geographic coordinates and spatial hierarchy information. When real-time project data changes, the data change monitoring engine immediately triggers the associated index query process; it retrieves the 3D associated index table based on the data source ID, quickly matches the corresponding BIM component ID and GIS spatial code, extracts the component's geometric center coordinates, GIS geographic coordinates, and spatial hierarchy information such as construction zones and monitoring points, and completes the accurate matching of data and spatial elements.

[0047] Step S145: Call the spatial positioning interface of the BIM-GIS fusion platform, highlight the model components and display the attribute pop-up based on the matched BIM component ID, and simultaneously mark the data change location on the map layer according to the GIS spatial code to complete the forward positioning of real-time data to spatial points; The system calls the spatial positioning interface of the BIM-GIS fusion platform, triggers the highlighting of model components and the display of attribute pop-ups based on the matched BIM component ID; simultaneously, according to the GIS spatial code, it marks the specific location of data changes in the GIS map layer, and uses dynamic labels to mark monitoring points and construction areas to complete the forward positioning of real-time data to spatial points, intuitively displaying the spatial location to which the data belongs.

[0048] Step S146: Configure the data reverse association rules triggered by spatial elements. When the user clicks on a BIM component or GIS spatial point, the platform extracts the corresponding identifier and performs a reverse query on the associated real-time data, pulls and displays the latest data details, and completes the reverse association from spatial point to real-time data. The project configures spatial elements to trigger data reverse association rules. When a user clicks on a BIM component or GIS spatial point on the platform, the system automatically extracts the corresponding BIM component ID or GIS spatial code; it then uses the 3D association index table to query the associated real-time data source, calls the data interface to pull the latest monitoring curves, construction record details, and other data, and displays them synchronously on the model and map interfaces, completing the reverse association from spatial points to real-time data.

[0049] Step S147: Establish a dynamic adaptation mechanism for association rules. For newly added real-time data types, BIM model updates, and GIS spatial adjustments, automatically update the 3D association index table and mapping rule base, and support manual supplementation and correction of association relationships. The project establishes a dynamic adaptation mechanism for association rules. When new real-time data types such as sensors or drone inspections are added, or when components are added / deleted from the BIM model or GIS construction zones are adjusted, the system automatically updates the 3D association index table and the two-way mapping rule base. At the same time, the platform builds a visual maintenance interface to support managers to manually supplement and correct association relationships, ensuring that association rules always adapt to the actual situation of the project during construction and guaranteeing the effectiveness of data and spatial association.

[0050] Step S1471: Construct a multi-dimensional change monitoring trigger to listen for new protocol registration, BIM model update, and GIS spatial adjustment events, capture parameters in real time, and trigger the adaptation process; Taking a high-rise residential construction project as an example, a multi-dimensional change monitoring trigger is constructed to listen for new protocol registration events in the data access middleware layer (such as protocol registration for adding LoRa sensors), component addition / deletion / version update interfaces in the BIM model management platform, and construction zone boundary adjustment and coding modification operations in the GIS spatial editing tool. The triggers use a sampling frequency of 50ms to capture metadata (sensor ID, monitoring parameters), BIM component change identifiers (new component ID, deleted component list), GIS spatial adjustment range (zone boundary coordinates, new coding), and other core parameters in real time. Once a change event is captured, the associated rule adaptation process is immediately triggered to ensure real-time response to changes.

[0051] Step S1472: Establish a mapping rule template library categorized by data type, develop an automatic adaptation engine, execute adaptation logic for different change scenarios, and synchronously adopt a transactional update mechanism to ensure data consistency; The project establishes a mapping rule template library categorized by data type, with pre-set standard templates for IoT sensors, construction records, and drone real-world views (e.g., IoT templates include sensor ID, BIM component ID, and GIS grid code fields). An automatic adaptation engine is developed to match similar templates to generate initial rules and write them to the rule library when a new data type is added. When a new BIM component is added, the GIS code is automatically associated and the index table is updated. When GIS zoning is adjusted, the associated entries are updated synchronously. A transactional update mechanism is adopted, locking the target entry before updating. If the update fails, a rollback is triggered and a log is recorded to ensure data consistency between the 3D association index table and the mapping rule library.

[0052] Step S1473: Build a visual manual maintenance interface, providing functions for adding, deleting, modifying, and querying index tables and mapping rules, batch importing and exporting, and searching and filtering, and supporting manual association of key identifiers; A visual manual maintenance interface is built, providing CRUD (Create, Read, Update, Delete) functions for the 3D association index table and mapping rules. It allows administrators to quickly associate BIM component IDs, GIS spatial codes, and data source IDs via drop-down selection boxes. The interface integrates batch import and export functions, allowing batch uploading of associated data via Excel. It also includes built-in search and filtering components by construction zone, data type, and component number, enabling quick location of items requiring correction. For example, when a newly added sensor is not automatically bound to a component, administrators can manually match the corresponding BIM component with the GIS zone through the interface, making the operation intuitive and efficient.

[0053] Step S1474: Embed the association validity verification logic. After automatic adaptation or manual modification, verify the validity of component ID, spatial code, data source ID and spatial coordinate deviation to prevent invalid association writing. Embedded association validity verification logic automatically checks whether the BIM component ID exists in the model, whether the GIS spatial code conforms to the project coding standard, and whether the data source ID is unique after automatic adaptation or manual modification. At the same time, it compares the spatial coordinate deviation of the associated data (the deviation between BIM component coordinates and GIS coordinates is ≤ ±5cm). If the verification finds "component ID does not exist", "GIS code is invalid", or the coordinate deviation exceeds the limit, a prompt will immediately pop up and prevent invalid association writing, guiding the administrator to correct and resubmit, ensuring the accuracy and validity of the association relationship.

[0054] Step S1475: Establish a change log traceability system to record all operation details, supporting conditional filtering queries and historical version rollback; Establish a change log traceability system to record in detail all automatic adaptations and manual operations, including change time, change type (automatic adaptation / manual modification), operator, original data content, and new data content. The log supports filtering and querying by time range, change type, construction zone, and other conditions. It also provides a historical version rollback function. When an adaptation error is found, the association relationship of a specified historical version can be restored with one click, which facilitates problem tracing and rapid fault repair and ensures the traceability of association rule adjustments.

[0055] Step S1476: Define adaptation priority rules, execute multiple changes in a preset order when they occur concurrently, enable incremental adaptation in high-frequency scenarios, and develop an abnormal adaptation handling mechanism to push alarms and adaptation suggestions. Define adaptation priority rules to execute multiple concurrent changes in the order of "BIM component changes > GIS spatial adjustments > new data types" to avoid adaptation conflicts; for high-frequency change scenarios such as batch addition of sensors, enable incremental adaptation mode to process only new / changed items without scanning the full data; develop an abnormal adaptation handling mechanism to automatically trigger alarms and push them to the management terminal for niche data types without corresponding templates or scenarios where the relationship between BIM and GIS is unclear, along with change details and adaptation suggestions (such as "It is recommended to select the IoT sensor mapping template") to guide manual adaptation.

[0056] Step S1477: Optimize adaptation performance by caching frequently accessed data to distributed nodes and adopting an asynchronous parallel processing mode for batch change scenarios.

