Construction dynamic twinning method based on Internet of Things and BIM

Abstract

The invention provides a construction dynamic twinning method based on the Internet of Things and BIM, and relates to the technical field of digital twinning, and the method comprises the steps: obtaining heterogeneous data streams collected by a sensor, and carrying out the analysis and classification of the data streams, and generating a standardized data set; performing semantic mapping on the standardized data set according to association rules of BIM component identifiers and data attributes to generate a component association data set; decrypting and decompressing the encrypted data packet at the cloud, analyzing the construction progress time sequence data by adopting a distributed computing framework, and generating a deviation detection result; dynamically adjusting a routing rule of a data link according to a deviation detection result, and redistributing a processing priority to optimize link configuration; updating a BIM component state based on the optimized link configuration, and generating a real-time updated digital twin model by adopting cloud computing resources; three-dimensional visual data are extracted from the digital twin model and mapped to a BIM model interface by adopting a rendering technology, so that the monitoring precision and the management efficiency of the construction process are effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of digital twin technology, and in particular to a construction dynamic twin method based on the Internet of Things and BIM. Background Art

[0002] Digital transformation in the construction sector is a key area for improving project management efficiency and quality. Dynamic construction twinning technology, based on the Internet of Things (IoT) and Building Information Modeling (BIM), is crucial as a core means of achieving full digitization. This technology collects real-time construction site data and integrates it with BIM models to create a dynamic mapping between the physical and digital worlds, providing precise decision-making support for construction management.

[0003] However, existing methods have significant limitations in practical applications and are unable to meet the needs of complex construction scenarios. Currently, most solutions rely on a single data source or simple integration methods, which are unable to cope with the complexity of multi-source heterogeneous data on the construction site. Traditional methods have shortcomings in data standardization, real-time synchronization, and dynamic adjustment, resulting in a disconnect between the digital twin model and the actual construction status. In addition, the computing and storage capabilities of existing systems are limited when processing massive amounts of data, making it difficult to achieve efficient data analysis and visualization. These shortcomings limit the in-depth application of dynamic twin technology in construction management.

[0004] The core challenges of integrating the IoT and BIM lie in the following technical factors: First, construction sites have a wide variety of sensor devices with varying data formats. Standardizing the processing of heterogeneous data and accurately mapping it to BIM model components is a fundamental challenge in building dynamic twin models. Second, the complex network environment on construction sites makes it difficult to ensure the real-time, reliability, and security of data transmission, directly impacting the dynamic updating of twin models. Third, the dynamic changes in construction progress and processes require adaptive data links, which existing technologies struggle to achieve with flexible optimization.

[0005] Therefore, how to establish a unified data exchange protocol to achieve standardization and mapping of heterogeneous data, how to optimize the data transmission mechanism to ensure real-time and reliability, and how to design adaptive data links to cope with dynamic changes in construction have become key issues that need to be urgently addressed in the construction dynamic twin method based on the Internet of Things and BIM. Summary of the Invention

[0006] The present invention provides a construction dynamic twinning method based on the Internet of Things and BIM, which mainly includes: Obtaining heterogeneous data streams collected by sensors, parsing and classifying the heterogeneous data streams using a preset format template, and generating a standardized data set; Performing semantic mapping on the standardized data set according to association rules between BIM component identifiers and data attributes to generate a component association data set; If the matching accuracy of the component association data set is lower than a preset threshold, the historical deviation data is used to train the machine learning model, optimize the association rule parameters and generate an optimized component association data set; Extracting real-time data streams from the optimized component-associated data set, compressing and encrypting the real-time data streams using edge computing nodes to generate encrypted data packets; Send the encrypted data packet to the cloud server through the main transmission channel, and switch to the backup transmission channel if the network jitter exceeds the preset threshold; Decrypting and decompressing the encrypted data packet in the cloud, analyzing the construction progress time series data using a distributed computing framework, and generating deviation detection results; Dynamically adjust data link routing rules based on the deviation detection results and reallocate processing priorities to optimize link configuration; Update BIM component status based on optimized link configurations, and use cloud computing resources to generate a real-time updated digital twin model; Three-dimensional visualization data is extracted from the digital twin model, and rendering technology is used to map the sensor data to the BIM model interface.

