A construction project archive digitization and information management system and method

Abstract

This invention discloses a construction project archive digitization and information management system and method, comprising: an engineering data integration layer, an engineering data management layer, an engineering data application layer, and an engineering data display layer. The engineering data integration layer acquires engineering data, implements cloud computing, and uploads it to the engineering data management layer. The engineering data management layer is used for data extraction, transformation, storage, and the generation of several digitized engineering archives. The engineering data application layer includes several management models, which are used to associate several digitized engineering archives and apply the associated digitized engineering archives to construction project management operations. The engineering data display layer includes a web interface and web controls, whereby the web controls are used to call the associated digitized engineering archives to perform construction project management operations and display the operation results on the web interface. This invention achieves the goal of improving the accuracy of digitized engineering archives and facilitating maintenance and control.

Description

Technical Field

[0001] This invention relates to the field of engineering information digitization technology, specifically to a construction project archive digitization and information management system and method. Background Technology

[0002] Engineering record management is a crucial component of engineering construction projects. Effective engineering record management provides scientific and reasonable reference materials and a sound basis for construction and planning. Engineering records serve a series of functions in planning, management, and construction within a construction project. Images, audio-visual materials, texts, and some photographic materials play a significant role in engineering record management; these are engineering records directly or indirectly generated during the project management process. Traditional engineering records are paper-based, requiring substantial storage space, making record management extremely demanding and labor-intensive. Paper-based records have high storage requirements because fires can cause significant losses. Traditional paper-based record management requires significant human and material resources for sorting and classifying records, and the large volume of paper records leads to difficulties in retrieval and preservation.

[0003] In today's rapidly developing modern engineering industry, traditional paper-based archiving and management methods are no longer sufficient to meet the demands of the ever-increasing volume of modern projects. Therefore, applying modern digital information technology to engineering archive management is an urgent and crucial step in the reform of modern engineering archive management. The digital construction and management of engineering archives can electronically archive and preserve traditional paper documents, making the creation and recording of archives more efficient and faster, and preventing damage or loss of documents.

[0004] Although the digitization of construction project archives has solved some of the technical problems existing in traditional paper archives, current digitization and information management technologies for construction project archives still have the following drawbacks:

[0005] (1) As the amount of engineering data connected to the public platform continues to increase, it is difficult to achieve efficient and accurate data integration, management and monitoring, and the accuracy of the final digital engineering archives is insufficient.

[0006] (2) Existing methods for digitizing construction project archives are difficult to maintain due to the susceptibility of errors in associated data.

[0007] (3) Existing methods for digitizing construction project archives may result in unsmooth view loading when a large amount of data is accessed, leading to inaccurate view control. Summary of the Invention

[0008] To overcome the shortcomings of existing technologies, this invention provides a construction project archive digitization and information management system and method, which solves the technical problems of insufficient accuracy, high maintenance difficulty, unsmooth view loading, and inaccurate control of digitized construction project archives generated by existing digitization methods, thereby achieving the goal of improving the accuracy of digitized construction project archives and making them easier to maintain and control.

[0009] To solve the above problems, the technical solution adopted by the present invention is as follows:

[0010] A construction project archive digitization and information management system, comprising:

[0011] The engineering data integration layer is used to acquire engineering data, implement cloud computing, and upload it to the engineering data management layer.

[0012] The engineering data management layer is used for data extraction, transformation, storage, and the generation of several digital engineering archives;

[0013] The engineering data application layer includes several management models, which are used to associate the several digital engineering files and apply the associated digital engineering files to construction project management operations.

[0014] The engineering data display layer includes a web interface and web controls. The web controls are used to call the associated digital engineering files to perform construction project management operations and display the operation results on the web interface.

[0015] In a preferred embodiment of the present invention, the engineering data integration layer includes: an information resource pool and a communication resource pool;

[0016] The information resource pool uses virtualization technology to implement cloud computing on engineering data and obtains cloud computing resources;

[0017] The communication resource pool connects to the public platform through a data interface to obtain engineering data, perform full-link monitoring of the cloud computing resources, and upload the data to the engineering data management layer.

[0018] In a preferred embodiment of the present invention, the engineering data management layer includes a data extraction module, a data management module, and a data analysis module;

[0019] The data extraction module is used to extract and transform data from uploaded cloud computing resources;

[0020] The data management module is used to store the extracted and transformed data;

[0021] The data analysis module is used to analyze the data stored in the data management module and generate several digital engineering archives.

