Intelligent construction industry brain construction method, system, equipment and medium

Abstract

The invention discloses an intelligent construction industry brain construction method and system, and relates to the technical field of intelligent construction and information, and the method comprises the steps: collecting the original data of a construction industry, constructing an intelligent data fusion and standardization system, carrying out the multi-source data fusion, and formulating a quality evaluation rule. Based on an intelligent data fusion and standardization system, constructing an industrial information model simulating a human brain neural network by using a deep neural network algorithm and a knowledge graph technology; and according to a demand analysis result of the industry information model, designing an intelligent construction industry brain collaboration platform, and completing construction of the intelligent construction industry brain. According to the method, a reliable data fusion and standardization system is established, the data quality and analyzability are improved, comprehensive modeling and demand analysis of construction industry information are realized, the problems of difficult data integration, insufficient data analysis capability and poor industry collaboration are effectively solved, the intelligent development of the construction industry is promoted, and the overall competitiveness of the industry is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent construction and information technology, and specifically to a method, system, equipment and medium for constructing an intelligent construction industry brain. Background Art

[0002] In the context of the rapid development of the construction industry, traditional construction methods face many challenges in terms of efficiency, quality and safety, and it is urgent to introduce advanced technologies to improve overall efficiency. The intelligent construction industry brain has emerged as the times require. Through technologies such as the Internet of Things, artificial intelligence, big data and cloud computing, a data-driven management and decision-making platform is built to achieve intelligent management of the entire life cycle of construction projects.

[0003] The brain of the smart construction industry monitors the progress, quality and safety of the construction site in real time by deploying sensors, collecting multi-dimensional data such as temperature, humidity and vibration, and using big data technology to store, process and analyze massive data to explore potential patterns. Artificial intelligence technologies such as machine learning and image recognition establish prediction models and decision support systems through learning and analyzing data, and optimize construction plans and resource allocation. Cloud computing provides powerful computing and storage capabilities to support large-scale data processing and real-time computing.

[0004] This intelligent system can provide early warning of potential problems, improve construction efficiency and quality, promote the informatization and digital transformation of the construction industry, and enhance overall competitiveness. By integrating construction industry data resources and using advanced big data, artificial intelligence and other technologies, an industrial brain with functions such as intelligent analysis, decision support and industrial collaboration is established. It aims to promote the development of the construction industry towards intelligence and efficiency and enhance the overall competitiveness of the industry. Summary of the invention

[0005] In view of the above-mentioned problems of high difficulty in data integration, insufficient data analysis capabilities and poor industrial synergy, the present invention is proposed.

[0006] Therefore, the technical problem solved by the present invention is: by constructing an intelligent data fusion and standardization system, massive data are converted into understandable semantic information, and an industrial information model that simulates the human brain neural network is established, and an intelligent construction industry brain collaborative platform is developed to realize intelligent analysis and decision-making support for the construction industry.

[0007] To solve the above technical problems, the present invention provides the following technical solutions: a method for constructing an intelligent construction industry brain, including collecting original data of the construction industry, building an intelligent data fusion and standardization system, performing multi-source data fusion and formulating quality assessment rules, based on the intelligent data fusion and standardization system, using deep neural network algorithms and knowledge graph technology to construct an industry information model that simulates the human brain neural network, and according to the demand analysis results of the industry information model, designing an intelligent construction industry brain collaborative platform to complete the construction of the intelligent construction industry brain; the data fusion and standardization system includes optimizing the MDI calculation formula, expanding the data quality scoring model, introducing a data anomaly detection and correction mechanism, and comprehensively considering the correlation coefficient and type difference of the data source; the industry information model includes capturing the complex relationships and time series characteristics of construction industry projects based on deep neural networks and knowledge graph technology; the intelligent construction industry brain collaborative platform includes implementing dynamic adjustment and optimization, and constructing a brain collaborative neural network based on a visual interface and natural language interaction.

[0008] As a preferred scheme of the method for constructing the intelligent construction industry brain described in the present invention, the construction of the intelligent data fusion and standardization system includes introducing the correlation coefficient of the construction industry data source, optimizing the MDI calculation formula, using a standardized method based on kernel density evaluation to process outliers, and using construction industry-related data sources to expand the data quality dimension, incorporating accuracy and interpretability into the scoring model, finding an adaptive weight balance point, and constructing a construction industry data model.