[0007] The technical solution provided by the embodiment of the present invention may have the following beneficial effects: The present invention discloses an intelligent construction management method based on the Internet of Things and BIM. By collecting heterogeneous sensor data streams and performing standardized processing, semantic association mapping with BIM components is achieved. Edge computing nodes are used to compress and encrypt real-time data for transmission. A distributed computing framework is used in the cloud to analyze construction progress time series data, dynamically adjust data link configurations, update BIM component status and generate digital twin models, and finally visualize sensor data on the BIM interface. This method integrates technologies such as the Internet of Things, BIM, edge computing, and cloud computing to achieve real-time collection, intelligent analysis, and dynamic management of construction site data, effectively improving the monitoring accuracy and management efficiency of the construction process, and providing technical support for the construction of smart construction sites. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] For better understanding and implementation, the technical solution of the present application is described in detail below with reference to the accompanying drawings.

[0009] Figure 1 This is a flow chart of a construction dynamic twinning method based on the Internet of Things and BIM of the present invention; Figure 2 Schematic diagram of a construction dynamic twinning method based on the Internet of Things and BIM of the present invention; Figure 3 This is another schematic diagram of the construction dynamic twin method based on the Internet of Things and BIM of the present invention. DETAILED DESCRIPTION

[0010] To further illustrate the technical means and effectiveness of the present invention in achieving its intended purpose, exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present application. Rather, they are merely examples of methods and systems consistent with certain aspects of the present application, as detailed in the appended claims.

[0011] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms "a," "an," "the," and "the" used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to any or all possible combinations of one or more of the associated listed items.

[0012] The following describes in detail the specific implementation methods, features and effects of the present invention in conjunction with the accompanying drawings and preferred embodiments.

[0013] like Figure 1-3 In this embodiment, a construction dynamic twinning method based on the Internet of Things and BIM may specifically include: Step S101 : obtaining heterogeneous data streams collected by sensors, parsing and classifying the heterogeneous data streams using a preset format template, and generating a standardized data set.

[0014] Heterogeneous data streams are acquired from sensors, and stream processing technology is used to verify the integrity of the data streams to obtain the initial data stream. The initial data stream is matched with a preset template. If the template matches the data stream characteristics, key fields are extracted to obtain structured data. A decision tree algorithm is used to decompose the structured data. If the characteristics meet the classification criteria, category labels are assigned to obtain classified data. Based on the classified data, data standardization technology is used to map it to a unified format to obtain a standardized dataset. Time series features are extracted from the standardized dataset. If the feature sequence meets the continuity criteria, a dynamic data stream is generated to obtain a dynamic dataset. The dynamic dataset is grouped using the K-means clustering algorithm. If the grouping results meet the similarity threshold, a grouped dataset is generated to obtain the final dataset. The final dataset is encoded using data formatting technology to obtain a formatted dataset.

[0015] Step S102 : performing semantic mapping on the standardized data set according to association rules between BIM component identifiers and data attributes to generate a component association data set.

[0016] The association rules between component identifiers and data attributes are obtained from the standardized dataset. Semantic mapping techniques are used to match component identifiers. If the rules match the attribute characteristics, the associated fields are extracted to obtain a component association dataset. Based on the component association dataset, the component identifiers are grouped using the K-means clustering algorithm. If the grouping results meet a preset similarity threshold, the component categories are determined to obtain a classified dataset. The associated fields of the component identifiers are extracted from the classified dataset and structured using a preset template. If the fields conform to the template format, a structured dataset is generated. Based on the structured dataset, rule verification techniques are used to verify the consistency between component identifiers and data attributes. If the consistency meets the preset conditions, a validation dataset is generated. Semantic features of component associations are extracted from the validation dataset and parsed using feature decomposition techniques. If the features meet the decomposition conditions, a semantic dataset is generated. Based on the semantic dataset, format conversion techniques are used to map the dataset to a unified format. If the format conversion results meet the standard specifications, a formatted dataset is obtained. Time series features of component associations are extracted from the formatted dataset. If the feature sequence meets the continuity condition, a dynamic association dataset is generated.

