Building engineering construction data management system and method based on cloud platform

Abstract

The embodiment of the invention relates to a building engineering construction data management system and method based on a cloud platform, and belongs to the technical field of electronic data processing, and the method comprises the steps: obtaining building project construction data, and forming a first construction project data group; fusing the target construction project data to the reference data to form a second construction project data group; calculating a construction compliance coefficient to obtain a construction information comprehensive compliance coefficient; analyzing compliance data characteristics by the cloud platform to obtain key construction data; screening an optimal processing node; and calling the construction information processing template to carry out marking processing on the data and then transmitting the data to a specified supervision terminal. According to the embodiment of the invention, experience reuse is realized through data clustering fusion, the comprehensiveness of evaluation is ensured, load balancing is realized through dynamic scheduling of cloud resources, and the problems of supervision delay, resource waste and hidden danger missing detection caused by complex data sources, compliance judgment lagging and resource scheduling mismatching are solved.

Description

Technical Field

[0001] This application relates to a cloud-based construction engineering data management system and method, which belongs to the technical field of electronic data processing. Background Technology

[0002] With the continuous improvement of the informatization level of the construction industry, modern construction projects generate massive amounts of structured and unstructured data during the construction process. The data covers a wide range of information such as construction progress, material consumption, manpower allocation, equipment operation, quality inspection and safety monitoring. How to effectively collect, process and analyze massive amounts of construction data is of great significance for achieving refined management, cost control and safety assurance.

[0003] Currently, in the field of construction data management, existing technical solutions typically use traditional relational databases combined with business intelligence tools for data processing. For example, construction data is acquired through manual entry or simple sensor collection, stored in a centralized database, and then construction projects are simply categorized using fixed data classification standards. Finally, various statistical reports are generated through preset report templates for managers to refer to.

[0004] However, with the continuous expansion of construction projects and the increase in data complexity, existing technical solutions have revealed the following prominent problems: At the data acquisition and processing level, existing technologies lack the ability to deeply integrate multi-source heterogeneous data. Construction data comes from a wide range of sources, including BIM models, schedules, bills of materials, sensor readings, and on-site images. These data differ significantly in format, frequency, and semantics. Existing technologies often can only process single types of data and cannot establish effective correlation analysis. This results in a single dimension of construction project characteristics, making it difficult to fully reflect the true state of the project. Especially when processing unstructured data, existing technologies usually use simple keyword matching or manual annotation, which results in high labor costs, low efficiency in matching and annotation, and is prone to errors.

[0005] In terms of data analysis methods, existing static classification standards are ill-suited to the dynamic changes inherent in the construction process. Building construction is a complex, dynamic process with intricate spatiotemporal relationships and logical dependencies among its various sub-projects. Current fixed classification methods cannot identify changes in construction patterns, nor can they intelligently group and merge data based on similarities. This results in data analysis findings often lagging behind actual construction progress, hindering the provision of forward-looking decision support for project management.

[0006] In terms of compliance monitoring, existing technologies mainly rely on manual comparison and experience-based judgment, which cannot conduct real-time and quantitative compliance assessments of various indicators during the construction process. Project managers often need to spend a lot of time reviewing regulatory documents and then manually comparing actual construction data, which can easily lead to judgment biases due to subjective factors. Furthermore, due to the lack of comprehensive compliance assessment indicators, it is difficult to make an accurate judgment on the overall compliance status of the project.

[0007] In terms of computing resource scheduling, existing technologies lack targeted optimization for the characteristics of construction data processing. Construction data processing exhibits distinct periodic peak characteristics; for example, computing demands increase dramatically during daily construction report generation periods or large model rendering periods. Existing technologies typically employ fixed server resource configurations, leading to either insufficient computing resources and processing delays during peak periods, or significant resource idleness during normal times. This is particularly problematic when handling computationally intensive tasks such as large BIM models or real-time video analysis, where existing technologies often respond slowly and fail to meet the real-time requirements of construction sites.

[0008] In terms of data standardization and output, existing technologies lack unified information processing standards. Different projects and departments often use their own independent data formats and output templates, which leads to serious obstacles to data sharing and collaborative work. This not only increases the difficulty of data integration but also requires regulatory personnel to spend a lot of time understanding and converting data reports in different formats, seriously affecting management efficiency.

[0009] The Chinese invention patent with publication number CN117519948A, titled "Method and System for Adjusting Computing Resources in Building Construction Based on Cloud Platform," involves the processing of construction engineering data. It determines the processing server corresponding to the construction data from the running target server and the idle target server, and calculates the load balancing deviation coefficient corresponding to the processing server. However, in the specific process of load balancing, it does not involve the constraints, processing, and grouping of the original data, nor does it involve defining data structuring specifications and processing rules based on the construction information processing template. It also lacks specific technical details on dividing data according to multiple dimensions, which requires improvement.