[0009] As a preferred solution of the method for constructing the intelligent construction industry brain of the present invention, the construction of the construction industry data model includes calculating a multi-source data fusion index, which is expressed as: ; ; in, is the multi-source data fusion index, is the corresponding weight, is the quality score of the i-th data source, is the data standardization conversion function, is the original data, is the mean, is the standard deviation; the data quality scoring model is expressed as: ; in, is the numerical value of the data quality scoring model, To construct the industry data integrity index, is the data timeliness index, is the data consistency index, , as well as is the corresponding weight coefficient.

[0010] As a preferred solution of the method for constructing the intelligent construction industry brain described in the present invention, the multi-source data fusion and the formulation of quality assessment rules include setting a double threshold value. When the construction industry data integrity index reaches 85% and the data timeliness index exceeds 90%, the comprehensive evaluation of machine, material, method and environment measurement is automatically triggered, and the construction industry data abnormality correction coefficient is introduced. When the detection value exceeds the historical average by 1.5 times, the intelligent correction program is started, key data points are identified, and predictions are made using machine learning. If the deviation between the detection value and the original value exceeds 30%, it is marked as a pending verification status and the weight is temporarily reduced. The spatiotemporal data quality coordination coefficient is set. When the construction project spans multiple geographical locations and the time span exceeds the preset threshold, the data quality differences in different time periods and geographical locations are analyzed, and the weights of each factor in the data quality scoring model are dynamically adjusted. For geographical locations with low construction industry data integrity, the data collection frequency and quality requirements are automatically increased. For time periods with low construction industry data timeliness, data backtracking correction is triggered.

[0011] As a preferred solution of the method for constructing the intelligent construction industry brain described in the present invention, the construction of the industry information model that simulates the human brain neural network includes setting an outlier detection threshold for the standardized data, combining the five-dimensional data of "machine, material, method, environment, and measurement", performing multi-dimensional cross-validation on the outliers, and using a deep neural network algorithm to automatically identify and correct the detected outliers, loading the original data into a complete industry information model, outputting the demand analysis results, and constructing an intelligent construction industry knowledge graph based on the defined ontology structure. After converting the structured information of the knowledge graph into a low-dimensional vector using the graph neural network GNN algorithm, the knowledge graph is embedded in the TransE algorithm to map entity relationships and attributes to a low-dimensional vector space, set a vectorization threshold, batch train the industry information model using the original data of the construction industry, and apply regularization technology to prevent overfitting.

[0012] As a preferred solution of the method for constructing the intelligent construction industry brain of the present invention, the output demand analysis result includes: the calculation demand analysis result is expressed as: ; in, The result of demand analysis is: is the Sigmoid activation function, Embedding for knowledge graph The weight matrix of Embed information into knowledge graphs, For entities, For attributes, Neural network features The weight matrix of For the original data The weight matrix of is the normalization function, is the original input data, is the graph structure, is the node feature; when the industry information model detects that the frequency of occurrence of a new entity relationship pattern exceeds the preset threshold, the reconstruction procedure of the intelligent construction industry knowledge graph is triggered. The reconstruction procedure includes evaluating the consistency of the new model with the existing knowledge system; if the consistency is lower than the set standard, the expert review mechanism is activated to automatically update the intelligent construction industry knowledge graph and adjust the graph neural network topology.

[0013] As a preferred scheme of the method for constructing the intelligent construction industry brain described in the present invention, the completion of the construction of the intelligent construction industry brain includes setting the demand weight dynamic index DWDI, real-time dynamic evaluation of the influence of knowledge graphs, neural networks and original data, and adapting to different project types.

[0014] Another object of the present invention is to provide an intelligent construction industry brain construction system, which can solve the current problems of uneven quality, diverse formats and difficulty in integration and analysis of data in the construction industry through intelligent data fusion and standardization system.

[0015] As a preferred solution for the intelligent construction industry brain construction system described in the present invention, it includes: the data governance module 100, which is responsible for collecting the original data of the construction industry, standardizing the data quality assessment, setting thresholds, dynamically adjusting weights, and realizing data monitoring and correction; the intelligent analysis module 200, which is used for outlier detection, building a smart map and vectorizing the data, and constructing an industry information model; the collaborative platform module 300, which is used for demand weight assessment and feasibility analysis, allows all stakeholders to evaluate and correct the demand analysis results in real time, integrate and collaborate on various modules, and complete the construction of the intelligent construction industry brain.