[0017] Step S103: If the matching accuracy of the component association data set is lower than a preset threshold, the historical deviation data is used to train the machine learning model, optimize the association rule parameters and generate an optimized component association data set.

[0018] A component association dataset is obtained from a repository, and a determination is made as to whether the matching accuracy of the component association dataset is below a preset threshold. If the matching accuracy is below the preset threshold, historical deviation data is obtained from the repository to obtain a deviation dataset. A random forest model is trained based on the deviation dataset, with the model hyperparameters set to default values and the model weights adjusted to obtain an optimized trained model. The component association rules are analyzed using the optimized trained model, and the rule parameters are extracted to obtain a parameter set. A determination is made as to whether the deviation value of the parameter set exceeds a preset range. If the deviation value exceeds the preset range, the parameter set is adjusted using a gradient descent algorithm, with the learning rate set to a preset value and the number of iterations set to a preset number to obtain an adjusted parameter set. The association rules are updated based on the adjusted parameter set to generate a new association rule set. The matching accuracy is recalculated using the new association rule set to obtain an updated matching accuracy. A determination is made as to whether the updated matching accuracy is below the preset threshold. If the updated matching accuracy is below the preset threshold, secondary deviation features are extracted from the deviation data and the random forest model is repeatedly trained to obtain a further optimized trained model.

[0019] Step S104: extracting a real-time data stream from the optimized component-related data set, compressing and encrypting the data stream using an edge computing node, and generating an encrypted data packet.

[0020] A structured real-time data stream is obtained from the optimized component-associated dataset. The structured real-time data stream generates a continuous data stream according to pre-established field filtering rules. The continuous data stream is received using the Apache Flink framework on the edge computing node, and streaming compression is performed using the LZ4 compression level to obtain a compressed data stream. If the volume of the compressed data stream exceeds a preset threshold, the compressed data stream is segmented according to a fixed size to obtain a set of ordered data segments. Each data segment in the set of ordered data segments is encrypted using the AES-256 algorithm and a pre-distributed key to obtain a set of encrypted data segments. The encrypted data segment set is packaged by sequence number at the edge computing node, and a metadata header containing the data length and number of segments is added to obtain a complete encrypted data packet. The SHA-256 algorithm is used to calculate the hash value of the metadata header of the complete encrypted data packet. If the hash value matches the pre-stored checksum, the complete encrypted data packet is sent to the authenticated target node address via the MQTT protocol.

[0021] It's important to note that the Apache Flink framework is an open-source stream processing framework designed for distributed data stream and batch processing. It supports high-throughput, low-latency real-time stream processing, as well as batch processing. Flink provides powerful streaming computing capabilities, making it suitable for scenarios such as large-scale data analysis, machine learning, and real-time monitoring.

[0022] LZ4 is a fast compression algorithm, especially suitable for streaming data compression. It allows users to set the compression level, which can trade off between compression speed and compression ratio.

[0023] The AES-256 algorithm and the SHA-256 algorithm are both commonly used encryption and hashing algorithms. The AES-256 algorithm is used to encrypt data and protect its confidentiality, using symmetric key encryption and decryption. The SHA-256 algorithm is used to generate hash values for data, ensuring its integrity and verifying its immutability.

[0024] Step S105 , sending the encrypted data packet to the cloud server via the primary transmission channel. If it is detected that the network jitter exceeds a preset threshold, switching to the backup transmission channel.