[0022] In a preferred embodiment of the present invention, the information resource pool includes: a metadata extraction module, a semantic annotation module, a virtual resource generation module, and a parallel computing module;

[0023] The metadata extraction module includes: a BERT pre-trained language model, a vector mapping layer, a DeepCAN layer, and a CRF layer;

[0024] The semantic annotation module includes: an NLPI R word segmentation system and association rules;

[0025] The NLP IR word segmentation system is used to clarify the requirements for completeness review in the construction engineering field and to construct an ontology conceptual model for information completeness review.

[0026] The association rules are used to make the information completeness review ontology interconnected, and provide data templates and inspection rules for BIM model information completeness review.

[0027] In a preferred embodiment of the present invention, the communication resource pool includes: a monitoring host and a monitoring database;

[0028] The monitoring host is generally used to obtain the resource data required for system operation, as well as to obtain the real-time transmission location of cloud computing resources.

[0029] The monitoring database is used to filter cloud computing resources multiple times through the entire link environment via the monitoring host, and to transform the freely distributed cloud computing resources that meet the system execution requirements into monitoring information files that meet the system application requirements.

[0030] In a preferred embodiment of the present invention, the data extraction module includes: an engineering data mart architecture;

[0031] The engineering data mart architecture includes: an underlying support network IDC, a distributed computing module, a virtualization module, and ETL tools;

[0032] The underlying support network IDC is used to obtain uploaded cloud computing resources;

[0033] The distributed computing module is used to divide the cloud computing resources into several sub-cloud computing resources;

[0034] The virtualization module includes: a virtualization network, a virtualization server, and virtualization storage, used for virtualization computing of each of the sub-cloud computing resources;

[0035] The ETL tool is used to extract, transform, and load data from each virtualized sub-cloud computing resource.

[0036] In a preferred embodiment of the present invention, the data analysis module includes: a data mining module and an engineering file generation module;

[0037] The data mining module is used to mine the data stored in the data management module, complete the classification and recording of the data, and obtain classified data.

[0038] The engineering archive generation module is used to generate the aforementioned digital engineering archives based on the categorized data.

[0039] As a preferred embodiment of the present invention, the plurality of management models include: project-level BIM model, function-level BIM model, component-level BIM model and part-level BIM model;

[0040] The aforementioned management models are associated with the aforementioned digital engineering archives through an engineering entity directory structure tree.

[0041] In a preferred embodiment of the present invention, when the Web control invokes the associated digital engineering file, it includes:

[0042] The engineering view microkernel manager, which supports reflection, supports the loading, configuration, removal, startup, suspension, and stopping of view components in the several management models, and proxies view objects to complete interoperability functions with controllers and the several management models.

[0043] The engineering view microkernel manager uses the XML-RPC protocol model to send metadata requests and event processing requests to the server-side controller via HTTP / POST.

[0044] A method for digitizing and managing construction project archives includes the following steps:

[0045] Engineering data is acquired through the engineering data integration layer, implemented in cloud computing, and then uploaded to the engineering data management layer.

[0046] The engineering data management layer is used for data extraction, transformation, storage, and the generation of several digital engineering archives.

[0047] Several management models in the engineering data application layer are used to associate the aforementioned digital engineering files, and the associated digital engineering files are applied to construction project management operations.

[0048] Construction project management operations are performed by calling the associated digital project archives through the web control of the project data display layer, and the operation results are displayed on the web interface of the project data display layer.

[0049] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0050] (1) By combining the monitoring host and the monitoring database, this invention can meet the data collection needs of cloud computing resources while realizing the timely acquisition and processing of monitoring information, thereby solving the technical problem that the existing technology has limited dynamic resource coordination capabilities within a unit of time and cannot comprehensively monitor and manage the cloud computing process.

[0051] (2) This invention uses a data mining module to mine data based on a dataset database, thereby achieving accurate classification and recording of data, thus obtaining accurate classification data, and generating more accurate digital engineering archives based on the accurate classification data.

[0052] (3) This invention solves the problem of organic integration of BIM model and digital engineering archive through engineering entity directory structure tree. By maintaining the engineering entity directory structure tree, the digital engineering archive can be automatically associated with the corresponding BIM model, making the associated data less prone to errors. Furthermore, when the digital engineering archive or BIM model is modified later, the accuracy of the association can be guaranteed.