[0016] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a step of a method for constructing an intelligent construction industry brain.

[0017] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for constructing an intelligent construction industry brain.

[0018] Beneficial effects of the invention: The intelligent construction industry brain construction method provided by the invention improves data quality and analyzability by establishing a reliable data fusion and standardization system, providing a reliable basis for subsequent analysis and decision-making; combines deep learning and knowledge graphs to achieve comprehensive modeling of construction industry information, effectively captures complex relationships and time series characteristics in construction projects, and conducts demand analysis; the development of an intelligent construction industry brain collaborative platform realizes dynamic adjustment of demand and project progress forecasting, optimal resource allocation, risk warning, quality control and cost management and other functions, providing project managers with intuitive and timely decision-making support. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0020] Figure 1 An overall flow chart of the method for constructing an intelligent construction industry brain provided for the first embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0022] Example 1, reference Figure 1 , which is an embodiment of the present invention, provides a method for constructing an intelligent construction industry brain, including: S1: Collect original data from the construction industry, build an intelligent data fusion and standardization system, conduct multi-source data fusion and formulate quality assessment rules.

[0023] Furthermore, collecting raw data from the construction industry includes collecting raw data from the construction industry from multiple data sources, including but not limited to sensor data, BIM model data, project management software data, supply chain data, and market research data, to ensure the comprehensiveness and diversity of data collection, covering the five dimensions of "machine, material, method, environment, and measurement."

[0024] It should be noted that building an intelligent data fusion and standardization system includes introducing correlation coefficients of construction industry data sources, optimizing the MDI calculation formula, using a standardized method based on kernel density evaluation to deal with outliers, using construction industry-related data sources to expand data quality dimensions, incorporating accuracy and interpretability into the scoring model, finding an adaptive weight balance point, and building a construction industry data model.

[0025] It should also be noted that building a construction industry data model includes calculating a fusion index based on multi-source data, expressed as: ; ; in, is the multi-source data fusion index, is the corresponding weight, For the The quality score of each data source, is the data standardization conversion function, is the original data, is the mean, is the standard deviation; the data quality scoring model is expressed as: ; in, is the numerical value of the data quality scoring model, To construct the industry data integrity index, is the data timeliness index, is the data consistency index, , as well as is the corresponding weight coefficient.

[0026] It should also be noted that the fusion of multi-source data and the formulation of quality assessment rules include setting dual thresholds. When the data integrity index of the construction industry reaches 85% and the data timeliness index exceeds 90%, the comprehensive assessment of machine, material, method and environmental measurement is automatically triggered, and the abnormal correction coefficient of construction industry data is introduced. When the detection value exceeds the historical average by 1.5 times, the intelligent correction program is started to identify key data points and use machine learning for prediction. If the deviation between the detection value and the original value exceeds 30%, it is marked as pending verification and the weight is temporarily reduced. The spatiotemporal data quality coordination coefficient is set. When a construction project spans multiple geographical locations and the time span exceeds the preset threshold, the data quality differences in different time periods and geographical locations are analyzed, and the weights of each factor in the data quality scoring model are dynamically adjusted. For geographical locations with low data integrity in the construction industry, the data collection frequency and quality requirements are automatically increased. For time periods with low timeliness of construction industry data, data backtracking correction is triggered.

[0027] It should also be noted that the fusion of multi-source data and the formulation of quality assessment rules include the introduction of the data consistency fluctuation index DCFI. When the data consistency fluctuation index DCFI shows an upward trend during the continuous assessment cycle and the latest value exceeds 1.5 times the historical average, the data consistency intelligent correction program is initiated; the data consistency intelligent correction program includes identifying the key data points that cause inconsistencies, and using machine learning algorithms to predict the correct values ​​based on historical data patterns and current construction project characteristics; if the deviation between the predicted value and the original value exceeds 30%, the system will automatically mark this data point as pending verification and temporarily reduce its weight in the data quality score.