[0025] An encrypted data packet is sent to the cloud server via the primary transmission channel to obtain the real-time network jitter value. The real-time network jitter value is determined to determine whether it exceeds a preset threshold, thereby obtaining the network status of the primary transmission channel. If the real-time network jitter value exceeds the preset threshold, the channel switching mechanism is activated, switching from the primary transmission channel to the backup transmission channel to confirm the completion of the switchover. An encrypted data packet is sent via the backup transmission channel to obtain the network jitter value of the backup transmission channel, determine the transmission stability, and obtain the status of the backup transmission channel. The sliding window function of the pandas library is used to analyze the network jitter value sequence of the backup transmission channel to determine whether the network jitter value sequence has returned to below the preset threshold. If the primary transmission channel is recoverable, the backup transmission channel is switched back to the primary transmission channel. The transmission performance of the primary transmission channel after the switchover is obtained, the transmission stability is determined, and the status of the primary transmission channel is obtained. The encrypted data packet from the primary or backup transmission channel is used to obtain the receipt confirmation information from the cloud server, determine the data integrity, and obtain the transmission result. Based on the transmission result, the Log4j logging tool is used to store the transmission status and channel switching records to determine business continuity and obtain the system operation status.

[0026] It's important to note that Pandas is a core Python library for data processing, analysis, and cleaning. It provides efficient and flexible data structures (such as Series and DataFrame), making data manipulation simple and fast. Log4j, on the other hand, is a logging tool that offers flexible configuration options and multiple log output methods, helping developers efficiently record, manage, and analyze log information. It also supports the recording and management of events such as transmission status and channel switching.

[0027] Step S106: decrypt and decompress the encrypted data packet in the cloud, use a distributed computing framework to analyze the construction progress time series data, and generate a deviation detection result.

[0028] Obtain an encrypted data packet from the cloud, encrypted using the AES algorithm. Decrypt the encrypted data packet using a preset key to obtain a decrypted data packet. Process the decrypted data packet using the ZIP decompression algorithm to generate construction progress time series data. Use the Apache Spark distributed computing framework to analyze the construction progress time series data, extract construction progress features, and obtain a progress feature set. If the deviation between the progress feature set and the preset schedule exceeds a preset threshold, it is marked as an anomaly, resulting in a deviation marker. Based on the deviation markers, use the K-Means clustering algorithm to group the anomaly data and generate an anomaly grouping result. By analyzing the anomaly grouping results, determine the progress adjustment direction and generate an adjustment plan. Based on the adjustment plan, use Matplotlib to generate a visual time series diagram to obtain the progress monitoring output.

[0029] It's important to note that Apache Spark is an open-source big data processing framework designed for fast, general-purpose data processing tasks. It supports distributed computing, can handle large datasets, and offers efficient performance. Matplotlib, on the other hand, is a widely used Python plotting library that provides an easy-to-use way to generate a variety of static, dynamic, and interactive charts.

[0030] Step S107 : dynamically adjusting the routing rules of the data link according to the deviation detection result, and reallocating the processing priority to optimize the link configuration.

[0031] Real-time network probe data is obtained, including a link traffic matrix. Three eigenvalues, namely bytes per second, packet loss rate, and latency, are extracted. The eigenvalues are Z-score normalized to obtain a standardized feature set. The standardized feature set is input into a K-Means clusterer with a preset K value of 2 to obtain an abnormal traffic marking result. If the abnormal traffic proportion in the abnormal traffic marking result exceeds a preset threshold of 5%, a draft routing rule is generated based on the historical routing table. The draft routing rule is input into the routing policy module of FRRouting to obtain a routing table update plan containing the next-hop address and weight values. Based on the weight values in the routing table update plan, server weight parameters are set on the HAProxy load balancer, and traffic control queues of corresponding priorities are created using the Linux TC module. Real-time monitoring data from HAProxy is obtained, and the load variance value of each link is calculated. If the load variance value exceeds the preset threshold of 0.2, the server weight parameters are recalculated using the minimum connection number algorithm. The recalculated weight parameters are written to the HAProxy configuration file, and a hot update operation is performed to obtain an updated link configuration snapshot. The link status data of the updated link configuration snapshot is continuously collected, the traffic feature value is recalculated and input into the K-Means clusterer to obtain a new abnormal traffic marking result.