[0053] (4) This invention introduces an engineering view microkernel manager and uses the XML-RPC protocol model to make metadata requests and event processing requests to the server-side controller via HTTP / POST, thereby ensuring both the smoothness and efficiency of view loading and the accuracy of view control.

[0054] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0055] Figure 1 This is the overall architecture diagram of the construction project archive digitization and information management system provided by the present invention;

[0056] Figure 2 This is a flowchart of engineering resource virtualization and cloud computing provided by the present invention;

[0057] Figure 3 This is a technical architecture diagram of the parallel computing module provided by the present invention;

[0058] Figure 4 This is a connection diagram between the monitoring host and the monitoring database provided by the present invention;

[0059] Figure 5 This is a diagram of the engineering data mart architecture provided by the present invention;

[0060] Figure 6 This is a flowchart illustrating the steps of the construction project archive digitization and information management method provided by the present invention.

[0061] The diagram numbers are as follows: 1. Engineering Data Integration Layer; 2. Engineering Data Management Layer; 3. Engineering Data Application Layer; 4. Engineering Data Display Layer; 5. Metadata Extraction Module; 6. Semantic Annotation Module; 7. Virtual Resource Generation Module; 8. Parallel Computing Module; 9. Engineering Information Resource Library; 10. Monitoring Host; 11. Monitoring Database; 12. Underlying Support Network IDC; 13. Distributed Computing Module; 14. Virtualization Module; 15. ETL Tool. Detailed Implementation

[0062] The construction project archive digitization and information management system provided by this invention, such as Figure 1 As shown, it includes: Engineering Data Integration Layer 1, Engineering Data Management Layer 2, Engineering Data Application Layer 3, and Engineering Data Display Layer 4.

[0063] The engineering data integration layer 1 includes an information resource pool and a communication resource pool. The information resource pool implements cloud computing on engineering data through virtualization technology and obtains cloud computing resources. The communication resource pool connects with the public platform through a data interface to obtain engineering data, and performs full-link monitoring of cloud computing resources, and uploads the data to the engineering data management layer 2.

[0064] The engineering data management layer 2 includes a data extraction module, a data management module, and a data analysis module. The data extraction module is used to extract and transform data from uploaded cloud computing resources. The data management module is used to store the extracted and transformed data. The data analysis module is used to analyze the data stored in the data management module and generate several digital engineering archives.

[0065] The engineering data application layer 3 includes several management models, which are used to associate several digital engineering files and apply the associated digital engineering files to construction project management operations.

[0066] The engineering data display layer 4 includes a web interface and web controls. The web controls are used to call the associated digital engineering files to perform construction project management operations and display the operation results on the web interface.

[0067] Specifically, construction project management operations include: reviewing, online interaction, and final acceptance.

[0068] Furthermore, the information resource pool includes: metadata extraction module 5, semantic annotation module 6, virtual resource generation module 7, and parallel computing module 8.

[0069] The metadata extraction module 5 is used to perform sequence labeling on the engineering data obtained from the public platform, abstract the labeled engineering data according to different granularities to form engineering metadata, and save the engineering metadata.

[0070] The semantic annotation module 6 is used to perform ontology-based semantic modeling of engineering metadata, and to perform semantic annotation through the engineering semantic identification system to obtain the annotated metadata. The annotated metadata is then stored in the engineering information resource library 9.

[0071] The virtual resource generation module 7 is used to obtain engineering entity resource data from the engineering information resource library 9, and generate virtual resources from the engineering entity resource data through the virtual resource model;

[0072] Parallel computing module 8 is used to perform parallel computing on virtual resources and store the computing results in cloud storage services.

[0073] Specifically, the virtual resource model includes: attribute entities, image entities, virtual machine specification entities, and network entities.

[0074] Specifically, the above metadata is five-dimensional tuple metadata, which includes: key fields, mapping relationships, identifiers, data names, and data attributes.

[0075] The process for generating virtual resources and cloud computing is as follows: Figure 2 As shown. The technical architecture of the parallel computing module 8 is as follows. Figure 3 As shown.

[0076] Furthermore, the metadata extraction module 5 includes: the BERT pre-trained language model, the vector mapping layer, the DeepCAN layer, and the CRF layer.