[0057] To optimize adaptation performance, high-frequency access data (such as data related to high-risk areas like tower cranes and deep foundation pits) from the mapping rule base and the 3D association index table are cached to distributed cache nodes (Redis cluster). During automatic adaptation, cached data is read first to reduce database access pressure. For batch change scenarios such as construction zoning adjustments, an asynchronous parallel processing mode is adopted, with the background performing adaptation operations in batches without blocking front-end interaction, ensuring that the platform response speed is not affected during the adaptation process.

[0058] Step S148: Embed spatial consistency verification logic, compare the deviation between the GIS geographic coordinates associated with real-time data and the global geodetic coordinates of BIM components, and trigger a coordinate calibration prompt and correct the location information when the deviation exceeds the preset threshold; In high-rise residential projects, spatial consistency verification logic is embedded in the BIM-GIS fusion platform to compare the deviation between the GIS geographic coordinates associated with real-time data from IoT sensors and the global geodetic coordinates of the corresponding BIM components in real time. For example, when comparing the GIS installation coordinates of the foundation pit settlement sensor with the global coordinates of the BIM support component, if the deviation exceeds the preset threshold (±5cm for ordinary points and ±3cm for high-risk points), the system automatically triggers a coordinate calibration prompt and starts the automatic correction logic to adjust the coordinate information to ensure the accurate association between data and spatial points.

[0059] Step S1481: Determine the National Geodetic Coordinate System as the sole verification benchmark, preset multi-dimensional coordinate deviation thresholds, support custom downward adjustment for high-risk points, and the threshold parameters can be flexibly configured through the system interface; The project explicitly adopts the CGCS2000 National Geodetic Coordinate System as the sole spatial verification benchmark, with preset multi-dimensional coordinate deviation thresholds: plane position (X / Y axis) ±5cm, elevation (Z axis) ±3cm, and comprehensive spatial deviation ≤±5cm. For high-risk monitoring points such as tower cranes and deep foundation pits, the threshold can be customized to ±3cm through the system configuration interface. All threshold parameters can be flexibly adjusted according to the construction accuracy requirements without modifying the underlying code, adapting to the accuracy requirements of different construction stages.

[0060] Step S1482: Develop a coordinate extraction and transformation submodule. During verification, automatically extract GIS geographic coordinates and BIM local coordinates, and unify them into global geodetic coordinates through a two-way conversion and adaptation module to ensure consistent coordinate reference. A coordinate extraction and transformation submodule was developed. When verification is triggered, it automatically extracts the real-time data-related GIS geographic coordinates (latitude and longitude / geocentric coordinates) from the 3D association index table, and simultaneously retrieves the local engineering coordinates of the BIM components. The bidirectional coordinate transformation and adaptation module is called, and the transformation parameters (translation, rotation angle, scaling ratio) measured in the project are input to accurately convert the local BIM coordinates into CGCS2000 global geodetic coordinates, ensuring the consistency of GIS and BIM coordinate benchmarks and providing a consistent basis for deviation calculation.

[0061] Step S1483: Design the three-dimensional spatial deviation calculation logic to complete the plane, elevation and comprehensive deviation calculations, and ensure millisecond-level response by executing efficiently in memory; The design of the three-dimensional spatial deviation calculation logic extracts the converted BIM component coordinates and GIS geographic coordinates, and calculates the planar deviation (X / Y axis combined deviation), elevation deviation (Z axis deviation) and comprehensive spatial deviation respectively. All calculation processes are completed in memory, using an optimized Euclidean distance algorithm to avoid disk I / O operations, ensuring that the calculation response time is ≤10ms, meeting the low latency requirements of real-time verification, and not affecting the overall operating efficiency of the platform.

[0062] Step S1484: Configure three types of verification mechanisms: data change trigger, timed verification, and manual trigger. When triggered, the unique identifier of the verification object is automatically marked. Data change triggers (automatically triggered when real-time data is updated or BIM / GIS is adjusted), scheduled verifications (full verification is performed on all points every morning, and sampling verification is performed on high-risk points every hour), and manual triggers (managers can select a specified zone / point to initiate an instant verification on the platform interface); when a verification is triggered, a unique identifier (component ID / monitoring point ID) of the verification object is automatically marked to facilitate subsequent tracking of verification results and anomaly handling.

[0063] Step S1485: Divide the comprehensive spatial deviation value into four levels: compliant, minor, moderate, and severe, and execute the corresponding graded processing rules of no action, pop-up prompt, automatic correction, and alarm suspension of data association. A comprehensive deviation of ≤5cm is considered compliant and will not trigger any action; a deviation of 5cm < ≤10cm is considered a minor deviation and will only trigger a pop-up notification on the platform interface; a deviation of 10cm < ≤20cm is considered a moderate deviation and will trigger a notification and initiate automatic preliminary correction; a deviation >20cm is considered a severe deviation and will immediately trigger an alarm and suspend data association for that location until manual intervention is required to ensure that deviations of different degrees are accurately controlled.

[0064] Step S1486: When there is a moderate deviation, trace the source and make targeted automatic corrections, update the coordinate information of the three-dimensional association index table, and record the correction details simultaneously; If the BIM coordinate transformation parameter drift is the cause, the historical transformation parameter records will be automatically retrieved, the transformation matrix will be recalculated, and the transformation matrix will be updated. If the GIS coordinate acquisition error is the cause, the GIS coordinates will be corrected using the BIM component coordinates (design baseline). If the sensor installation offset is the cause, the BIM associated coordinates will be corrected using the on-site measured GIS coordinates. After correction, the coordinate information in the 3D associated index table will be automatically updated, and the coordinate values ​​before and after correction, the cause of the deviation, and the correction timestamp will be recorded synchronously to ensure that the correction is traceable.

[0065] Step S1487: Build a multi-dimensional alert system, mark points and display information according to the deviation level, and push alarms through multiple channels for severe deviations; A multi-dimensional alert system is established. Minor deviations are marked in yellow, moderate deviations in orange, and pop-up windows display the deviation value, cause analysis, and correction suggestions. Severe deviations are marked in red. In addition to the on-screen prompts, alarm information is simultaneously pushed to the administrator's mobile APP, WeChat, and SMS, along with the location ID, deviation value, and a link to manual calibration, ensuring that administrators are aware of and can handle serious deviations in a timely manner.

[0066] Step S1488: Design a verification process for the correction results, including automatic correction followed by re-inspection, manual correction followed by manual confirmation, and the system retaining complete verification records. The design includes a verification process for corrected results. After automatic correction, the coordinate deviation calculation is immediately re-executed. If the corrected deviation is less than or equal to the preset threshold, it is considered valid; otherwise, it is upgraded to a severe deviation and an alarm is triggered. After manual correction, the administrator needs to confirm the verification results on the platform and enter the correction description (such as "re-measure and calibrate the sensor position on site"). The system retains a complete verification record to ensure that the correction effect meets the standards and avoids invalid correction.

[0067] Step S1489: Establish a verification and correction log system to record core information and support multi-condition filtering and querying; Establish a verification and correction log system to record the trigger time, location identifier, BIM coordinates, GIS coordinates, deviation value, judgment result, processing method (prompt / automatic correction / alarm), processor and correction result for each verification. The log supports filtering and querying by time range, deviation level, construction zone and other conditions, which makes it easier for managers to analyze the high-incidence areas and causes of deviations and provide data support for subsequent optimization.

[0068] Step S14810: Optimize verification performance by caching high-frequency coordinates and batch asynchronous processing, and periodically analyze logs and optimize coordinate transformation parameters.