[0032] It's important to note that FRRouting (FRR) is a powerful routing protocol implementation suite that allows users to manage and configure complex routing policies in Linux systems. The routing policy module provides network administrators with a flexible way to control routing decisions, filtering, and path selection, suitable for a variety of routing protocol environments.

[0033] HAProxy is an efficient load balancer used to distribute traffic to multiple backend servers. You can set weights for each server to control the traffic distribution ratio. Linux TC is a powerful traffic control tool that can be used to classify network traffic, allocate bandwidth, and manage priorities, thereby ensuring reasonable utilization and efficient distribution of traffic.

[0034] Step S108: Update the BIM component status based on the optimized link configuration, and use cloud computing resources to generate a real-time updated digital twin model.

[0035] A parameter set is obtained from the link configuration, which includes link delay and computing task allocation information. The parameter set is parsed to determine the status update rules of the BIM components. The status update rules include computing priority adjustment conditions. If the link delay in the parameter set exceeds the preset threshold, the computing tasks are reallocated through cloud computing resources to obtain an adjusted resource scheduling plan. Based on the resource scheduling plan, the computing priority of the BIM components is dynamically adjusted to generate updated component status data. Synchronization operations are performed on the updated component status data through cloud computing resources to obtain real-time updated data synchronization results. The data synchronization results are mapped to the virtual model to generate a preliminary digital twin model. If the update frequency of the preliminary digital twin model is lower than the preset threshold, the model update frequency is clustered and analyzed using the K-means algorithm to obtain optimized model generation parameters. Based on the optimized model generation parameters, the rendering frequency of the digital twin model is adjusted to generate the final real-time updated digital twin model.

[0036] Step S109: extracting three-dimensional visualization data from the digital twin model and mapping the sensor data to the BIM model interface using rendering technology.

[0037] Specifically, three-dimensional visualization data is obtained from the digital twin model, and the data structure is parsed using a JSON parsing tool to determine the correspondence between data points and the BIM model, thereby obtaining mapping parameters for the three-dimensional visualization data. Based on the mapping parameters, the sensor data is matched with the geometric nodes of the BIM model using WebGL to generate preliminary visualization interface data. If the refresh frequency of the preliminary visualization interface data is lower than a preset threshold, the refresh frequency is optimized using an ARIMA model to obtain an adjusted refresh frequency parameter. Based on the adjusted refresh frequency parameter, the visualization interface data is dynamically updated using WebSocket to generate real-time synchronized interface display data. The real-time synchronized interface display data is distributedly stored using HDFS to obtain distributed stored interface data. Keyframe data is extracted from the distributed stored interface data, and the BIM model interface is updated using a Diff algorithm to generate a final dynamically updated visualization interface. If the data synchronization delay of the final dynamically updated visualization interface exceeds a preset threshold, cloud computing resources are reallocated using the Ketama algorithm to obtain an optimized resource scheduling solution, and the final dynamically updated visualization interface is updated.

[0038] It should be noted that WebGL (Web Graphics Library) is a powerful web graphics rendering technology that allows developers to efficiently render 3D graphics in the browser. WebGL allows the geometric data in BIM models to be combined with real-time sensor data to achieve dynamic, interactive visualization. This approach not only enhances the real-time monitoring capabilities of BIM models but also provides users with a more intuitive data display experience. WebSocket is a communication protocol used to establish a persistent, full-duplex (bidirectional) communication channel between a client (such as a browser) and a server. It allows two parties to exchange data in real time over an open connection, overcoming the limitations of the traditional HTTP protocol's request-response model. It is particularly suitable for application scenarios with high real-time requirements.

[0039] HDFS (Hadoop Distributed File System) is a core component of the Hadoop ecosystem, used for large-scale data storage. Designed to run on commodity hardware, HDFS can efficiently and fault-tolerantly store and manage large amounts of data, making it particularly well-suited for big data processing. A diff algorithm is used to compare two files or datasets and identify differences between them. Its core purpose is to identify the differences between two versions of text or data.