[0010] The patent application CN119444470A, titled "An Intelligent Drilling Construction Data Security Monitoring System Based on a Control Cloud Platform," relates to the field of construction data security monitoring. It can achieve comprehensive, real-time, and efficient monitoring and analysis of drilling construction data, better identify potential safety risks, promptly detect and handle safety hazards during construction, and improve the accuracy and timeliness of early warning information. However, it does not consider the impact of server load balancing on data processing efficiency in the specific processing process, and there is room for improvement.

[0011] In conclusion, existing technologies can no longer meet people's needs and urgently need to be improved. Summary of the Invention

[0012] The main objective of this application is to provide a cloud-based construction engineering data management system and method that can integrate multi-source data, achieve intelligent analysis, support dynamic resource scheduling, and provide standardized output for construction engineering data management, thereby addressing the shortcomings of existing technologies.

[0013] The embodiments of this application adopt the following technical solutions: According to one aspect of the embodiments of this application, a method for managing construction project data based on a cloud platform is provided, comprising: acquiring construction project data, constructing a multi-source construction element dataset, extracting multi-dimensional features from the multi-source construction element dataset to generate a corresponding construction project feature list, grouping the construction project data based on the construction project feature list using a clustering calculation method to form a first construction project data group; determining the feature matching degree between target construction project data and reference construction project data, and when the feature matching degree reaches a preset fusion threshold, merging the target construction project data into the reference construction project data to form a second construction project data group; comparing and analyzing the first construction project data group with preset construction benchmark indicators, and calculating based on the second construction project data group... The construction compliance coefficient is calculated using a weighted statistical algorithm to obtain a comprehensive compliance coefficient for construction information. The compliance data characteristics of the comprehensive compliance coefficient are analyzed through the cloud platform. Based on these characteristics, corresponding key construction data is obtained, and the computational resource requirements for these key construction data are calculated. The network transmission delay distance between different server nodes in the cloud platform is obtained. Combining the remaining processing capacity of the service nodes with the computational resource requirements, the optimal processing node is selected from running and idle service nodes. The load balancing deviation of the optimal processing node is calculated, and a dynamic allocation scheme for computational resources is implemented based on this deviation. The corresponding construction information processing template is retrieved from the construction storage database, and the construction data of the construction project is tagged according to the template and transmitted to the designated construction supervision terminal.

[0014] According to at least one specific embodiment of the present application, the step of acquiring construction project data, constructing a multi-source construction element dataset, extracting multi-dimensional features from the multi-source construction element dataset to generate a corresponding construction project feature list, and grouping the construction project data based on the construction project feature list using a clustering calculation method to form a first construction project data group, further includes: calculating weight coefficients for the multi-dimensional features in the multi-source construction element dataset, combining the weight coefficients and using a weighted feature fusion algorithm to generate optimized feature vectors for each construction sub-project, and then optimizing the features... Vectors are used for preliminary clustering calculations to obtain an initial clustered dataset and a set of sample numbers for each subclass within the initial clustered dataset. The 25th percentile, 75th percentile, and mean of the sample number set are calculated to determine the corresponding high-density cluster judgment threshold and sparse cluster judgment threshold. The initial clustered dataset is then divided into high-density clusters, sparse clusters, and core clusters. The cutting threshold for high-density clusters and the exclusion threshold for sparse clusters are obtained. Based on the cutting threshold and the exclusion threshold, the high-density clusters and the core clusters are merged to generate the first construction project data group.

[0015] According to at least one specific embodiment of the present application, the step of obtaining the cutting threshold of high-density clusters and the exclusion threshold of sparse clusters, and merging the high-density clusters and the core clusters according to the cutting thresholds and the exclusion thresholds to generate a first construction project data group specifically involves: obtaining the cutting threshold of high-density clusters and the exclusion threshold of sparse clusters; segmenting the high-density clusters based on the cutting thresholds; removing the sparse clusters based on the exclusion thresholds; merging the segmented high-density clusters and the removed core clusters; extracting the centroid features after the merge operation; and grouping the construction project data according to the centroid features to generate a first construction project data group.