[0069] To optimize verification performance, coordinate transformation results of high-frequency verification points such as tower cranes and deep foundation pits are cached to distributed cache nodes to reduce redundant calculations. In batch verification scenarios, an asynchronous parallel processing mode is adopted, with the verification logic executed in batches in the background without blocking the front-end interaction. Verification logs are automatically analyzed weekly to identify high-frequency deviation types (such as transformation parameter drift), and the parameters of the bidirectional coordinate transformation adaptation module are optimized to reduce the deviation rate.

[0070] Step S149: Deploy the three-dimensional association index table in a distributed cache to establish a hot data cache pool for high-frequency changing real-time data, optimize the performance of association queries, and control the spatial positioning delay after data changes; The three-dimensional relational index table is deployed on a Redis distributed cache cluster, and a master-slave replication architecture is used to ensure high availability. For high-frequency change real-time data such as tower crane tilt angle and foundation pit settlement collected at 1Hz, a hot data cache pool is established, with a cache validity period of 5 minutes, and the cache is refreshed synchronously when updated. The cache accelerates relational queries, ensuring that the spatial positioning latency after data change is ≤300ms, meeting the real-time interaction requirements and improving the platform response speed.

[0071] Step S1410: Establish an exception handling mechanism to log and trigger alarms for exceptions such as missing indexes and invalid associations, prompting administrators to supplement association information.

[0072] An anomaly handling mechanism is established. When the corresponding spatial point is not found in the 3D association index table after data changes (index missing), or the association fails due to the deletion of the associated BIM component, the system automatically records the anomaly log (including anomaly type, trigger time, and data identifier) ​​and pushes alarm information through the platform alarm bar and the administrator's APP, prompting the supplementation of association information. At the same time, a quick processing entry is provided, and the administrator can directly jump to the association maintenance interface to complete the supplementation, ensuring the integrity of the two-way association.

[0073] Step S150: Incrementally update data and models by component / location / region, and synchronously refresh storage and display; Incremental updates are carried out at three levels: component unit, monitoring point, and construction area. When updating at the component level, only the changed BIM component attributes and related data are synchronized, and the full model is not loaded. When updating at the point level, only the changed IoT monitoring data and identification status are refreshed. When updating at the area level, the GIS spatial adjustments and related data changes within the area are synchronized. After the update, it is automatically synchronized to the spatiotemporal data base and BIM-GIS display interface. Partial refresh technology is used to avoid the lag caused by full loading and to ensure update efficiency and smooth visualization.

[0074] Step S160: Cross-validate multi-source data, identify anomalies through spatial matching and logical verification, automatically remove anomalies and trigger supplementary data collection; Taking the construction of foundation pits for high-rise residential buildings as an example, multi-source data is cross-validated: IoT settlement sensor data, BIM support component design data, GIS foundation pit zoning data, UAV real-scene modeling outline data, and construction pouring records are correlated and compared; data with coordinate deviations exceeding 10cm are removed through spatial matching verification, and abnormal data with settlement values ​​exceeding the standard limit (daily change ≤ 5mm) are identified through logical verification. Abnormal data is automatically isolated and removed, and supplementary collection instructions are sent to edge nodes. After the data is supplemented, it is re-verified to ensure the accuracy of the data entering the database.

[0075] Step S161: Define the verification scope of BIM model, GIS geospatial data, IoT monitoring, construction business records, and UAV real-scene modeling data. Using the project's global geodetic coordinate system as a unified spatial reference, formulate a quantifiable verification benchmark library that includes equipment range, construction specifications, spatiotemporal correlation, and data integrity. The scope of verification is clearly defined, covering BIM model data, GIS geospatial data, IoT monitoring data, construction business records, and UAV real-scene modeling data. Using the CGCS2000 National Geodetic Coordinate System as a unified spatial benchmark, a quantifiable verification benchmark library is established, including: equipment range thresholds (such as ±10° for tower crane tilt sensors), construction specification limits (such as daily settlement variation of deep foundation pits ≤5mm), spatiotemporal correlation logic (such as construction record time must be within the corresponding process period), and data integrity requirements (including spatiotemporal primary keys, timestamps, and equipment IDs), providing a unified standard for cross-verification.

[0076] Step S162: Construct a multi-source data association map based on the spatiotemporal correlation primary key, bind multiple types of data at the same spatial location and the same time node to form a spatiotemporal association network that can be synchronously cross-compared; Based on the spatiotemporal correlation primary key, a multi-source data correlation map is constructed: IoT settlement data, BIM support component design thresholds, GIS foundation pit zoning codes, construction pouring records, and local data from UAV real-scene modeling at the same foundation pit monitoring point are bound to form a spatiotemporal correlation network of "spatial code-time stamp-data type-data value"; for example, the settlement data of a certain monitoring point is associated with the construction pouring record with the same time stamp and the BIM component threshold with the same spatial code, ensuring that different data sources can be cross-compared based on the same spatiotemporal dimension.

[0077] Step S163: Design spatial matching verification and multi-dimensional logical verification rules, compare spatial position deviation and attribution rationality after unifying coordinates, and complete numerical, time, association and integrity verification according to data type; All data coordinates were uniformly converted to the CGCS2000 coordinate system. The coordinate deviations between the drone-based real-world component outlines and the BIM model components were compared (general points ≤ 10cm, high-risk points ≤ 5cm). The consistency between the construction record zoning code and the GIS zoning code was verified. Multi-dimensional logical verification rules were established: numerical verification (data within equipment range / specification limits), time verification (collection time within the construction period / process period), correlation verification (multi-source data mutually verify each other), and integrity verification (no missing core fields), comprehensively covering data anomaly scenarios.

[0078] Step S164: Build a cross-validation engine that first performs spatial matching and then logical verification, execute the verification in sequence, mark the anomaly type, and record the anomaly data identifier, deviation value and conflict details; A cross-validation engine is built, which executes the process in the order of "spatial matching first, then logical verification": First, spatial matching verification is performed on the associated data to remove data with excessive coordinate deviation or abnormal spatial attribution; then, logical verification is performed on the data that passed the spatial verification, and the anomaly type (numerical anomaly, time anomaly, association anomaly, integrity anomaly) is marked, and the anomaly details (anomaly data ID, deviation value / conflict point, associated baseline data) are recorded; the engine adopts a parallel computing architecture to improve verification efficiency and meet the needs of real-time data processing.

[0079] Step S165: Isolate and automatically remove abnormal data, classify the abnormality into mild, moderate and severe levels, trigger the corresponding graded supplementary data collection rules, and send instructions to edge nodes to complete data supplementary collection; For data identified as abnormal, it is first temporarily isolated to the abnormal data buffer (preserving the original data traceability), and then automatically removed from the normal data link; supplementary data collection is triggered according to the degree of abnormality: for mild abnormalities (single missing data, spatial deviation ≤15cm), a single supplementary data collection instruction is issued to the edge node (timeout 5 minutes); for moderate abnormalities (3 consecutive abnormal values, spatial deviation 15-20cm), 3 consecutive supplementary data collections are triggered (1 minute interval) and a prompt is pushed; for severe abnormalities (batch spatial matching failure, logical conflict of high-risk points), full supplementary data collection is triggered, data entry in the area is suspended and an alarm is set, and backup sensors are activated to temporarily supplement data.