[0040] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0041] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as a preferred embodiment as above, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to equivalent embodiments using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A construction dynamic twinning method based on the Internet of Things and BIM, characterized by: The method comprises: Obtaining heterogeneous data streams collected by sensors, parsing and classifying the heterogeneous data streams using a preset format template, and generating a standardized data set; Performing semantic mapping on the standardized data set according to association rules between BIM component identifiers and data attributes to generate a component association data set; If the matching accuracy of the component association data set is lower than a preset threshold, the historical deviation data is used to train the machine learning model, optimize the association rule parameters and generate an optimized component association data set; Extracting real-time data streams from the optimized component-associated data set, compressing and encrypting the real-time data streams using edge computing nodes to generate encrypted data packets; Send the encrypted data packet to the cloud server through the main transmission channel, and switch to the backup transmission channel if the network jitter exceeds the preset threshold; Decrypting and decompressing the encrypted data packet in the cloud, analyzing the construction progress time series data using a distributed computing framework, and generating deviation detection results; Dynamically adjust data link routing rules based on the deviation detection results and reallocate processing priorities to optimize link configuration; Update BIM component status based on optimized link configurations, and use cloud computing resources to generate a real-time updated digital twin model; Three-dimensional visualization data is extracted from the digital twin model, and rendering technology is used to map the sensor data to the BIM model interface.

2. The method according to claim 1, characterized in that The obtaining of heterogeneous data streams collected by sensors, parsing and classifying the heterogeneous data streams using a preset format template, and generating a standardized data set includes: Obtain heterogeneous data streams from sensors, use stream processing technology to verify the integrity of the data streams, and obtain the initial data streams; Matching the initial data stream with a preset template, if the template is consistent with the data stream characteristics, extracting key fields to obtain structured data; A decision tree algorithm is used to perform feature decomposition on the structured data. If the feature meets the classification conditions, a category label is assigned to obtain classified data. According to the classified data, mapping to a unified format using data standardization technology to obtain a standardized data set; Extracting time series features from the standardized data set, and if the time series features meet a continuity condition, generating a dynamic data stream and obtaining a dynamic data set; The dynamic data set is grouped using a K-means clustering algorithm. If the grouping result meets a similarity threshold, a grouped data set is generated to obtain a final data set. The final data set is encoded using a data formatting technology to obtain a formatted data set.

3. The method according to claim 1, characterized in that The step of performing semantic mapping on the standardized data set according to the association rules between BIM component identifiers and data attributes to generate a component-associated data set includes: Obtaining association rules between component identifiers and data attributes from the standardized data set, matching the component identifiers using semantic mapping technology, and extracting associated fields if the rules are consistent with the attribute characteristics to obtain a component association data set; According to the component association data set, the component identifiers are grouped using a K-means clustering algorithm. If the grouping result meets a preset similarity threshold, the component category is determined to obtain a classification data set; Extracting the associated fields of the component identification from the classified data set, performing structured processing on the associated fields using a preset template, and generating a structured data set if the associated fields conform to the template format; Based on the structured data set, the consistency between the component identification and the data attribute is verified using rule verification technology, and if the consistency meets the preset conditions, a verification data set is obtained; Extracting semantic features associated with components from the verification data set, parsing the semantic features using feature decomposition technology, and obtaining a semantic data set if the semantic features meet decomposition conditions; According to the semantic data set, a format conversion technology is used to map the semantic data set to a unified format, and if the format conversion result meets the standard specification, a formatted data set is obtained; A time series feature of component association is extracted from the formatted data set, and a dynamic association data set is generated if the time series feature meets a continuity condition.