[0028] It should also be noted that the fusion of multi-source data and the formulation of quality assessment rules include setting the spatiotemporal data quality coordination coefficient STDQC. When it is activated when a construction project spans several geographical locations or the time span exceeds a preset threshold, the data quality differences in different time periods and geographical locations are analyzed, and the weights of each factor in the data quality scoring model are dynamically adjusted; when it is detected that the data integrity of a certain geographical location is lower than that of other locations, the collection frequency and quality requirements of the data at this location are automatically increased; if it is detected that the data timeliness of a certain time period is low, the data backtracking mechanism is triggered, and the high-quality data of adjacent time periods are used for interpolation and correction.

[0029] S2: Based on the intelligent data fusion and standardization system, use deep neural network algorithms and knowledge graph technology to build an industrial information model that simulates the human brain neural network.

[0030] Furthermore, after data preparation, outlier detection and correction, the construction of the industrial information model requires the completion of knowledge graph construction, deep neural network construction, and model fusion application steps.

[0031] It should be noted that the construction of the knowledge graph includes defining the ontology of the industrial information model, including entities, attributes, and relationships. Based on the ontology definition, entities, attributes, and relationships are extracted from the original data to construct an industrial information knowledge graph. The graph neural network GNN algorithm is used to convert the structured information of the knowledge graph into a low-dimensional vector, and the TransE algorithm is used to map the entity relationships and attributes to a low-dimensional vector space.

[0032] It should also be noted that the construction of a deep neural network includes selecting a suitable deep neural network model, designing the network structure of the deep neural network, training the deep neural network model using raw data from the construction industry, and optimizing parameters.

[0033] It should also be noted that model fusion includes fusing knowledge graph embedded information, deep neural network features and original input data to build a comprehensive industry information model.

[0034] It should also be noted that the construction of an industrial information model that simulates the human brain neural network includes setting an outlier detection threshold for the standardized data, combining the five-dimensional data of "machine, material, method, environment, and measurement", performing multi-dimensional cross-validation on the outliers, and using a deep neural network algorithm to automatically identify and correct the detected outliers, loading the original data into a complete industrial information model, and outputting the demand analysis results; based on the defined ontology structure, building an intelligent construction industry knowledge graph, using the graph neural network GNN algorithm to convert the structured information of the knowledge graph into a low-dimensional vector, and then embedding the knowledge graph into the TransE algorithm to map entity relationships and attributes to a low-dimensional vector space; setting a vectorization threshold, batch training the industrial information model using the original data of the construction industry, and applying regularization technology to prevent overfitting.

[0035] It should also be noted that the output requirements analysis results include the calculation requirements analysis results expressed as: ; in, The result of demand analysis is: is the Sigmoid activation function, Embedding for knowledge graph The weight matrix of Embed information into knowledge graphs, For entities, For attributes, Neural network features The weight matrix of For the original data The weight matrix l, is the normalization function, is the original input data, is the graph structure, is the node feature; when the industry information model detects that the frequency of occurrence of a new entity relationship pattern exceeds the preset threshold, the reconstruction procedure of the intelligent construction industry knowledge graph is triggered. The reconstruction procedure includes evaluating the consistency of the new model with the existing knowledge system; if the consistency is lower than the set standard, the expert review mechanism is activated to automatically update the intelligent construction industry knowledge graph and adjust the graph neural network topology.

[0036] S3: Based on the demand analysis results of the industrial information model, design the intelligent construction industry brain collaborative platform to complete the construction of the intelligent construction industry brain.

[0037] Furthermore, completing the construction of the intelligent construction industry brain includes setting up the demand weight dynamic index DWDI to evaluate in real time the influence of knowledge graph embedding, neural network characteristics and raw data in demand analysis; when the demand weight dynamic index DWDI detects a change in the contribution of a component, it triggers the weight allocation algorithm, introduces domain-specific factors, and dynamically adjusts the weight ratio of each component according to different construction project types.

[0038] It should be noted that completing the construction of the intelligent construction industry brain includes setting the demand feasibility threshold RFT to achieve difficulty, resource investment and market acceptance; when the demand analysis results reach the demand feasibility threshold RFT, the refined analysis program is triggered, and the threshold adaptive mechanism is introduced. According to historical project data and industry development trends, the demand feasibility threshold RFT is regularly updated.

[0039] It should also be noted that completing the construction of the smart construction industry brain includes developing a demand feedback intelligent interface RFII, allowing all stakeholders to evaluate and correct the demand analysis results in real time; setting the demand consensus index RCI. When the demand consensus index RCI is lower than the preset value, the multi-party negotiation mechanism is automatically initiated to promote the unification and optimization of demand.