[0077] The BERT pre-trained language model is used to obtain the context dynamics of engineering data, generate character embedding representations through context dynamics, and map them to the DeepCAN layer through a vector mapping layer;

[0078] The DeepCAN layer is used to perform parallel computation on the mapped engineering data through multiple convolutional kernels and capture the local continuous features of the entity. At the same time, it uses deep CNN networks to stack and increase the receptive field, and extracts the global contextual high-level semantic features of the engineering data through a multi-head attention mechanism.

[0079] The CRF layer is used to decode the output of the DeepCAN layer, obtain the entity label prediction, and perform sequence labeling on the engineering data.

[0080] Specifically, the DeepCAN layer includes: a non-linear sublayer, a multi-head attention sublayer, residual connections, and layer normalization. The non-linear sublayer is composed of multiple stacked CNNs.

[0081] The metadata extraction module 5 provided by this invention dynamically generates embedded representations of characters based on the engineering data context, providing a better input representation for the DeepCAN layer. It combines convolutional networks and multi-head attention mechanisms to accurately extract global contextual features of engineering data. Finally, it decodes the data through the CRF layer to obtain accurate label predictions, thereby achieving accurate sequence labeling.

[0082] Furthermore, semantic annotation module 6 includes: the NLPI R word segmentation system and association rules.

[0083] The NLPI R word segmentation system is used to clarify the information requirements for completeness review in the construction engineering field, and based on the concepts and relationships of components and attributes, it uses an ontology editing platform to perform semantic modeling and construct an ontology conceptual model for information completeness review.

[0084] Association rules are used to make the information completeness review ontology interconnected, and provide data templates and inspection rules for BIM model information completeness review;

[0085] Among them, the NLP IR word segmentation system is also used to segment and annotate the text of construction engineering design specifications and BIM model standards.

[0086] Specifically, the NLPI R word segmentation system clarifies the attribute information that BIM model standards and construction engineering design specifications should include in the BIM model of a construction project by analyzing and annotating, thereby ensuring the completeness and standardization of the review requirements information.

[0087] Specifically, this invention, through the design of an information resource pool architecture, provides a bottom-up metadata extraction module 5, a semantic annotation module 6, a virtual resource generation module 7, and a parallel computing module 8, thereby realizing the virtualization and parallel computing of engineering data, and further achieving efficient resource sharing, on-demand allocation, unified management, and dynamic scheduling. Among them, the virtual resource generation module 7 allows the hardware resources of the physical server to be shared by multiple virtual servers, and can be uniformly allocated through the virtual resource generation module 7, which greatly improves the utilization rate of server hardware and can effectively reduce the investment in server purchase and infrastructure. The parallel computing module 8 fully decomposes the computing tasks of virtual resources and completes parallel computing, thereby quickly obtaining several computing results for merging and output.

[0088] Furthermore, the communication resource pool includes: monitoring host 10 and monitoring database 11.

[0089] The monitoring host 10 directly acquires resource information related to the monitoring results through the resource data collector at preset intervals, and obtains the resource data required for system operation from the resource information;

[0090] The monitoring host 10 actively requests API from the stored cloud computing resources through the resource data collector based on the configuration of the cloud computing host, and obtains the real-time transmission location of the cloud computing resources based on the consumption of the monitored object data in the communication protocol.

[0091] The monitoring database 11 is used to filter cloud computing resources through the entire link environment multiple times through the monitoring host 10, select cloud computing resources that do not meet the system execution requirements, and return them to the original cloud computing environment through the monitoring host 10. The cloud computing resources that meet the system execution requirements are transformed from a freely distributed form into monitoring information files that meet the system application requirements, and the monitoring information files are temporarily stored.

[0092] The output of monitoring database 11 when transforming cloud computing resources is shown in Formula 1:

[0093]

[0094] In the formula, μ is the directional acquisition coefficient of the resource data collector. The average storage amount of resource information related to monitoring database 11. To monitor the rated resource storage conditions of database 11, Z max This represents the maximum value of cloud computing resource transmission per unit of time.

[0095] Where E represents the performance condition of the communication resource pool, as shown in Formula 2:

[0096]

[0097] In the formula, μ is the directional acquisition coefficient of the resource data collector. This refers to the characteristic value of cloud computing resources collected per unit time. ω represents the average cloud computing resource transmission rate under the end-to-end environment, and ω is the end-to-end time series coefficient in monitoring database 11. This represents the dynamic quantification conditions for cloud computing resources.

[0098] Specifically, the connection relationship between monitoring host 10 and monitoring database 11, such as... Figure 4 As shown.