[0080] Step S166: Re-verify the supplementary data. If it passes, it is included in the normal data link; if it fails, it is upgraded to an anomaly and manual intervention is triggered. The system retains the processing results. After the supplementary data collection is triggered, the supplementary data collection execution module generates a supplementary data collection instruction containing the spatiotemporal primary key, data type, and precision requirements, and sends it to the corresponding edge node. The edge node interrupts normal data collection and prioritizes the supplementary data collection. The supplementary data is marked as "supplementary data collection" and then re-enters the cross-validation process. If the validation passes, it is included in the normal data link and the isolation status of the corresponding original abnormal data is cleared. If it is still abnormal, the abnormality level is upgraded and manual intervention is triggered. After viewing the details, the administrator can choose to remove or adjust the rules to re-collect the data, and the system records the processing results.

[0081] Step S167: Establish an anomaly tracing and rule optimization mechanism, analyze the causes of anomalies and iteratively verify the rules, optimize the verification performance by combining caching, asynchronous parallelism and full sampling, and record the verification, anomaly, supplementary sampling and processing logs throughout the process.

[0082] Establish an anomaly tracing mechanism, regularly collect statistics on the types of abnormal data, high-incidence construction zones, and high-incidence data sources, and analyze the causes (such as sensor failure, coordinate transformation error, and manual data entry error); iteratively optimize verification rules for high-frequency anomaly types (such as adjusting sensor value thresholds); optimize verification performance: cache correlation maps and verification rules, perform asynchronous parallel verification of batch data, perform sampling verification of low-priority data, and perform full verification of high-priority data; establish a full-process log system to record all information on anomaly identification, removal, re-collection, and processing, and support multi-condition filtering and querying.

[0083] Step S170: Adaptively adjust the model display accuracy, overlay real-time data visualization labels, locate risks, highlight the model, and push warnings when thresholds are exceeded.

[0084] Taking high-rise residential construction projects as an example, the BIM-GIS fusion platform automatically adapts the model display accuracy: a low-precision model is loaded during global browsing, and a high-precision model is switched when focusing on the tower crane area; real-time data labels from IoT sensors are overlaid (normal green circle, slightly excessive yellow triangle, and severely excessive red diamond). When the deep foundation pit settlement data exceeds the 5mm threshold, the system automatically locates the risk points, highlights the corresponding BIM support components and GIS areas, and simultaneously pushes warnings to the management personnel's APP, realizing the linkage of accuracy adaptation, data visualization, and risk warning.

[0085] Step S171: Divide the model display accuracy into three levels of low, medium and high according to the construction scenario and terminal equipment type, and define the display rules of BIM model and GIS elements under each accuracy. At the same time, set adaptive trigger conditions such as view zoom, interactive operation and data update frequency to form an accuracy adaptation rule library. The project is divided into three accuracy levels based on construction scenario (global browsing, zone focus, risk viewing) and terminal type (PC / mobile / large screen): low accuracy BIM with 10% facets, GIS only showing zone outlines; medium accuracy with 50% facets, GIS with key points added; and high accuracy with full loading. Trigger conditions are set: low accuracy > 50 meters, medium accuracy 20-50 meters, and high accuracy < 20 meters; automatic switching to high accuracy occurs when clicking on a component or when high-risk data is updated at 1Hz, forming a structured accuracy adaptation rule base that supports dynamic matching by scenario and terminal.

[0086] Step S172: Construct a model accuracy adaptive rendering engine integrated into the visualization kernel of the BIM-GIS fusion platform, monitor the performance of terminal devices, view parameters and data interaction requirements in real time, dynamically adjust the model loading strategy according to the accuracy adaptation rule base, and preload medium / high accuracy model caches for high-frequency interest areas. The adaptive rendering engine is integrated into the platform's visualization kernel to monitor terminal CPU / memory usage, viewpoint height, and data update frequency in real time. The loading strategy is dynamically adjusted according to the rule base: low-precision models are batch-loaded using instantiated rendering, while high-precision models are loaded on demand using LOD technology. Medium / high-precision models of high-frequency areas such as deep foundation pits and tower cranes are preloaded to the local cache to ensure a stable rendering frame rate of over 30 frames per second, with no delay in precision switching, adapting to different terminals and interaction requirements.

[0087] Step S1721: Design a five-layer modular architecture consisting of a monitoring and perception layer, a rule parsing layer, a scheduling and execution layer, a cache management layer, and a performance feedback layer. Each layer interacts through standardized interfaces, embedding the overall architecture into the rendering pipeline of the BIM-GIS fusion platform's visualization kernel, and decoupling it from the platform's existing model loading and view control modules. The design employs a five-layer architecture: a monitoring and perception layer, a rule parsing layer, a scheduling and execution layer, a cache management layer, and a performance feedback layer. Each layer interacts via standardized APIs. The entire architecture is embedded within the platform's visualization kernel rendering pipeline, decoupled from the existing model loading and view control modules, without affecting core functionality. The monitoring layer captures terminal and interaction states, the rule layer matches precision rules, the scheduling layer outputs loading instructions, the cache layer manages preloaded data, and the feedback layer sends back performance data, forming a closed-loop control system.

[0088] Step S1722: Develop a real-time monitoring submodule for terminal device performance, call the terminal system API to collect five core indicators at a sampling frequency of 100ms: CPU utilization, memory utilization, GPU load, current rendering frame rate, and network bandwidth, preset performance thresholds for different terminal types, and send a performance degradation signal to the rule parsing layer and record performance data when the indicators exceed the thresholds. A terminal performance monitoring submodule was developed, which calls the system API to collect five types of indicators, including CPU utilization and memory usage, at a sampling frequency of 100ms. Preset thresholds are set: PC memory > 85%, mobile > 80%, and frame rate < 25 frames per second. If any indicator exceeds the limit, a degradation signal is sent. For example, when mobile memory usage reaches 82%, a signal is automatically pushed to the rule layer, triggering a reduction in model accuracy from high to medium, while simultaneously recording performance data for optimization.

[0089] Step S1723: Develop a sub-module for monitoring view parameters and data interaction requirements, listen to view control events on the BIM-GIS platform and parse view height, angle, and field of view parameters, capture user interaction operations and data update frequency events and bind trigger tags, and transmit the above parameters and tags to the rule parsing layer in real time. The system monitors platform view control events (scrolling, dragging), analyzes and standardizes view height, angle, and field of view in real time; it captures user clicks / selection of components, 1Hz high-risk data updates, and binds them with tags such as "view distance < 20 meters" and "click on tower crane". For example, when a user selects a deep foundation pit area, the submodule extracts the "zone focus" tag, transmits it to the rule parsing layer in real time, and triggers a high-precision loading command.

[0090] Step S1724: Develop a precision adaptation rule base parsing and matching submodule, encapsulate precision rules into a JSON format structured rule set, support combination matching of construction scenarios, terminal types, and trigger conditions, match monitoring signals according to terminal type priority and trigger condition supplementary priority, and output target precision level instructions according to the performance priority principle when multiple conditions conflict. The accuracy rules are encapsulated as a JSON structured rule set, supporting combined matching (e.g., "mobile + viewing angle > 50 meters = low accuracy", "PC + clicking on a deep foundation pit = high accuracy"). The parsing module matches monitoring signals according to "terminal type priority, trigger conditions supplementation," and executes according to "performance priority" when multiple conditions conflict. For example, a high-performance PC triggers high accuracy, while a mobile device with memory exceeding the limit is forced to downgrade to medium accuracy, outputting a clear accuracy level instruction.