[0016] According to at least one specific embodiment of the present application, the step of comparing and analyzing the first construction project data group with preset construction benchmark indicators, calculating the construction compliance coefficient based on the second construction project data group, and calculating the comprehensive compliance coefficient of construction information based on the construction compliance coefficient using a weighted statistical algorithm, further includes: extracting key construction process indicators of each sub-project from the second construction project data group, establishing a mapping relationship table between actual indicators and benchmark indicators, performing quantitative analysis on each mapping relationship in the mapping relationship table, and calculating the absolute deviation and relative deviation rate of the key construction process indicators relative to the benchmark values; calculating the compliance coefficient of each mapping relationship based on the absolute deviation and the relative deviation rate, generating a set of construction compliance coefficients according to the influence factors of different key construction process indicators; determining the indicator importance weights of different elements in the set of construction compliance coefficients based on the analytic hierarchy process, multiplying the compliance coefficients by the corresponding weight values ​​using a weighted summation formula and summing them to obtain a weighted total score; converting the weighted total score into a standardized value in the range of 0-1 to generate a comprehensive compliance coefficient of construction information representing the overall construction quality.

[0017] According to at least one specific embodiment of this application, the step of determining the importance weights of different elements in the construction compliance coefficient set based on the analytic hierarchy process (AHP), and multiplying the compliance coefficients by their corresponding weight values ​​using a weighted summation formula to obtain a weighted total score, further includes: constructing a judgment matrix to perform pairwise importance comparisons on the compliance coefficients of construction cycle, material consumption, labor efficiency, and equipment utilization rate in the set, wherein the elements in the judgment matrix are assigned values ​​according to the 1-9 scaling method; solving for the largest eigenvalue of the judgment matrix and its corresponding normalized eigenvector; calculating the consistency ratio of the judgment matrix; and confirming the normalized eigenvector as the indicator importance weight of the compliance coefficient when the consistency ratio is less than 0.1.

[0018] According to at least one specific implementation of the embodiments of this application, the step of confirming the normalized feature vector as the indicator importance weight of the compliance coefficient is specifically as follows: after obtaining the indicator importance weight, each compliance coefficient is multiplied by its corresponding indicator importance weight using a weighted summation formula and then summed to obtain the weighted total score.

[0019] According to at least one specific implementation of the embodiments of this application, the step of parsing the compliance data features of the comprehensive compliance coefficient of construction information through the cloud platform, obtaining the corresponding key construction data based on the compliance data features, calculating the computing resource requirements of the key construction data, obtaining the network transmission delay distance between different server nodes in the cloud platform, and selecting the optimal processing node from running and idle service nodes by combining the remaining processing capacity of the service nodes and the computing resource requirements, further includes: collecting the dynamic operating parameters of each node in real time through resource monitoring agents deployed on each server node of the cloud platform, generating a multi-dimensional resource vector for each node, forming a real-time resource profile of all nodes on the platform; and performing feature analysis on the key construction data to be processed. The process involves parsing and identifying the corresponding data type, data scale, and required processing algorithm. Based on the algorithm complexity model and data volume, the required computing resources for the data processing task are calculated. Optimal processing nodes are then selected based on the obtained real-time resource profile and required computing resources. After determining the optimal processing node, dynamic scheduling of computing resources is performed through a container orchestration engine. During task execution, a monitoring agent continuously tracks the node's resource usage and task execution progress. When insufficient node resources or task execution anomalies are detected, a container migration mechanism is triggered in real time to reschedule the task to another suitable node. After the task is completed, the allocated container resources are released, and the node resource information in the resource status database is updated, completing the entire resource scheduling lifecycle management.

[0020] According to at least one specific embodiment of the present application, the step of selecting the optimal processing node based on the obtained real-time resource profile and computing resource demand capacity further includes: setting constraints for node selection, the constraints including that the node's remaining memory is greater than the task's memory usage, the node's available CPU computing power is greater than the task's estimated CPU computing power, and the network transmission latency is lower than the task's maximum allowable latency; using a weighted scoring algorithm to comprehensively evaluate the candidate nodes selected through the constraints, and calculating the comprehensive adaptation score of each candidate node based on the matching degree between the node's remaining computing resources and the task's required resources, the network transmission latency from the node to the data source, and the node's current load balancing degree, using preset weight coefficients, and selecting the node with the highest comprehensive adaptation score as the optimal processing node.

[0021] According to at least one specific embodiment of the present application, the calculation of the load balancing deviation of the optimal processing node, the implementation of a dynamic allocation scheme for computing resources based on the load balancing deviation, the retrieval of the corresponding construction information processing template from the construction storage database, the marking of the construction data of the construction project according to the template, and the transmission to the designated construction supervision terminal, further includes: collecting resource utilization data of each node in real time through a monitoring agent deployed on the cloud platform node, calculating the load balancing deviation of the cluster based on the collected data, and when the deviation exceeds a preset threshold, scheduling the computing tasks on the overloaded nodes to low-load nodes and allocating additional virtualized computing resources to the overloaded nodes. The task scheduler prioritizes the construction data processing tasks input to the cloud platform. Based on task priority and resource requirement characteristics, a task scheduling strategy model is established using a reinforcement learning algorithm based on deep Q-networks. The matching degree between tasks and nodes is detected, and corresponding optimization scheduling decisions are made to allocate computationally intensive tasks to high-performance servers. The metadata definition module calls the construction information processing template from the database. The construction information processing template is used to define data structuring specifications and processing rules. The raw construction data is parsed according to the template specifications, key information is extracted and type tags are added to generate standardized construction information data packets, which are asynchronously transmitted to the designated construction supervision terminal through a message queue.