4. The method according to claim 1, wherein If the matching accuracy of the component association data set is lower than a preset threshold, the historical deviation data is used to train the machine learning model, optimize the association rule parameters and generate an optimized component association data set, including: Obtaining a component-related data set from a repository, and determining whether a matching accuracy of the component-related data set is lower than a preset threshold; If the matching accuracy is lower than a preset threshold, obtaining historical deviation data from a repository to obtain a deviation data set; Training a random forest model based on the biased dataset, setting the model's hyperparameters to default values, and adjusting the model weights to obtain an optimized training model; Analyzing component association rules using the optimized training model, extracting rule parameters, and obtaining a parameter set; Determining whether the deviation value of the parameter set exceeds a preset range; If the deviation value exceeds the preset range, the parameter set is adjusted by the gradient descent algorithm, the learning rate is set to a preset value, the number of iterations is set to a preset number, and an adjusted parameter set is obtained; updating the association rules according to the adjusted parameter set to generate a new association rule set; Recalculating the matching accuracy using the new association rule set to obtain an updated matching accuracy; Determining whether the updated matching accuracy is lower than a preset threshold; If the updated matching accuracy is lower than a preset threshold, secondary deviation features are extracted from the deviation data, and the random forest model is repeatedly trained to obtain a further optimized training model.

5. The method according to claim 1, wherein The method of extracting a real-time data stream from the optimized component-associated data set, compressing and encrypting the real-time data stream using an edge computing node, and generating an encrypted data packet includes: Acquire a structured real-time data stream from the optimized component association data set, wherein the real-time data stream generates a continuous data stream according to a pre-established field filtering rule; Using a framework on an edge computing node to receive the continuous data stream, and performing streaming compression to obtain a compressed data stream; If the volume of the compressed data stream exceeds a preset threshold, the compressed data stream is segmented into pieces of fixed size to obtain a set of ordered data segments; Encrypting each data segment in the ordered set of data segments using an AES-256 algorithm in conjunction with a pre-distributed key to obtain an encrypted data segment set; The edge computing node packages the encrypted data fragment set according to the sequence number, adds a metadata header containing the data length and the number of fragments, and obtains a complete encrypted data packet; The SHA-256 algorithm is used to calculate the hash value of the metadata header of the complete encrypted data packet. If the hash value matches the pre-stored check code, the complete encrypted data packet is sent to the authenticated target node address.

6. The method according to claim 1, characterized in that The encrypted data packet is sent to the cloud server through the primary transmission channel, and if the network jitter detected exceeds a preset threshold, the backup transmission channel is switched to, including: Sending encrypted data packets to the cloud server through the main transmission channel to obtain a real-time network jitter value, determining whether the real-time network jitter value exceeds a preset threshold, and obtaining the network status of the main transmission channel; If the real-time network jitter value exceeds a preset threshold, a channel switching mechanism is activated to switch from the primary transmission channel to the backup transmission channel, and a switching completion status is determined; Sending an encrypted data packet through the backup transmission channel, obtaining a network jitter value of the backup transmission channel, judging transmission stability, and obtaining a status of the backup transmission channel; The sliding window function of the pandas library is used to analyze the network jitter value sequence of the backup transmission channel to determine whether the network jitter value sequence has recovered to below a preset threshold; If the main transmission channel is restored and feasible, switch back to the main transmission channel from the backup transmission channel, obtain the transmission performance of the main transmission channel after switching, determine the transmission stability, and obtain the status of the main transmission channel; Obtain the receipt confirmation information from the cloud server through the encrypted data packet of the main transmission channel or the backup transmission channel, determine the data integrity, and obtain the transmission result; According to the transmission results, a log recording tool is used to store the transmission status and channel switching records, determine business continuity, and obtain the system operation status.

7. The method according to claim 1, characterized in that Decrypting and decompressing the encrypted data packet in the cloud, analyzing the construction progress time series data using a distributed computing framework, and generating deviation detection results include: Obtain an encrypted data packet from the cloud, where the encrypted data packet is encrypted using the AES algorithm; Decrypting the encrypted data packet using a preset key to obtain a decrypted data packet; Processing the decrypted data packet through a ZIP decompression algorithm to generate construction progress time series data; Analyzing the construction progress time series data and extracting construction progress features to obtain a progress feature set; If the deviation between the progress feature set and the preset progress plan exceeds a preset threshold, it is marked as abnormal and a deviation mark is obtained; According to the deviation mark, the abnormal data is grouped using the K-Means clustering algorithm to generate an abnormal grouping result; Determine the progress adjustment direction by analyzing the abnormal grouping results and generate an adjustment plan; A visual timing diagram is generated for the adjustment plan to obtain a progress monitoring output.