[0091] Step S1725: Develop a dynamic scheduling submodule for model loading strategy, design differentiated loading logic for low, medium and high precision levels. Low precision mode simplifies the number of BIM model faces and batch renders, only loading GIS partition outlines. Medium precision mode loads core components, GIS partitions and key points and loads them in blocks. High precision mode loads all information on demand and enables LOD technology. After receiving the precision level instruction, the loading strategy is switched within 100ms, and visual discontinuities are avoided through gradual transition animation. Differentiated loading logic is designed for three levels of precision: Low-precision BIM surfaces are reduced to 10%, textures are removed, and only partition outlines are loaded for GIS, using instantiated rendering to reduce draw calls; Medium-precision loads core components, retains basic textures, adds key points to GIS, and loads the visible area model in blocks; High-precision loads all information on demand, enabling LOD technology. The strategy switches within 100ms after receiving the command, using gradual animation to avoid visual discontinuities.

[0092] Step S1726: Develop a high-frequency interest area model cache management submodule to identify high-frequency interest areas such as deep foundation pits and tower cranes and mark their spatial range. When the platform is initialized, preload medium / high precision models to the local cache of the terminal, dynamically allocate cache capacity according to terminal performance, adopt LRU cache eviction policy, listen for model version update events and update the cache. When the cache is not hit, load low precision models first for transition, and supplement medium / high precision models asynchronously in the background. Identify and mark high-frequency areas of interest such as deep foundation pits and tower cranes. During platform initialization, preload medium / high-precision models into the terminal cache (PC memory + disk, mobile memory + SD card), and allocate cache capacity according to terminal performance (PC maximum 5GB, mobile 1GB). Use LRU strategy to evict models that have not been accessed for 2 hours, and listen for model update events to refresh the cache synchronously. When the cache is not hit, load a low-precision model as a transition, and asynchronously supplement the medium / high-precision model in the background.

[0093] Step S1727: Develop a rendering performance closed-loop feedback submodule to monitor rendering frame rate, model loading time, CPU / GPU utilization and other indicators in real time. When the frame rate is below 30 frames and lasts for more than 2 seconds, automatically optimize the loading strategy, collect data such as precision switching time and cache hit rate to generate performance reports, and provide feedback to the rule parsing layer on precision threshold optimization suggestions. Real-time monitoring of rendering frame rate, loading time, and CPU / GPU utilization. When the frame rate drops below 30 frames per second for 2 seconds, the texture resolution is automatically reduced and the number of batch rendering components is decreased until the frame rate recovers. Hourly performance reports are generated to count the precision switching time and cache hit rate. Bottlenecks such as a sudden drop in high-precision loading frame rate in a certain area are identified, and optimization suggestions such as "adjusting the precision trigger threshold in this area" are fed back to the rule layer.

[0094] Step S1728: Develop the engine's fault tolerance and degradation mechanism. When rule base parsing, model loading, or cache reading encounters anomalies, switch to the backup strategy, load the medium-precision model and disable non-core rendering effects, record the exception log and push it to the operation and maintenance end, enable the offline cached low / medium-precision model in a weak network environment, and asynchronously update the high-precision data after the network is restored. When rule base parsing, model loading, or cache reading encounters anomalies, the backup strategy is automatically switched, loading a medium-precision model and disabling non-core rendering effects, and an error log is recorded and pushed to the operations and maintenance end. In weak network environments, offline cached low / medium-precision models are enabled to avoid loading failures, and high-precision data is asynchronously updated after the network is restored to ensure the availability of basic visualization functions.

[0095] Step S1729: Connect the engine with the model management, rendering pipeline, and view control interfaces of the BIM-GIS visualization kernel, conduct full-scale testing of the engine functions under different terminals and different construction scenarios, and iteratively optimize the rule base thresholds and loading strategy parameters.

[0096] The engine was integrated with the platform's visualization kernel's model management, rendering pipeline, and view control interfaces to achieve decoupling and integration. On PC, Android / iOS mobile devices, and large-screen terminals, scenarios such as global browsing, partition focusing, and risk warning were tested to verify the smoothness of precision switching, frame rate stability, and cache hit rate. For the mobile device lag issues that occurred during testing, the low-precision model simplification algorithm and cache allocation strategy were optimized, and the rule base threshold was iterated.

[0097] Step S173: Construct a real-time data visualization labeling system classified by data type and risk level, bind spatiotemporal association primary keys to the labels and specify anchor coordinates, and configure label floating interaction rules; Construct a real-time data visualization identification system, classifying data by data type (IoT monitoring, construction records) and risk level: a green circle (8px in diameter) indicates a normal state, a yellow triangle (10px on the side) indicates a slightly excessive threshold, and a red rhombus (12px on the side) indicates a severely excessive threshold. Each identification is bound to a spatiotemporal key of "spatial code + timestamp", anchored to the geometric center of the BIM component or GIS geodetic coordinates, and the collection time, value, threshold, and deviation rate are displayed when the mouse hovers over them.

[0098] Step S174: Based on the spatiotemporal correlation primary key, accurately overlay the visual identifiers onto the corresponding BIM components / GIS spatial points, and develop a batch rendering module for identifiers to refresh the identifier styles and values ​​according to the data update frequency; Based on spatiotemporal correlation primary keys, the visual identifiers are accurately anchored to the coordinates of the corresponding BIM components / GIS points; a batch rendering module is developed to refresh the identifier styles and values ​​in real time according to the data update frequency (1Hz / 10Hz); it supports hiding / showing a certain type of data identifier separately, and automatically avoids model occlusion when overlaying (the identifier is always displayed on the top layer, and is offset to the visible area when occluded), ensuring that the identifiers are clearly visible and do not affect model browsing.

[0099] Step S175: Build a dynamic threshold judgment engine, connect to the threshold benchmark library composed of construction safety specifications and equipment range, compare data values ​​with corresponding thresholds in real time, classify the risk level into four levels according to the degree of deviation: normal, slightly exceeding the threshold, severely exceeding the threshold, and extreme exceeding the limit, label the data and associate it with the corresponding BIM component ID / GIS spatial code. A dynamic threshold determination engine is built, and the system is connected to the construction safety specifications and equipment range threshold library (such as daily settlement variation of deep foundation pit ≤5mm, tower crane tilt angle ≤3°). The system extracts IoT monitoring and actual construction measurement data values ​​in real time, compares them with the corresponding thresholds, and classifies them according to the degree of deviation: within the threshold is normal, exceeding 5%-10% is slightly exceeding the limit, exceeding 10% is severely exceeding the limit, and exceeding 20% ​​is extremely exceeding the limit. The system also adds risk tags to the data and associates them with the corresponding BIM component ID and GIS spatial code.

[0100] Step S176: When the data exceeds the threshold, match the corresponding BIM component / GIS spatial area through the spatiotemporal primary key, call the spatial positioning interface to jump the model view to the risk point and adapt to high-precision display, and at the same time perform differentiated highlight rendering on the component / area according to the risk level. When data exceeds the threshold, the corresponding BIM component and GIS area are quickly matched using the spatiotemporal primary key. The platform's spatial positioning interface is called to automatically jump the model view to the risk point (focusing on the center and switching to high-precision display). Differentiated highlighting is performed according to the risk level: for minor exceedances, a yellow glowing outline (2px wide) is added to the component; for severe exceedances, a red semi-transparent coloring and outline are added to the area. When locating multiple risk points in batches, severe risks are displayed first.