[0099] This invention combines a monitoring host 10 and a monitoring database 11 to meet the data collection needs of cloud computing resources while enabling timely acquisition and processing of monitoring information. This solves the technical problem that existing technologies have limited dynamic resource coordination capabilities within a unit of time and cannot comprehensively monitor and manage the cloud computing process.

[0100] Furthermore, the data extraction module includes an engineering data mart architecture.

[0101] The engineering data mart architecture includes: underlying support network IDC 12, distributed computing module 13, virtualization module 14, and ETL tools 15;

[0102] The underlying support network IDC 12 is used to obtain uploaded cloud computing resources and support the upper-layer distributed computing module 13, virtualization module 14 and ETL tool 15;

[0103] Distributed computing module 13 is used to divide cloud computing resources into several sub-cloud computing resources;

[0104] The virtualization module 14 includes a virtualization network, a virtualization server, and virtualization storage, and performs virtualization computing on each sub-cloud computing resource through the virtualization network, virtualization server, and virtualization storage.

[0105] ETL tool 15 is used to extract, transform, and load data from each virtualized sub-cloud computing resource.

[0106] Specifically, the data extraction module based on the engineering data mart architecture provided by this invention supports the upper-layer distributed computing module 13, virtualization module 14, and ETL tool 15 through the underlying support network IDC 12, making the data extraction module more balanced, efficient, and stable. Secondly, by combining the distributed computing module 13 with the virtualization module 14, cloud computing resources are divided into several sub-cloud computing resources, and virtualization computing is performed on each sub-cloud computing resource, thereby enabling dynamic deployment and allocation of resources such as servers and storage, achieving cloud computing resource interaction with the most reasonable resource consumption and the shortest time, while rapidly configuring, providing, or releasing cloud computing resources. Finally, the ETL tool 15 is used for more refined data extraction, transformation, and loading.

[0107] Specifically, the engineering data mart architecture, such as Figure 5 As shown.

[0108] Furthermore, the data analysis module includes: a data mining module and an engineering file generation module.

[0109] The data mining module is used to obtain the engineering archive catalog items, and based on the engineering archive catalog items, it mines the data stored in the data management module based on the user information feedback mechanism to complete the data classification and cataloging, and obtain classified data.

[0110] The project file generation module is used to generate several digital project files based on categorized data;

[0111] The data mining module includes: a dataset library; the dataset library includes training datasets, test datasets, and classification datasets;

[0112] The data mining module uses the training and test datasets to construct a binary matrix by obtaining the n feature vectors with the highest weights through a classification algorithm. It then obtains a likelihood matrix based on the binary matrix and uses the likelihood matrix to perform association analysis on the data, determine the data classification, and update the classification dataset.

[0113] Specifically, n takes values ​​from 5 to 7.

[0114] Specifically, this invention uses a data mining module to mine data based on a dataset database, thereby achieving accurate data classification and recording, resulting in accurate classification data, and generating more accurate digital engineering archives based on the accurate classification data.

[0115] Furthermore, several management models include: project-level BIM model, function-level BIM model, component-level BIM model, and part-level BIM model;

[0116] Among them, the project-level BIM model is used to carry the project, sub-projects and project-related information; the functional-level BIM model is used to carry modules with complete functions and spatial information; the component-level BIM model is used to carry components and related information; and the part-level BIM model is used to carry parts belonging to components or products and related information.

[0117] Several management models are linked to several digital engineering archives through an engineering entity directory structure tree;

[0118] The engineering entity directory structure tree consists of several engineering entities, including: project, section, work area, specialty, work point, and components.

[0119] Specifically, this invention solves the problem of organically integrating BIM models and digital engineering archives through an engineering entity directory structure tree. By maintaining the engineering entity directory structure tree, digital engineering archives can be automatically associated with corresponding BIM models, making the associated data less prone to errors. Furthermore, the accuracy of the association can be guaranteed when the digital engineering archives or BIM models are modified later.

[0120] Furthermore, when the Web control invokes the associated digitized project file, it includes:

[0121] The engineering view microkernel manager, which supports reflection, enables the loading, configuration, removal, startup, suspension, and stopping of view components in several management models, and proxies view objects to complete interoperability functions with controllers and several management models.

[0122] The interoperability function exists in the form of metadata. The engineering view microkernel manager proxies interoperability instructions based on the metadata and dynamically generates component objects based on the meta-object protocol according to the component type metadata description.