[0022] According to another aspect of the embodiments of this application, a cloud-based construction project data management system is provided to implement the cloud-based construction project data management method described in any one of the claims, comprising: a first construction project data grouping generation module, which acquires construction project data, constructs a multi-source construction element dataset, performs multi-dimensional feature extraction on the multi-source construction element dataset, generates a corresponding construction project feature list, and groups the construction project data based on the construction project feature list using a clustering calculation method to form a first construction project data group; a second construction project data grouping generation module, which determines the feature matching degree between target construction project data and reference construction project data, and when the feature matching degree reaches a preset fusion threshold, merges the target construction project data into the reference construction project data to form a second construction project data group; and a construction information comprehensive compliance coefficient generation module, which compares the first construction project data group with a preset construction benchmark. The system performs comparative analysis of indicators, calculates construction compliance coefficients based on the second construction project data grouping, and calculates the comprehensive compliance coefficient of construction information based on the construction compliance coefficients using a weighted statistical algorithm. The optimal processing node selection module analyzes the compliance data characteristics of the comprehensive compliance coefficient of construction information through the cloud platform, obtains the corresponding key construction data based on the compliance data characteristics, calculates the computing resource requirements of the key construction data, obtains the network transmission delay distance between different server nodes in the cloud platform, and selects the optimal processing node from running and idle service nodes by combining the remaining processing capacity of the service nodes and the computing resource requirements. The load balancing reliability calculation module calculates the load balancing deviation of the optimal processing node, implements a dynamic allocation scheme for computing resources based on the load balancing deviation, calls the corresponding construction information processing template from the construction storage database, marks the construction data of the construction project according to the template, and transmits it to the designated construction supervision terminal.

[0023] The beneficial technical effects of the embodiments of this application are: This application embodiment acquires multi-source heterogeneous data and performs clustering calculations, matches and fuses features to generate compliance coefficient weighting, and then manages construction data through dynamic scheduling of cloud resources. By automatically merging and integrating massive amounts of messy construction data into reference data in real time, historical experience is immediately injected into the current project, enabling data asset reuse. Online comparison with preset construction indicators generates construction compliance coefficients, which are then weighted to obtain a comprehensive compliance coefficient, ensuring that the evaluation results reflect both individual risks and overall performance. When processing data on the cloud platform, this application embodiment calculates the computational resource requirements of key data, quantifies the network transmission latency and remaining processing capacity between server nodes, selects the optimal processing node, smoothly migrates computational tasks to lightly loaded nodes to achieve load balancing, and finally calls the processing template matching the key data from the construction storage database to perform tokenization processing on the raw data and transmits it to the designated monitoring terminal through a communication link. This solves the technical problems of monitoring delays, computational waste, and missed detection of hidden dangers caused by the diverse sources of construction data, delayed compliance judgments, and the disconnect between cloud resources and business load. Attached Figure Description

[0024] To more clearly illustrate the specific implementation methods of the embodiments of this application or the technical solutions in the prior art, the drawings used in the description of the specific implementation methods or the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart of the technical solutions provided in steps S1 to S5.

[0026] Figure 2 This is a flowchart of the optimization technical solutions provided in steps S11 to S13.

[0027] Figure 3 This is a flowchart of the optimization technical solutions provided in steps S31 to S34.

[0028] Figure 4 This is a flowchart of the optimization technical solution provided in steps S41 to S44.

[0029] Figure 5 This is a flowchart of the optimization technical solutions provided in steps S51 to S53.

[0030] Figure 6 This is an architecture diagram of a cloud-based construction engineering data management system. Detailed Implementation

[0031] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the embodiments of this application, and not all embodiments. Based on the specific implementation methods in the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of the embodiments of this application.

[0032] like Figure 1 The cloud-based construction data management method shown includes: Step S1: Obtain construction data of building projects, construct a multi-source construction element dataset, extract multi-dimensional features from the multi-source construction element dataset, generate a corresponding construction project feature list, and group the construction data of building projects based on the construction project feature list using a clustering calculation method to form the first construction project data group.