[0101] Step S177: Design a tiered early warning push mechanism, configure push channels and corresponding content such as platform pop-ups, mobile APP / WeChat for Enterprise, SMS / voice calls according to risk level, and record the push time, recipient, and read status of all early warning information to form a closed loop record; Design a tiered early warning push mechanism: minor exceedances will only trigger a pop-up notification on the platform (including location name, data value, and threshold); severe exceedances will trigger an early warning to the administrator's APP / WeChat Work (with a model location link) in addition to the pop-up notification; extreme exceedances will trigger SMS / voice call warnings, with added deviation rate, standard basis, and emergency suggestions; all warnings will record the push time, recipient, and read status, forming a closed-loop management system for easy traceability.

[0102] Step S178: Optimize visualization and early warning performance by using GPU batch rendering, progressive model loading, and early warning anti-shake mechanism, and support users to customize the label style, threshold judgment deviation ratio, early warning push channel and triggering conditions on the platform interface; GPU batch rendering technology is used to process large batches of labels, reducing terminal pressure; model accuracy adjustment adopts progressive loading (low precision first, then high precision details) to avoid lag; a warning anti-shake mechanism is set up so that if the same point exceeds the threshold repeatedly within 5 minutes, it will only push once to prevent interference; the platform interface supports user-customizable configuration: adjust the label style, threshold deviation ratio, warning push channel (such as turning off SMS warnings) to adapt to personalized needs.

[0103] Step S179: Establish a dynamic adaptation mechanism for display and early warning, monitor changes in the performance of terminal equipment during the construction phase in real time, automatically update the accuracy adaptation rules and threshold benchmark library, regularly statistically analyze the frequency of early warnings and the number of accuracy switching, and iteratively optimize the accuracy rules and threshold parameters.

[0104] Establish a dynamic adaptation mechanism to monitor changes in the construction phase in real time (such as switching from foundation construction to the main structure), automatically update the accuracy rules (the main structure area defaults to medium accuracy) and the threshold library (stress threshold for formwork supports); monitor terminal performance (such as mobile device memory > 80%), automatically downgrade accuracy, and simplify label rendering; count the frequency of early warnings and the number of accuracy switches every week, analyze high-frequency risk points and performance bottlenecks, and iteratively optimize accuracy rules and threshold parameters.

[0105] Based on the same inventive concept, please refer to Figure 2 This paper shows a schematic block diagram of a BIM-GIS fusion-based building construction information management system 100 provided in this application embodiment for executing the above-described BIM-GIS fusion-based building construction information management method. The BIM-GIS fusion-based building construction information management system 100 may include a communication unit 110, a machine-readable storage medium 120, and a processor 130.

[0106] In this embodiment, both the machine-readable storage medium 120 and the processor 130 are located within the BIM-GIS integrated building construction information management system 100 and are separately configured. However, it should be understood that the machine-readable storage medium 120 may also be independent of the BIM-GIS integrated building construction information management system 100 and may be accessed by the processor 130 via a bus interface. Alternatively, the machine-readable storage medium 120 may be integrated into the processor 130 and may communicate and interact with external systems through the communication unit 110.

[0107] The processor 130 is the control center of the BIM-GIS integrated building construction information management system 100. It connects to various parts of the system via various interfaces and lines. By running or executing software programs and / or modules stored in the machine-readable storage medium 120, and by calling data stored in the machine-readable storage medium 120, it performs various functions and processes data of the BIM-GIS integrated building construction information management system 100, thereby providing overall monitoring of the system. Optionally, the processor 130 may include one or more processing cores; for example, the processor 130 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may also not be integrated into the processor. The machine-readable storage medium 120 is used to store machine-executable instructions for executing the solution of this application, and the processor 130 is used to execute the machine-executable instructions stored in the machine-readable storage medium 120 to realize the building construction information management method based on BIM-GIS fusion provided in the aforementioned method embodiments.

[0108] It should be noted that, in order to simplify the description of the present invention and thus help to understand one or more embodiments of the invention, multiple features may sometimes be grouped into one embodiment, drawing or description thereof in the foregoing description of the embodiments of the present invention.