[0040] Example 2 is an embodiment of the present invention, which provides a method for constructing an intelligent construction industry brain. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0041] Specifically, a large-scale construction project is used as the test object. The project is a comprehensive commercial center covering an area of ​​50,000 square meters, including office buildings, shopping malls and underground parking lots. The total construction area of ​​the project is about 200,000 square meters, and the planned construction period is 24 months.

[0042] First, we collected all kinds of data related to the construction project, including design drawings, construction progress, material procurement, personnel deployment, equipment use, safety management, and quality control information. The data sources include project management systems, BIM models, IoT sensors, on-site video surveillance, and supply chain management systems. To ensure the accuracy and completeness of the data, we adopted a multi-source data cross-validation method.

[0043] Furthermore, we will build an intelligent data fusion and standardization system, adopt a distributed storage architecture, and combine blockchain technology to ensure data security and traceability; through unified data standards and interface specifications, we will achieve seamless integration of data from different sources and in different formats; and introduce natural language processing technology to convert unstructured data into structured information.

[0044] Furthermore, based on data fusion and standardization, deep neural network algorithms and knowledge graph technology are used to build an industrial information model that simulates the human brain neural network. The multi-layer convolutional neural network structure, combined with the attention mechanism and long short-term memory network (LSTM), can effectively capture the complex relationships and time series characteristics in construction projects; the knowledge graph incorporates the professional knowledge and experience of the construction industry, enhancing the interpretability and reasoning ability of the model. Based on the demand analysis results of the industrial information model, the intelligent construction industry brain collaborative platform is designed and developed, which adopts a microservice architecture to support multi-terminal access and real-time data updates. The core functional modules include project schedule forecasting, resource optimization allocation, risk warning, quality control and cost management. Through the visual interface and natural language interaction, the platform can provide intuitive and timely decision support for project managers.

[0045] Specifically, as shown in Table 1, by comparing the experimental data, it can be clearly seen that the intelligent construction industry brain is superior to traditional project management methods in many aspects; in terms of project completion time, after adopting the intelligent construction industry brain, the project completion time was shortened from 26 months to 22 months, 15.38% earlier. This improvement is due to the industry brain's ability to accurately predict and dynamically adjust project progress. By analyzing historical data and real-time information, the system can promptly identify potential delay factors and provide optimization solutions to improve project execution efficiency.

[0046] Table 1 Comparison between the present invention and the traditional method

[0047] Furthermore, in terms of cost savings, the intelligent construction industry brain helped the project save 8.5% of costs. This achievement is mainly due to the intelligent optimization of resource allocation and the refined management of the procurement process. By predicting material demand and price fluctuations, procurement can be carried out at the best time, avoiding the additional costs caused by inventory backlogs and emergency procurement. In terms of resource utilization, under traditional methods, the resource utilization rate is 72%, while after adopting the intelligent construction industry brain, this indicator has increased to 89%, an increase of 23.61%. The industry brain can accurately allocate resources by real-time monitoring and predicting the use of various resources, reducing idle resources and waste.

[0048] Furthermore, in terms of the quality problem detection rate, it increased from 65% to 92%, an increase of 41.54%. The intelligent construction industry brain uses deep learning algorithms to identify tiny quality anomalies from massive construction data. Combined with the professional knowledge in the knowledge graph, it can not only detect potential quality problems early, but also provide targeted solutions, greatly improving the level of engineering quality control. In terms of the incidence of safety accidents and safety management, the intelligent construction industry brain also performed well. The incidence of safety accidents has dropped from 1.8% to 0.5%, a decrease of 72.22%. By analyzing historical accident data, real-time monitoring of construction sites, identifying dangerous behavior patterns and other means, it can warn of potential safety risks in advance and automatically generate safety management measures, effectively preventing the occurrence of various safety accidents.

[0049] Example 3 is an embodiment of the present invention, which provides an intelligent construction industry brain construction system, including a data governance module 100, an intelligent analysis module 200 and a collaborative platform module 300.

[0050] Among them, S4: the data governance module 100 includes a data collection sub-module 101, a data quality assessment sub-module 102, a data standardization sub-module 103, a weight dynamic sub-module 104, a data monitoring and correction sub-module 105, and a data fusion sub-module 106.