[0033] Step S2 involves determining the feature matching degree between the target construction project data and the reference construction project data. When the feature matching degree reaches a preset fusion threshold, the target construction project data is fused into the reference construction project data to form a second construction project data group. In step S2, the reference construction project data refers to a set of baseline data that has undergone feature extraction and classification processing in a historical database or current project database. The baseline data set has clear construction mode characteristics and represents a combination of digital features of a typical construction mode. For example, the applicant selects 10 completed high-rise building projects of the same type from the company's historical database, extracts complete data records of their standard floor concrete construction stage, and constitutes the reference construction project data. Construction cycle characteristics: Single-layer construction cycle: 5±0.5 days, process connection interval: <4 hours, night construction ratio: <15%.

[0034] Resource consumption characteristics: Average daily concrete consumption: 120±10 cubic meters, formwork turnover rate: 85%±3%, labor allocation: 25±2 people / shift.

[0035] Quality and safety characteristics: Concrete strength compliance rate: 100%, Verticality deviation: <8mm, Safety inspection pass rate: 98%±1%.

[0036] Step S3: Compare and analyze the data group of the first construction project with the preset construction benchmark indicators, calculate the construction compliance coefficient based on the data group of the second construction project, and calculate the comprehensive compliance coefficient of construction information based on the construction compliance coefficient using a weighted statistical algorithm.

[0037] Step S4: Analyze the compliance data characteristics of the comprehensive compliance coefficient of construction information through the cloud platform, obtain the corresponding key construction data based on the compliance data characteristics, calculate the computing resource requirements of the key construction data, obtain the network transmission delay distance between different server nodes in the cloud platform, and select the optimal processing node from the running service nodes and idle service nodes by combining the remaining processing capacity of the service nodes and the computing resource requirements.

[0038] Step S5: Calculate the load balancing deviation of the optimal processing node, implement a dynamic allocation scheme for computing resources based on the load balancing deviation, call the corresponding construction information processing template from the construction storage database, mark the construction data of the construction project according to the template, and transmit it to the designated construction supervision terminal.

[0039] The technical solutions provided in steps S1 to S5 constitute the entire process from acquiring construction data to making intelligent decisions. They refine, integrate, and analyze the raw data, and finally, with the support of intelligent resource scheduling, transform the raw data into information that can be used for decision-making. This realizes a data-driven construction data management method with self-optimization capabilities.

[0040] Step S1 groups the construction data and integrates the grouped data. By using clustering algorithms, the inherent patterns and similar data subjects in the massive data are discovered, thus overcoming the subjective limitations of traditional classification based on human experience. Step S2 uses feature matching to dynamically and intelligently integrate new data (target project) with historical experience (reference project). The combination of steps S1 and S2 enables prediction and early warning of the current project's trajectory based on historical data, achieving a technological advancement from passive recording to proactive prediction.

[0041] Step S3 processes the fused data and performs corresponding compliance analysis, resulting in a hierarchical data logic for construction engineering data. Step S3 conducts micro-level data analysis on the fused and optimized data group (second construction project data group), quantitatively assessing the deviation between the fused data and standards and specifications, ensuring that the compliance inspection results have both a global perspective on construction engineering and the ability to identify and address detailed issues in construction engineering.

[0042] Step S4 automatically identifies key construction data with poor compliance and high risk based on the comprehensive compliance coefficient of construction information, and allocates more cloud platform computing resources to key construction data. Dynamic load balancing is used to prevent the system from encountering bottlenecks when handling peak requests, and resource allocation is dynamically driven according to the business needs of building construction.

[0043] Step S5 calls the construction information processing template to mark the processing results of step S4, transforming technical data into structured management information and transmitting it to the monitoring terminal, so that the data results from steps S1 to S4 can be transformed into strategic basis to support decision-making at the construction site.

[0044] In summary, steps S1 to S5 construct an intelligent system that goes from automatic perception to intelligent analysis, and then to efficient cloud platform data processing to form key construction management information. This significantly improves the level of precision in construction project management, decision-making efficiency, and resource utilization efficiency of the cloud platform in processing construction data.

[0045] like Figure 2 As shown, preferably, in step S1, construction project data is acquired, a multi-source construction element dataset is constructed, multi-dimensional features are extracted from the multi-source construction element dataset to generate a corresponding construction project feature list, and the construction project data is grouped based on the construction project feature list using a clustering calculation method to form a first construction project data group, further including: Step S11: Calculate the weight coefficients of the multidimensional features in the multi-source construction element dataset, combine the weight coefficients and use the weighted feature fusion algorithm to generate the optimized feature vectors of each construction sub-project, perform preliminary clustering calculations on the optimized feature vectors, and obtain the initial clustering dataset and the set of sample numbers of each subclass in the initial clustering dataset.