8. The method according to claim 1, characterized in that The method of dynamically adjusting the routing rules of the data link and reallocating processing priorities to optimize the link configuration according to the deviation detection result includes: Acquire real-time network probe data and extract eigenvalues; the real-time network probe data includes a link traffic matrix; the eigenvalues include bytes per second, packet loss rate, and delay; Performing Z-score normalization on the feature values to obtain a normalized feature set; Input the standardized feature set into a K-Means clusterer with a preset K value of 2 to obtain an abnormal traffic marking result; If the abnormal traffic proportion in the abnormal traffic marking result exceeds a preset threshold of 5%, a routing rule draft is generated according to the historical routing table; Inputting the routing rule draft into a routing policy module to obtain a routing table update plan including a next hop address and a weight value; According to the weight values in the routing table update scheme, setting server weight parameters on the load balancer and creating flow control queues of corresponding priorities; Obtain real-time monitoring data and calculate the load variance value of each link; If the load variance value exceeds a preset threshold, the server weight parameter is recalculated using the minimum connection number algorithm; Write the recalculated server weight parameters into the configuration file and perform a hot update operation to obtain the updated link configuration snapshot; The link status data of the updated link configuration snapshot is continuously collected, the traffic feature value is recalculated and input into the K-Means clusterer to obtain a new abnormal traffic marking result.

9. The method according to claim 1, characterized in that The method of updating the BIM component status based on the optimized link configuration and generating a real-time updated digital twin model using cloud computing resources includes: Acquire a parameter set from a link configuration, the parameter set including link delay and computing task allocation information; Parsing the parameter set to determine a status update rule for the BIM component, the status update rule including a calculation priority adjustment condition; If the link delay in the parameter set exceeds a preset threshold, computing tasks are reallocated through cloud computing resources to obtain an adjusted resource scheduling solution; Dynamically adjust the calculation priority of BIM components according to the adjusted resource scheduling plan to generate updated component status data; Performing synchronization operations on the updated component status data through cloud computing resources to obtain real-time updated data synchronization results; Mapping the real-time updated data synchronization results to the virtual model to generate a preliminary digital twin model; If the update frequency of the preliminary digital twin model is lower than a preset threshold, cluster analysis is performed on the update frequency of the model using the K-means algorithm to obtain optimized model generation parameters; According to the optimized model generation parameters, the rendering frequency of the digital twin model is adjusted to generate a final real-time updated digital twin model.

10. The method according to claim 1, characterized in that Extracting three-dimensional visualization data from the digital twin model and mapping the sensor data to the BIM model interface using rendering technology includes: Obtain 3D visualization data from the digital twin model, analyze the data structure, determine the correspondence between data points and the BIM model, and obtain the mapping parameters of the 3D visualization data; According to the mapping parameters, the sensor data is matched with the geometric nodes of the BIM model to generate preliminary visualization interface data; If the refresh frequency of the preliminary visualization interface data is lower than a preset threshold, the refresh frequency is optimized by an ARIMA model to obtain an adjusted refresh frequency parameter; Dynamically updating the visual interface data according to the adjusted refresh frequency parameters to generate real-time synchronized interface display data; Distributed storage is performed on the real-time synchronized interface display data to obtain distributed stored interface data; Extract key frame data from the distributed stored interface data, use a Diff algorithm to update the BIM model interface, and generate a final dynamically updated visual interface; If the data synchronization delay of the final dynamically updated visual interface exceeds a preset threshold, the cloud computing resources are reallocated using the Ketama algorithm to obtain an optimized resource scheduling solution, and the final dynamically updated visual interface is updated.

Images