[0123] The engineering view microkernel manager uses the XML-RPC protocol model to send metadata requests and event handling requests to the server-side controller via HTTP / POST.

[0124] Metadata requests include: the engineering view microkernel manager sends a metadata request to the server-side controller via an XML RPC request; after receiving the metadata request, the server-side controller calls the reflection middleware to reflect the class attributes or procedures, automatically generates MOP-based prototype declarations and all types of metadata related to the remote procedures, and returns the metadata as an XML RPC response.

[0125] The event handling request includes: when a user interacts with a view component object, the view component object sends an event notification to the engineering view microkernel manager. After receiving the event notification, the engineering view microkernel manager sends an event request to the server-side controller using an XML RPC request. The server-side controller calls the reflection middleware to perform metadata analysis on the event request, generates the corresponding procedure call information, calls the corresponding event handling procedure, generates the procedure return value, encapsulates the procedure return value, view control instructions, and call instructions to obtain event metadata, and returns the event metadata to the engineering view microkernel manager as an XML RPC response for parsing, and controls the view based on the parsing results.

[0126] This invention introduces an engineering view microkernel manager and uses the XML-RPC protocol model to send metadata requests and event handling requests to the server-side controller via HTTP / POST, thereby ensuring both the smoothness and efficiency of view loading and the accuracy of view control.

[0127] The construction project archive digitization and information management method provided by this invention, such as Figure 6 As shown, it includes the following steps:

[0128] Step S1: Implement cloud computing on engineering data using virtualization technology through the information resource pool and obtain cloud computing resources; connect with the public platform through the communication resource pool using data interfaces to obtain engineering data, and perform full-link monitoring of cloud computing resources, and upload to the engineering data management layer 2;

[0129] Step S2: Extract and transform the uploaded cloud computing resources using the data extraction module; store the extracted and transformed data using the data management module; analyze the data stored in the data management module using the data analysis module to generate several digital engineering archives.

[0130] Step S3: Associate several digital project files through several management models, and apply the associated digital project files to construction project management operations;

[0131] Step S4: Use the Web control to call the associated digital project archives to perform construction project management operations, and display the operation results on the Web interface;

[0132] The engineering data integration layer 1 includes an information resource pool and a communication resource pool; the engineering data management layer 2 includes a data extraction module, a data management module, and a data analysis module; the engineering data application layer 3 includes several management models; and the engineering data display layer 4 includes a web interface and web controls.

[0133] The above embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of protection of the present invention. Any non-substantial changes and substitutions made by those skilled in the art based on the present invention shall fall within the scope of protection claimed by the present invention.