[0046] Step S12: Calculate the 25th percentile, 75th percentile, and mean point of the sample set, determine the corresponding high-density cluster judgment threshold and sparse cluster judgment threshold, and divide the initial clustering dataset into high-density clusters, sparse clusters, and core clusters.

[0047] Step S13: Obtain the cutting threshold of high-density clusters and the exclusion threshold of sparse clusters. Based on the cutting threshold and the exclusion threshold, merge the high-density clusters and the core clusters to generate the first construction project data group.

[0048] The optimization technical solutions provided in steps S11 to S13 enable intelligent processing and optimization of construction data grouping. Based on the clustering calculation, a judgment of high-density / sparse clusters is also performed. More representative data groups are extracted from the mixed original data, and the high-density clusters and core clusters are merged to generate the first construction project data group.

[0049] Step S11 is for feature optimization and preliminary clustering. The introduction of weight coefficients and weighted fusion changes the static data processing method that treats all features equally in traditional clustering, highlights the influence of key features, and makes the results of preliminary clustering more reflective of the essential differences in construction projects.

[0050] Step S12, based on step S11, performs cluster identification and classification. Based on statistical principles (quantiles), step S12 filters clusters, classifying all clusters into three different types: high-density, sparse, and core. High-density clusters indicate potential data bloat and contamination within the data, while sparse clusters may exhibit low data density and lack of representativeness. Core clusters typically represent healthy and complete data. This cluster classification in step S12 provides a rational arrangement of the construction data, laying the foundation for data reconstruction and merging in step S13. Step S13 merges high-density and core clusters based on cutting and exclusion thresholds, generating the first construction project data group. This first group eliminates noise and outliers, ensuring full utilization of the high-quality data from the core clusters.

[0051] For example, in step S13, the cutting threshold of high-density clusters and the exclusion threshold of sparse clusters are obtained. Based on the cutting threshold and the exclusion threshold, the high-density clusters and the core clusters are merged to generate a first construction project data group, specifically: Step S131: Obtain the cutting threshold of high-density clusters and the exclusion threshold of sparse clusters, perform segmentation processing on high-density clusters based on the cutting thresholds, and perform elimination processing on sparse clusters based on the exclusion thresholds.

[0052] Step S132: Merge the high-density clusters after segmentation with the core clusters after removal, extract the centroid features after merging, and group the construction data of the building project according to the centroid features to generate the first construction project data group.

[0053] In the optimization solutions provided in steps S131 to S132, step S131 solves the "pattern ambiguity" problem caused by data mixing by cutting high-density clusters; by removing sparse clusters, it eliminates data interference caused by noise and outliers, removing negative factors affecting grouping quality and providing a data foundation for constructing high-quality groups. Step S132 performs data set operations on the output of step S131, extracts new centroid features of the set, and redefines the first construction project data grouping based on the inherent core features of the extracted data. As can be seen from the above, the optimization solutions provided in steps S131 to S132 optimize the potentially defective original data structure, providing a more optimized data grouping structure. This makes the first construction project data grouping a clearer and more accurate data template reflecting the essence of different construction modes, providing high-quality data for subsequent data fusion, compliance judgment, and other operations, and improving the quality and reliability of construction engineering data.

[0054] Although most of the data in the sparse clusters obtained in step S131 is noisy, valuable data points may still exist. This leads to the loss of valuable data when the sparse clusters identify and remove noise. In step S132, after the merging operation, the center point features are directly extracted. However, before extracting the center point features, the sparse clusters have already cut and removed the possible noisy data, causing a change in the overall data distribution. Therefore, the extracted center point features may no longer accurately reflect the overall structure and characteristics of the original dataset. Thus, it is necessary to further optimize the optimization techniques provided in steps S131 to S132. Step S133, Noise data recovery of sparse clusters: Based on the exclusion threshold, outlier detection is performed on sparse clusters to identify data points in sparse clusters with a density higher than the local background density, obtain the corresponding candidate valid data, and recover the candidate valid data into the core cluster.

[0055] Step S134, Dynamic weighted center point calibration: Calculate the initial center point of each group based on the optimized dataset, use the reciprocal of the distance between the data point and the initial center point as the weight, and perform iterative calibration of the initial center point through a weighted average algorithm. When the change in the center point position between two consecutive iterations is less than the convergence threshold, output the iterative center point feature set.

[0056] Step S135, Grouping Quality Assessment and Feedback Optimization: The clustering quality of each data point in the first construction project data group is evaluated by the profile coefficient analysis method. Abnormal data points with profile coefficients below the confidence threshold in the profile coefficient analysis method are marked as pending data. The pending data are re-inputted into step S11 for iterative optimization. When the average profile coefficient in the profile coefficient analysis method reaches the preset quality standard, the final first construction project data group is output.