Claims

1. A building construction information management method based on BIM-GIS fusion, characterized in that, The method includes: Multi-source heterogeneous data is uniformly accessed to the middle layer, with a built-in BIM and GIS bidirectional coordinate transformation and adaptation module. Standardized access protocols and protocol parsing components are configured for different types of data sources, such as IoT time-series data, UAV 3D point cloud data, and construction text form data. Based on the geospatial data of the construction site, various types of data are integrated and bound together through a spatiotemporal correlation primary key to construct an integrated spatiotemporal data foundation. Real-time data is collected from nearby edge nodes and transmitted via a lightweight message queue, with high-risk monitoring data being pushed first. Establish a two-way link between BIM-GIS and real-time data, and automatically locate the corresponding spatial points when the data changes; Data and models are updated incrementally by component / location / region, and storage and display are refreshed synchronously. Cross-validation of multi-source data identifies anomalies through spatial matching and logical verification, automatically removes them, and triggers re-collection. The system adaptively adjusts the model display accuracy, overlays real-time data visualization labels, and identifies risks, highlights the model, and pushes warnings when thresholds are exceeded. 2.The BIM-GIS fusion-based construction informatization management method according to claim 1, characterized in that, The aforementioned approach uses the geographical space of the construction site as a benchmark, integrates various types of data, and binds them through spatiotemporal association primary keys to construct an integrated spatiotemporal data foundation, including: A unified spatial reference system is determined based on the geospatial data of the entire construction site. The national geodetic coordinate system is selected as the global spatial reference. The conversion parameters between the local BIM engineering coordinate system and the global geodetic coordinate system are obtained through field measurements. The unique calibration of the spatial reference for the entire site is completed, eliminating spatial reference deviations from multi-source data. Spatial attribute standardization processing is carried out on BIM component information, GIS topographic data, IoT monitoring data, UAV real-scene modeling data and construction business records. This process matches the corresponding spatial points of the construction area to construction records and business forms that lack spatial coordinates, binds the spatial coordinates of the installation location of IoT sensors, and accurately registers the UAV point cloud and BIM model to the global spatial benchmark, thereby achieving spatial normalization of all data. A spatiotemporal hierarchical architecture is constructed, with five levels of structure in the spatial dimension: construction site, construction zone, individual building, component unit, and monitoring point; and four levels of nodes in the temporal dimension: total construction period, sub-project, sub-process, and real-time data collection time, forming a spatiotemporal hierarchical framework covering the entire domain and the entire life cycle. A spatiotemporal association primary key is generated by combining spatial grid coding, timestamp and unique identifier of data object. Spatial location coding, data acquisition or effective timestamp, component number, sensor ID and business record coding are fused and encoded so that each data unit corresponds to a unique spatiotemporal association primary key. According to the preset spatiotemporal hierarchy, the normalized data of various types are accurately mounted to the corresponding spatiotemporal nodes. By establishing a strong binding relationship between BIM component parameters, GIS geographic information, IoT real-time data, real scene model and construction business data through the spatiotemporal association primary key, the orderly collection of multi-source data in the same spatiotemporal system is realized. A spatiotemporal hybrid index is built based on spatiotemporal associated primary keys, supporting multi-dimensional associated retrieval by spatial location, time node, and data type, ensuring rapid data location and associated retrieval; Perform spatiotemporal consistency verification on the mounted data, check the accuracy of spatial binding and the rationality of time correspondence, automatically identify and correct spatiotemporal misalignment and primary key conflict anomalies, and ensure the reliability of data association; The multi-source data that has been bound and verified will be integrated and encapsulated to form an integrated spatiotemporal data foundation with unified storage, unified association, and unified scheduling, providing standardized data support for subsequent data linkage, updates, visualization, and intelligent early warning. 3.The BIM-GIS fusion-based construction informatization management method according to claim 2, characterized in that, The aforementioned spatiotemporal dual-layer hierarchical architecture is constructed, with a spatial dimension dividing the site into five levels: construction area, construction zone, individual building, component unit, and monitoring point; and a temporal dimension dividing the project into four levels: total construction period, sub-projects, sub-processes, and real-time data acquisition time. This forms a comprehensive, full-cycle spatiotemporal hierarchical framework, including: The entire construction site is delineated based on the overall construction plan and GIS geographic data vectorization. Construction zones are divided in conjunction with the construction organization design. Boundaries are drawn using GIS to ensure that the zones do not overlap and are fully covered. Using the BIM model as the core, extract information about individual buildings and associate it with construction zones, disassemble the model to divide it into component units, assign unique numbers and collect spatial coordinates to complete the spatial binding of components with individual buildings. By combining construction risks and quality control points, monitoring points are set up, and the geodetic coordinates are obtained by actual measurement and associated with the corresponding components / areas. A unique identifier is assigned, and the spatial five-level structural division and positioning are completed. The total construction period is determined based on the contract and schedule. The project is broken down into sub-projects and further subdivided into individual work processes according to the specifications. Time connection relationships are set, and the collection frequency and ISO8601 timestamp are unified. The four-level node decomposition and calibration of the completion time are completed. Establish spatiotemporal hierarchical coding rules. Spatial coding adopts site, zone, unit, component, and location hierarchical coding, while temporal coding adopts a combination of construction period, division, process, and timestamp. A two-way spatiotemporal mapping relationship is established through coding. A spatiotemporal hierarchy management module was developed based on the BIMGIS platform, a visual configuration interface was built, supporting hierarchy display and adjustment, and built-in spatial topology and temporal logic verification functions. A dynamic update mechanism for spatiotemporal hierarchy is established. When the construction plan changes, the module automatically updates the hierarchical boundaries, codes, and relationships, and synchronously links the data layer to achieve rapid association between the hierarchy and multi-source data. 4.The BIM-GIS fusion-based building construction informatization management method according to claim 1, characterized in that, The edge nodes collect real-time data from nearby locations, which is then streamed via a lightweight message queue. High-risk monitoring data is prioritized for push, including: Mesh deployment adapts to industrial-grade edge gateways with temperatures ranging from -20℃ to 60℃, covering IoT devices and mobile terminals within 50 meters, and accessing the construction IoT private network via both 5G and wired dual links; IoT devices use Modbus / MQTT interfaces to collect data at 1Hz for high-risk scenarios and 10Hz for normal scenarios; mobile devices upload JSON data via HTTP / HTTPS interfaces, and edge devices parse, normalize, and add metadata. Deploy lightweight message queues of MQTT / ZeroMQ at the edge and in the cloud to build a two-level architecture. Each message is ≤1KB, and the cache limit is 1 hour of high-risk data. If the limit is exceeded, local persistence is triggered. Set three-level data priority, add identifiers to message headers, and automatically mark edges; Configure multiple priority independent queues, adopt preemptive scheduling, with the highest level latency ≤500ms and the lowest level latency ≤1s. If the threshold is exceeded, an alarm will be triggered and a backup link will be switched. Deploy a preprocessing engine at the edge to remove outliers, downsample ordinary data, and proactively push data when special / first-level data exceeds limits or undergoes sudden changes; Special-grade / Level 1 data is configured with dual 5G and wired links, persistent queues, alarms for three failed retransmissions of special-grade data, and data is encrypted with TLS 1.3 and a checksum. Data is distributed in the cloud according to priority. The highest priority triggers model linkage and early warning, the first priority updates the construction status, the ordinary batch is put into the database, the transmission indicators are monitored, and the highest priority automatically optimizes when the latency exceeds the standard. Monitor queue performance, adjust bandwidth allocation when high-priority queues are piling up, generate daily reports, and optimize edge node deployment and scheduling rules. 5.The BIM-GIS fusion-based construction informatization management method according to claim 1, characterized in that, The establishment of a two-way association between BIM-GIS and real-time data, and the automatic location of corresponding spatial points when data changes, includes: By sorting out the correlation dimensions between BIM components, GIS spatial elements and real-time data, and using the spatiotemporal correlation primary key as a unified identification benchmark, a three-dimensional correlation index table of BIM component ID, GIS spatial code and real-time data source ID is established to clarify the BIM component attributes and GIS spatial range corresponding to each type of real-time data, so as to ensure that the correlation relationship is unique and traceable. Define a bidirectional mapping rule base, and configure corresponding mapping rules for IoT sensor data, construction record data, and drone real-world data respectively to achieve bidirectional traceability of data and spatial elements; Deploy a data change monitoring engine to capture data change events in real time and extract the identifiers and content of the changed data by monitoring the data transmission channels of edge nodes, BIM model update interfaces, and GIS spatial feature editing interfaces. When real-time data changes, the data change monitoring engine triggers a related index query, retrieves the three-dimensional related index table based on the source ID of the changed data, matches the corresponding BIM component ID and GIS spatial code, and obtains the component coordinates, GIS geographic coordinates and spatial hierarchy information. The spatial positioning interface of the BIM-GIS fusion platform is called to highlight the model components and display the attributes in the pop-up window based on the matched BIM component ID. At the same time, the data change location is marked on the map layer according to the GIS spatial code to complete the forward positioning of real-time data to spatial points. Configure the data reverse association rules triggered by spatial elements. When a user clicks on a BIM component or GIS spatial point, the platform extracts the corresponding identifier and performs a reverse query on the associated real-time data, pulls and displays the latest data details, and completes the reverse association from spatial point to real-time data. Establish a dynamic adaptation mechanism for association rules. For new real-time data types, BIM model updates, and GIS spatial adjustments, automatically update the 3D association index table and mapping rule library, and support manual supplementation and correction of association relationships. Embedded spatial consistency verification logic compares the deviation between the GIS geographic coordinates associated with real-time data and the global geodetic coordinates of BIM components. When the deviation exceeds a preset threshold, a coordinate calibration prompt is triggered and the location information is corrected. Deploy the three-dimensional association index table in a distributed cache to create a hot data cache pool for frequently changing real-time data, optimize the performance of association queries, and control the spatial positioning delay after data changes; Establish an exception handling mechanism to log and trigger alarms for exceptions such as missing indexes and invalid associations, prompting administrators to supplement the association information. 