[0051] Furthermore, the data collection submodule 101 is used to collect original data of the construction industry from multiple data sources, the data quality assessment submodule 102 is used to perform quality assessment on the collected data, including the completeness, accuracy, consistency, and timeliness of the data, and set the threshold for data quality assessment, the data standardization submodule 103 is used to use a standardized method to unify the data from different data sources and eliminate dimensional differences, the weight dynamic submodule 104 is used for data quality assessment results, dynamically adjusts the weights of different data sources, and ensures the rationality of data fusion, the data monitoring and correction submodule 105 is used to monitor data quality in real time, identify and correct abnormal data, and ensure the continuous stability of data quality, and the data fusion submodule 106 is used to fuse data from multiple data sources to form a complete and unified data set.

[0052] It should be noted that the data collection submodule 101 is the data basis between each small module, the data quality assessment submodule 102 provides data quality information for the data standardization submodule 103, and guides the data monitoring and correction submodule 105 to correct the data, provides weight information to the weight dynamic submodule 104 to provide corrected data to the data standardization submodule 103, and the data fusion submodule 106 provides fused data to the intelligent analysis module 200.

[0053] It should also be noted that the data governance module 100 provides a reliable data foundation for the intelligent analysis module 200 and is a prerequisite for the effective operation of the entire system.

[0054] S5: The intelligent analysis module 200 includes an outlier detection submodule 201, a knowledge graph construction submodule 202, a deep neural network construction submodule 203, and a model training submodule 204.

[0055] It should be noted that the outlier detection submodule 201 performs outlier detection on the standardized data, the knowledge graph construction submodule 202 extracts entities, attributes, and relationships from the data collection submodule 101 based on the ontology definition, and constructs an industry information knowledge graph, the deep neural network construction submodule 203 is used to select a suitable deep neural network model, and the model training submodule 204 uses the original data of the construction industry to train the deep neural network model and optimize the parameters.

[0056] It should also be noted that the intelligent analysis module 200 relies on the data provided by the data governance module 100 to output the industrial information model and provide decision support for the collaborative platform module 300.

[0057] S6: The collaborative platform module 300 includes a demand weight evaluation submodule 301 , a feasibility analysis submodule 302 , and a module integration collaborative submodule 303 .

[0058] It should be noted that the demand weight assessment submodule 301 assesses the weights of different demands, determines priorities, and provides demand weight information for the feasibility analysis submodule 302. The feasibility analysis submodule 302 performs a feasibility analysis on the demand analysis results, assesses the feasibility and potential risks of implementation, and provides an integration foundation for the module integration and collaboration submodule 303. The module integration and collaboration submodule 303 relies on the results of the data governance module 100, the intelligent analysis module 200, and the collaborative platform module 300 to ultimately realize the construction of the intelligent construction industry brain.

[0059] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0060] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0061] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk case (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0062] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logical function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A method for constructing an intelligent construction industry brain, characterized in that: include: Collect original data from the construction industry, build an intelligent data fusion and standardization system, conduct multi-source data fusion and formulate quality assessment rules; Based on the intelligent data fusion and standardization system, we use deep neural network algorithms and knowledge graph technology to build an industrial information model that simulates the human brain neural network; According to the demand analysis results of the industrial information model, the intelligent construction industry brain collaborative platform is designed to complete the construction of the intelligent construction industry brain; The data fusion and standardization system includes optimizing the MDI calculation formula, expanding the data quality scoring model, introducing a data anomaly detection and correction mechanism, and comprehensively considering the correlation coefficient and type differences of data sources; The industrial information model includes deep neural network and knowledge graph technology to capture the complex relationships and temporal characteristics of construction industry projects; The intelligent construction industry brain collaborative platform includes implementing dynamic adjustments and optimizations, building a brain collaborative neural network based on visual interfaces and natural language interactions.

2. The method for constructing an intelligent construction industry brain according to claim 1, characterized in that: The construction of intelligent data fusion and standardization system includes: The correlation coefficient of the construction industry data source was introduced, the MDI calculation formula was optimized, and the outliers were processed using a standardized method based on kernel density evaluation; We also use construction industry-related data sources to expand the data quality dimension, incorporate accuracy and explainability into the scoring model, find the adaptive weight balance point, and build a construction industry data model.