Claims

1. A construction project archive digitization and information management system, characterized in that, include: The engineering data integration layer is used to acquire engineering data, implement cloud computing, and upload it to the engineering data management layer. The engineering data management layer is used for data extraction, transformation, storage, and the generation of several digital engineering archives; The engineering data application layer includes several management models, which are used to associate the several digital engineering files and apply the associated digital engineering files to construction project management operations. The engineering data display layer includes a web interface and web controls. The web controls are used to call the associated digital engineering archives to perform construction project management operations and display the operation results on the web interface. The engineering data integration layer includes: an information resource pool and a communication resource pool; The information resource pool performs cloud computing on engineering data through metadata extraction, semantic annotation, virtual resource generation, and parallel computing, and obtains cloud computing resources. The information resource pool includes: a metadata extraction module, a semantic annotation module, a virtual resource generation module, and a parallel computing module; The metadata extraction module is used to perform sequence labeling on the engineering data obtained from the public platform, abstract the labeled engineering data according to different granularities to form engineering metadata, and save the engineering metadata. The semantic annotation module is used to perform ontology-based semantic modeling of engineering metadata, and to perform semantic annotation through the engineering semantic identification system to obtain annotated metadata, which is then stored in the engineering information resource library. The virtual resource generation module is used to obtain engineering entity resource data from the engineering information resource database and generate virtual resources from the engineering entity resource data through the virtual resource model; The parallel computing module is used to perform parallel computing on virtual resources and store the computing results in the cloud storage service. The virtual resource model includes: attribute entities, image entities, virtual machine specification entities, and network entities; The project metadata is a five-dimensional tuple metadata, which includes: key fields, mapping relationships, identifiers, data names, and data attributes; The metadata extraction module includes: a BERT pre-trained language model, a vector mapping layer, a DeepCAN layer, and a CRF layer; The BERT pre-trained language model is used to obtain the context dynamics of engineering data, generate character embedding representations through context dynamics, and map them to the DeepCAN layer through a vector mapping layer; The DeepCAN layer is used to perform parallel computation on the mapped engineering data through multiple convolutional kernels and capture the local continuous features of the entity. At the same time, it uses deep CNN networks to stack and increase the receptive field, and extracts the global contextual high-level semantic features of the engineering data through a multi-head attention mechanism. The CRF layer is used to decode the output of the DeepCAN layer, obtain the entity label prediction, and perform sequence labeling on the engineering data; The semantic annotation module includes: an NLPIR word segmentation system and association rules; The NLPIR word segmentation system is used to clarify the requirements for completeness review in the construction engineering field, and based on the concepts and relationships of components and attributes, it uses an ontology editing platform to perform semantic modeling and construct an ontology concept model for information completeness review. The association rules are used to make the information completeness review ontology interconnected, and provide data templates and inspection rules for BIM model information completeness review; The NLPIR word segmentation system is also used to segment and annotate the text of construction engineering design specifications and BIM model standards. The communication resource pool connects to the public platform through a data interface to obtain engineering data, and performs full-link monitoring of the cloud computing resources, and uploads the data to the engineering data management layer. The communication resource pool includes: a monitoring host and a monitoring database; The monitoring host is generally used to obtain the resource data required for system operation, as well as to obtain the real-time transmission location of cloud computing resources. The monitoring database is used to filter cloud computing resources multiple times through the entire link environment via the monitoring host, and to transform the freely distributed cloud computing resources that meet the system execution requirements into monitoring information files that meet the system application requirements. Among them, the output of the monitoring database when transforming cloud computing resources takes into account the average storage volume of resource information related to the monitoring database, the rated storage conditions of the monitoring database, the maximum value of cloud computing resource transmission per unit time, and the performance conditions of the communication resource pool. The management models include: project-level BIM model, function-level BIM model, component-level BIM model and part-level BIM model; The aforementioned management models are associated with the aforementioned digital engineering archives through an engineering entity directory structure tree; The management models undergo information completeness review by the NLPIR word segmentation system and association rules in the semantic annotation module, and the engineering entity directory structure tree consists of several engineering entities; When the web control invokes the associated digital engineering file, it includes: The engineering view microkernel manager, which supports reflection, supports the loading, configuration, removal, startup, suspension, and stopping of view components in the several management models, and proxies view objects to complete interoperability functions with controllers and the several management models. The interoperability function exists in the form of metadata. The engineering view microkernel manager proxies interoperability instructions based on the metadata and dynamically generates component objects based on the meta-object protocol according to the component type metadata description. The engineering view microkernel manager sends metadata and event processing requests to the server-side controller via the XML-RPC protocol model using HTTP / POST.

2. The construction project archive digitization and information management system according to claim 1, characterized in that, The engineering data management layer includes a data extraction module, a data management module, and a data analysis module; The data extraction module is used to extract and transform data from uploaded cloud computing resources; The data management module is used to store the extracted and transformed data; The data analysis module is used to analyze the data stored in the data management module and generate several digital engineering archives.

3. The construction project archive digitization and information management system according to claim 2, characterized in that, The data extraction module includes: an engineering data mart architecture; The engineering data mart architecture includes: an underlying support network IDC, a distributed computing module, a virtualization module, and ETL tools; The underlying support network IDC is used to obtain uploaded cloud computing resources; The distributed computing module is used to divide the cloud computing resources into several sub-cloud computing resources; The virtualization module includes: a virtualization network, a virtualization server, and virtualization storage, used for virtualization computing of each of the sub-cloud computing resources; The ETL tool is used to extract, transform, and load data from each virtualized sub-cloud computing resource.

4. The construction project archive digitization and information management system according to claim 2, characterized in that, The data analysis module includes: a data mining module and an engineering file generation module; The data mining module is used to mine the data stored in the data management module, complete the classification and recording of the data, and obtain classified data. The engineering file generation module is used to generate the aforementioned digital engineering files based on the categorized data.