6.The BIM-GIS fusion-based building construction informatization management method according to claim 5, characterized in that, The established dynamic adaptation mechanism for association rules automatically updates the 3D association index table and mapping rule base for newly added real-time data types, BIM model updates, and GIS spatial adjustments. It also supports manual supplementation and correction of association relationships, including: Build multi-dimensional change monitoring triggers to listen for new protocol registration, BIM model updates, and GIS spatial adjustment events, capture parameters in real time, and trigger the adaptation process. Establish a mapping rule template library categorized by data type, develop an automatic adaptation engine, execute adaptation logic for different change scenarios, and synchronously adopt a transactional update mechanism to ensure data consistency; A visual manual maintenance interface is built, providing functions for adding, deleting, modifying, and querying index tables and mapping rules, batch importing and exporting, and searching and filtering. It also supports manually associating key identifiers. Embedded association validity verification logic automatically adapts or is manually modified to verify the validity of component ID, spatial code, data source ID and spatial coordinate deviation, preventing invalid associations from being written. Establish a change log traceability system to record all operation details and support conditional filtering and historical version rollback; Define adaptation priority rules, execute multiple changes in a preset order when they occur concurrently, enable incremental adaptation in high-frequency scenarios, and develop an abnormal adaptation handling mechanism to push alarms and adaptation suggestions. To optimize adaptation performance, frequently accessed data is cached on distributed nodes, and asynchronous parallel processing is adopted for batch change scenarios. 7.The BIM-GIS fusion-based construction informatization management method according to claim 5, characterized in that, The embedded spatial consistency verification logic compares the deviation between the GIS geographic coordinates associated with real-time data and the global geodetic coordinates of the BIM components. When the deviation exceeds a preset threshold, it triggers a coordinate calibration prompt and corrects the location information, including: The National Geodetic Coordinate System is clearly defined as the sole verification benchmark. Multi-dimensional coordinate deviation thresholds are preset, and high-risk points can be customized and adjusted. Threshold parameters can be flexibly configured through the system interface. A coordinate extraction and transformation submodule was developed. During verification, GIS geographic coordinates and BIM local coordinates are automatically extracted and then unified into global geodetic coordinates through a two-way conversion and adaptation module to ensure consistent coordinate reference. Design the logic for calculating three-dimensional spatial deviations, complete the calculation of plane, elevation and comprehensive deviations, and ensure millisecond-level response by executing efficiently in memory. The system configures three types of verification mechanisms: data change trigger, scheduled verification, and manual trigger. When triggered, the system automatically marks the unique identifier of the verification object. Based on the comprehensive spatial deviation value, it is divided into four levels: compliant, minor, moderate and severe, and corresponding graded processing rules are implemented: no action, pop-up prompt, automatic correction, and alarm suspension of data association. When there is a moderate deviation, the source is traced and the automatic correction is carried out accordingly, the coordinate information of the 3D association index table is updated, and the correction details are recorded synchronously. Establish a multi-dimensional alert system, differentiate the marking of points and display of information according to the deviation level, and push alarms through multiple channels for severe deviations; The design includes a verification process for the correction results, automatic correction followed by re-inspection, manual correction followed by manual confirmation, and the system retains complete verification records. Establish a verification and correction log system to record core information and support multi-condition filtering and querying; Optimize verification performance by caching high-frequency coordinates and batch asynchronous processing, and regularly analyze logs and optimize coordinate transformation parameters. 8.The BIM-GIS fusion-based building construction informatization management method according to claim 1, characterized in that, The cross-validated multi-source data identifies anomalies through spatial matching and logical verification, automatically removes anomalies, and triggers supplementary data collection, including: Define the verification scope of BIM models, GIS geospatial data, IoT monitoring data, construction business records, and UAV real-scene modeling data. Use the project's global geodetic coordinate system as a unified spatial reference and develop a quantifiable verification benchmark library that includes equipment range, construction specifications, spatiotemporal correlation, and data integrity. A multi-source data association map is constructed based on the spatiotemporal association primary key, which binds multiple types of data at the same spatial location and the same time node to form a spatiotemporal association network that can be synchronously cross-compared. Design spatial matching and multi-dimensional logical verification rules, compare spatial position deviation and attribution rationality after unifying coordinates, and complete numerical, time, association and integrity verification according to data type. Build a cross-validation engine that first performs spatial matching and then logical verification, executes the verification sequentially, marks the anomaly type, and records the anomaly data identifier, deviation value, and conflict details; Abnormal data is isolated and automatically removed, classified into mild, moderate and severe levels according to the degree of abnormality, triggering corresponding level supplementary data collection rules, and issuing instructions to edge nodes to complete data supplementary collection; The supplementary data is re-verified. If it passes, it is included in the normal data link. If it fails, it is upgraded to an anomaly and manual intervention is triggered. The system retains the processing results. Establish an anomaly tracing and rule optimization mechanism, analyze the causes of anomalies and iteratively verify the rules, optimize the verification performance by combining caching, asynchronous parallelism and full sampling, and record the verification, anomaly, supplementary sampling and processing logs throughout the process. 9.The BIM-GIS fusion-based construction informatization management method according to claim 1, wherein, The adaptive adjustment of model display accuracy, overlaying real-time data visualization indicators, and the identification of risks, model highlighting, and early warning push when thresholds are exceeded include: The model display accuracy is divided into three levels: low, medium and high, according to the construction scenario and terminal equipment type. The display rules of BIM model and GIS elements under each accuracy are defined. At the same time, adaptive triggering conditions such as view zoom, interactive operation and data update frequency are set to form an accuracy adaptation rule base. A model accuracy adaptive rendering engine is built and integrated into the visualization kernel of the BIM-GIS fusion platform. It monitors the performance of terminal devices, view parameters and data interaction requirements in real time, dynamically adjusts the model loading strategy according to the accuracy adaptation rule base, and preloads medium / high accuracy model caches for high-frequency interest areas. Construct a real-time data visualization labeling system categorized by data type and risk level, bind spatiotemporal association primary keys to the labels and specify anchor coordinates, and configure label floating interaction rules; Based on the spatiotemporal correlation primary key, the visual identifiers are accurately overlaid onto the corresponding BIM components / GIS spatial points, and a batch rendering module for identifiers is developed to refresh the identifier styles and values ​​according to the data update frequency. A dynamic threshold determination engine is built, which is connected to the threshold benchmark library composed of construction safety specifications and equipment range. The data values ​​are compared with the corresponding thresholds in real time, and the risk levels are divided into four levels according to the degree of deviation: normal, slightly exceeding the threshold, severely exceeding the threshold, and extreme exceeding the limit. The data is labeled and associated with the corresponding BIM component ID / GIS spatial code. When the data exceeds the threshold, the corresponding BIM component / GIS spatial area is matched by the spatiotemporal primary key, the spatial positioning interface is called to jump the model view to the risk point and adapt to high-precision display, and at the same time, the component / area is highlighted according to the risk level. Design a tiered early warning push mechanism, configure push channels and corresponding content such as platform pop-ups, mobile APP / WeChat Work, SMS / voice calls according to risk level, and record the push time, recipient, and read status of all early warning information to form a closed loop record; The system employs GPU batch rendering, progressive model loading, and anti-shake mechanisms to optimize visualization and early warning performance. It also supports users in customizing identifier styles, threshold judgment deviation ratios, early warning push channels, and triggering conditions on the platform interface. Establish a dynamic adaptation mechanism for display and early warning, monitor changes in the performance of terminal equipment during the construction phase in real time, automatically update the accuracy adaptation rules and threshold benchmark library, regularly statistically analyze the frequency of early warnings and the number of accuracy switching, and iteratively optimize the accuracy rules and threshold parameters.

10. A building construction information management system based on BIM-GIS fusion, characterized in that, include: processor; A machine-readable storage medium for storing machine-executable instructions of the processor; The processor is configured to execute the BIM-GIS integrated building construction information management method according to any one of claims 1 to 9 by executing the machine-executable instructions.

Images