3. The method for constructing an intelligent construction industry brain according to claim 1 or 2, characterized in that: The construction of the construction industry data model includes: The calculation is based on the multi-source data fusion index, expressed as: ； ； in, is the multi-source data fusion index, is the corresponding weight, is the quality score of the i-th data source, is the data standardization conversion function, is the original data, is the mean, is the standard deviation; The data quality scoring model is expressed as: ； in, is the numerical value of the data quality scoring model, To construct the industry data integrity index, is the data timeliness index, is the data consistency index, , as well as is the corresponding weight coefficient.

4. The method for constructing an intelligent construction industry brain according to claim 3, characterized in that: The multi-source data fusion and quality assessment rule formulation include: Set dual thresholds. When the data integrity index of the construction industry reaches 85% and the data timeliness index exceeds 90%, the comprehensive evaluation of machine, material, method and environment testing will be automatically triggered. Introducing the abnormal correction coefficient of construction industry data. When the detection value exceeds the historical average by 1.5 times, the intelligent correction program is activated to identify key data points and use machine learning to make predictions. If the deviation between the detected value and the original value exceeds 30%, it will be marked as pending verification and the weight will be temporarily reduced. The spatiotemporal data quality coordination coefficient will be set. When the construction project spans multiple geographical locations and the time span exceeds the preset threshold, the data quality differences in different time periods and geographical locations will be analyzed, and the weights of each factor in the data quality scoring model will be dynamically adjusted; For geographical locations with low integrity of construction industry data, the frequency and quality requirements of data collection are automatically increased. For time periods with low timeliness of construction industry data, data backtracking correction is triggered.

5. The method for constructing an intelligent construction industry brain according to claim 1 or 4, characterized in that: The industrial information model for simulating the human brain neural network includes: For the standardized data, we set the outlier detection threshold, combined with the five-dimensional data of "machine, material, method, environment, and measurement", to perform multi-dimensional cross-validation on the outliers, and use the deep neural network algorithm to automatically identify and correct the detected outliers, load the original data into a complete industrial information model, and output the demand analysis results; Based on the defined ontology structure, we build a knowledge graph for the intelligent construction industry. We use the graph neural network (GNN) algorithm to transform the structured information of the knowledge graph into a low-dimensional vector. Then, we embed the knowledge graph into the TransE algorithm to map entity relationships and attributes into a low-dimensional vector space. Set the vectorization threshold, batch train the industry information model by building the original data of the industry, and apply regularization technology to prevent overfitting.

6. The method for constructing the intelligent construction industry brain according to claim 5, characterized in that: The output requirements analysis results include: The calculation requirement analysis results are expressed as: ； in, The result of demand analysis is: is the Sigmoid activation function, Embedding for knowledge graph The weight matrix of Embed information into knowledge graphs, For entities, For attributes, Neural network features The weight matrix of For the original data The weight matrix of is the normalization function, is the original input data, is the graph structure, is the node feature; When the industry information model detects that the frequency of occurrence of a new entity relationship pattern exceeds the preset threshold, the reconstruction procedure of the intelligent construction industry knowledge graph is triggered. The reconstruction procedure includes evaluating the consistency of the new model with the existing knowledge system; if the consistency is lower than the set standard, the expert review mechanism is activated to automatically update the intelligent construction industry knowledge graph and adjust the graph neural network topology.

7. The method for constructing an intelligent construction industry brain according to claim 6, characterized in that: The construction of the intelligent construction industry brain includes: Set the demand weight dynamic index DWDI to dynamically evaluate the influence of knowledge graphs, neural networks and raw data in real time to adapt to different project types.

8. Intelligent construction industry brain construction system, characterized by: It includes a data governance module (100), an intelligent analysis module (200) and a collaborative platform module (300); The data governance module (100) is responsible for collecting raw data of the construction industry, evaluating and standardizing data quality, setting thresholds, dynamically adjusting weights, and realizing data monitoring and correction; The intelligent analysis module (200) is used for outlier detection, building a smart map and vectorizing data to construct an industrial information model; The collaborative platform module (300) is used for demand weight assessment and feasibility analysis, allowing all stakeholders to evaluate and correct the demand analysis results in real time, integrate and collaborate with each module, and complete the construction of the intelligent construction industry brain.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of a method for constructing an intelligent construction industry brain as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a method for constructing an intelligent construction industry brain as described in any one of claims 1 to 7 are implemented.

Images