[0057] The optimization techniques provided in steps S133 to S135 enable the data grouping process of the first construction project to have the ability to self-correct and continuously optimize, overcoming the data loss and model bias problems that may be caused by one-time processing. In particular, they optimize the defects of simple elimination in step S131. By using local density analysis, valid data that may be misjudged is recovered from sparse clusters that are regarded as noise, ensuring the integrity of subsequent data analysis. Specifically, the data recovered in step S133 changes the data distribution. Step S134 uses iterative weighted averaging to make the final centroid feature not only based on the location of the data, but also taking into account the density and structure of its distribution, so as to more robustly represent the optimized data grouping.

[0058] like Figure 3 As shown, preferably, in step S3, the first construction project data group is compared and analyzed with the preset construction benchmark indicators, the construction compliance coefficient is calculated based on the second construction project data group, and the comprehensive compliance coefficient of construction information is calculated based on the construction compliance coefficient using a weighted statistical algorithm, further including: Step S31: Extract key construction process indicators for each sub-project from the second construction project data group, establish a mapping relationship table between actual indicators and benchmark indicators, perform quantitative analysis on each mapping relationship in the mapping relationship table, and calculate the absolute deviation and relative deviation rate of key construction process indicators relative to the benchmark value.

[0059] Step S32: Calculate the compliance coefficient of each mapping relationship based on the absolute deviation and relative deviation rate, and generate a set of construction compliance coefficients according to the influencing factors of different key construction process indicators.

[0060] Step S33: Based on the analytic hierarchy process, determine the importance weights of different elements in the construction compliance coefficient set. Then, use the weighted summation formula to multiply the compliance coefficients by their corresponding weight values ​​and sum them up to obtain the weighted total score.

[0061] Step S34: Convert the weighted total score into a standardized value in the range of 0-1 to generate a comprehensive compliance coefficient of construction information that characterizes the overall construction quality.

[0062] Step S31 calculates the absolute deviation and relative deviation rate, providing precise data descriptions for key construction process indicators, preventing ambiguous data descriptions, and providing quantifiable inputs for subsequent steps, ensuring that compliance judgments are based on solid objective data. Step S32 transforms the output data from Step S31 into unified individual compliance coefficients using predefined functions or rules, completing the transformation from raw data to evaluation scores. Step S33 determines the weight of each indicator in the overall evaluation based on the analytic hierarchy process (AHP), and obtains the final comprehensive compliance coefficient for construction information through weighted fusion. This comprehensive compliance coefficient accurately reflects which aspects of compliance during the construction process are performed well and which are poorly, assessing the overall impact of compliance on overall quality. Step S34 fixes the weighted total score calculated in step S33 within the range of 0-1, making the comprehensive compliance coefficient of construction information a standardized, dimensionless management KPI. Even non-professionals without professional knowledge of construction do not need to fully understand the technical parameters related to the technical details of construction. They can quickly and accurately grasp the overall construction quality status of the project simply by looking at the magnitude and trend of the comprehensive compliance coefficient of construction information.

[0063] The analytic hierarchy process (AHP) in step S31 is a multi-index decision analysis method that decomposes a complex decision problem into different hierarchical structures, determines the relative importance of different elements in each level by pairwise comparison, and calculates their respective weights using mathematical methods.

[0064] First, the construction compliance assessment issue needs to be hierarchically structured. For example, the target layer (highest layer) is the comprehensive compliance coefficient of construction information, the criterion layer (middle layer) is the key construction process indicators such as construction cycle, material consumption, labor efficiency, and equipment utilization rate, and the scheme layer (lowest layer) is the specific actual measurement value and benchmark value of each indicator.

[0065] For the indicators at the aforementioned criterion level, the relative importance of each pair of indicators is compared pairwise through expert evaluation or system preset. The comparison uses a 1-9 scale to convert subjective judgments into specific numerical values, forming a judgment matrix. Then, the weight vector is calculated and a consistency check (CR) is performed. If CR < 0.1, the consistency of the judgment matrix is ​​considered acceptable, and the calculated weights are reasonable; otherwise, the judgment matrix needs to be adjusted.

[0066] For example, in step S33, the importance weights of different elements in the construction compliance coefficient set are determined based on the analytic hierarchy process (AHP). The compliance coefficients are then multiplied by their corresponding weight values ​​using a weighted summation formula, and the sums are accumulated to obtain the weighted total score. This further includes: Step S331: By constructing a judgment matrix, the compliance coefficients of construction cycle, material consumption, labor efficiency and equipment utilization rate in the set are compared pairwise for importance. The elements in the judgment matrix are assigned values ​​according to the 1-9 scale method.