5. A method for digitizing and managing construction project archives, characterized in that, Includes the following steps: Engineering data is acquired through the engineering data integration layer, implemented in cloud computing, and then uploaded to the engineering data management layer. The engineering data management layer is used for data extraction, transformation, storage, and the generation of several digital engineering archives. Several management models in the engineering data application layer are used to associate the aforementioned digital engineering files, and the associated digital engineering files are applied to construction project management operations. Construction project management operations are performed by calling the associated digital engineering archives through the Web control of the engineering data display layer, and the operation results are displayed on the Web interface of the engineering data display layer. The engineering data integration layer includes: an information resource pool and a communication resource pool; The information resource pool performs cloud computing on engineering data through metadata extraction, semantic annotation, virtual resource generation, and parallel computing, and obtains cloud computing resources. The information resource pool includes: a metadata extraction module, a semantic annotation module, a virtual resource generation module, and a parallel computing module; The metadata extraction module is used to perform sequence labeling on the engineering data obtained from the public platform, abstract the labeled engineering data according to different granularities to form engineering metadata, and save the engineering metadata. The semantic annotation module is used to perform ontology-based semantic modeling of engineering metadata, and to perform semantic annotation through the engineering semantic identification system to obtain annotated metadata, which is then stored in the engineering information resource library. The virtual resource generation module is used to obtain engineering entity resource data from the engineering information resource database and generate virtual resources from the engineering entity resource data through the virtual resource model; The parallel computing module is used to perform parallel computing on virtual resources and store the computing results in the cloud storage service. The virtual resource model includes: attribute entities, image entities, virtual machine specification entities, and network entities; The project metadata is a five-dimensional tuple metadata, which includes: key fields, mapping relationships, identifiers, data names, and data attributes; The metadata extraction module includes: a BERT pre-trained language model, a vector mapping layer, a DeepCAN layer, and a CRF layer; The BERT pre-trained language model is used to obtain the context dynamics of engineering data, generate character embedding representations through context dynamics, and map them to the DeepCAN layer through a vector mapping layer; The DeepCAN layer is used to perform parallel computation on the mapped engineering data through multiple convolutional kernels and capture the local continuous features of the entity. At the same time, it uses deep CNN networks to stack and increase the receptive field, and extracts the global contextual high-level semantic features of the engineering data through a multi-head attention mechanism. The CRF layer is used to decode the output of the DeepCAN layer, obtain the entity label prediction, and perform sequence labeling on the engineering data; The semantic annotation module includes: an NLPIR word segmentation system and association rules; The NLPIR word segmentation system is used to clarify the requirements for completeness review in the construction engineering field, and based on the concepts and relationships of components and attributes, it uses an ontology editing platform to perform semantic modeling and construct an ontology concept model for information completeness review. The association rules are used to make the information completeness review ontology interconnected, and provide data templates and inspection rules for BIM model information completeness review; The NLPIR word segmentation system is also used to segment and annotate the text of construction engineering design specifications and BIM model standards. The communication resource pool connects to the public platform through a data interface to obtain engineering data, and performs full-link monitoring of the cloud computing resources, and uploads the data to the engineering data management layer. The communication resource pool includes: a monitoring host and a monitoring database; The monitoring host is generally used to obtain the resource data required for system operation, as well as to obtain the real-time transmission location of cloud computing resources. The monitoring database is used to filter cloud computing resources multiple times through the entire link environment via the monitoring host, and to transform the freely distributed cloud computing resources that meet the system execution requirements into monitoring information files that meet the system application requirements. Among them, the output of the monitoring database when transforming cloud computing resources takes into account the average storage volume of resource information related to the monitoring database, the rated storage conditions of the monitoring database, the maximum value of cloud computing resource transmission per unit time, and the performance conditions of the communication resource pool. The management models include: project-level BIM model, function-level BIM model, component-level BIM model and part-level BIM model; The aforementioned management models are associated with the aforementioned digital engineering archives through an engineering entity directory structure tree; The management models undergo information completeness review by the NLPIR word segmentation system and association rules in the semantic annotation module, and the engineering entity directory structure tree consists of several engineering entities; When the web control invokes the associated digital engineering file, it includes: The engineering view microkernel manager, which supports reflection, supports the loading, configuration, removal, startup, suspension, and stopping of view components in the several management models, and proxies view objects to complete interoperability functions with controllers and the several management models. The interoperability function exists in the form of metadata. The engineering view microkernel manager proxies interoperability instructions based on the metadata and dynamically generates component objects based on the meta-object protocol according to the component type metadata description. The engineering view microkernel manager sends metadata and event processing requests to the server-side controller via the XML-RPC protocol model using HTTP / POST.

Images