[0067] Step S332: Solve for the largest eigenvalue of the judgment matrix and its corresponding normalized eigenvector, calculate the consistency ratio of the judgment matrix, and when the consistency ratio is less than 0.1, confirm the normalized eigenvector as the indicator importance weight of the compliance coefficient.

[0068] In the optimization scheme provided in steps S331 to S332, step S331 constructs a judgment matrix. Through a pairwise comparison using a 1-9 scale, qualitative judgments in the construction process are transformed into a structured mathematical matrix, achieving quantitative transformation. This provides structured input for subsequent precise calculations, ensuring the effectiveness and reliability of the weights. Steps S332 and S331 form a collaborative calculation-verification relationship. Using mathematical methods for solving eigenvectors, preliminary weight vectors are extracted from the judgment matrix provided in step S331. Then, a consistency ratio test is introduced. This pairwise comparison process overcomes, to some extent, the logical contradictions that may exist in the construction data processing process.

[0069] For example, consider three variables A, B, and C: A is the construction period, B is material consumption, and C is labor efficiency. These three variables are key indicators in construction data. Existing technologies typically use equal weighting or subjectively assigned weights, making it difficult to accurately reflect the relative importance of A, B, and C in actual projects. To improve this, in a certain project, experts in the field of construction were invited to conduct pairwise comparisons of the three indicators using the 1-9 scale. Step 1, construct the judgment matrix: Construction period (A) is significantly more important than material consumption (B), so it is assigned a value of 5. Material consumption (B) is slightly more important than labor efficiency (C), so it is assigned a value of 3; Construction cycle (A) is significantly more important than labor efficiency (C), so it is assigned a value of 7; ABC A 1 5 7 B 1 / 5 1 3 C 1 / 7 1 / 3 1 Step 2, perform consistency checks and weight calculations: The maximum eigenvalue of the judgment matrix is ​​calculated to be λ_max = 3.065, and the corresponding eigenvector is [0.811, 0.267, 0.121]^T.

[0070] Calculate the consistency index: CI = (λ_max - n) / (n - 1) = (3.065 - 3) / 2 = 0.0325 Query the random consistency index RI (RI=0.58 when n=3) and calculate the consistency ratio: CR = CI / RI = 0.0325 / 0.58 = 0.056 < 0.1 Step 3: Examples of logical contradictions If an expert makes a logical contradiction in their judgment, for example: Construction period (A) is more important than material consumption (B) (assign a value of 5). Material consumption (B) is more important than labor efficiency (C) (assigned a value of 3). However, labor efficiency (C) is more important than construction period (A) (assigned value 2). At this point, the calculated consistency ratio CR = 0.318 > 0.1. The optimized technical solution will prompt experts to re-evaluate the judgment matrix to ensure logical consistency.

[0071] After reassessment, the new weights were determined as follows: construction period 0.68, material consumption 0.22, and labor efficiency 0.10. Implementing the project according to the new weights yielded significantly better results than the old weights.

[0072] As can be seen from the above examples, the optimization solutions provided in steps S331 to S332 effectively avoid potential data logic contradictions in construction projects through judgment matrices and consistency checks, ensuring the scientific and reliable allocation of weights and accurately reflecting the relative importance of each indicator in actual projects.

[0073] As a further improvement, in step S332, the importance weight of the indicator as the compliance coefficient is confirmed by the normalized feature vector. Specifically, after obtaining the importance weight of the indicator, each compliance coefficient is multiplied by its corresponding importance weight using the weighted summation formula and then summed to obtain the weighted total score.

[0074] In construction engineering practice, it has been found that the technical aspects of step S332 may have some deficiencies. For example, it may not consider objective evidence such as monitoring data and historical compliance records generated during actual construction, leading to a disconnect between the weight allocation and the actual project situation. To prevent potential problems arising from the above issues, this application provides a further optimization and improvement scheme for step S332: Step S333: Obtain multi-source monitoring data during the construction process; standardize the construction cycle, material consumption, labor efficiency, and equipment operating parameters through the data preprocessing module to generate a normalized data matrix; calculate the covariance matrix based on the normalized matrix and perform eigenvalue decomposition to obtain a set of eigenvalues ​​and eigenvectors; use a semimetric analysis algorithm to calculate the correlation distance between indicators and automatically select key evaluation indicators based on preset thresholds.

[0075] Step S334: The key indicators after screening are sorted by importance using a sorting algorithm, and an importance ratio sequence between adjacent indicators is established. The first weight coefficient set is obtained by recursive calculation based on the benchmark ratio. The judgment matrix is ​​constructed using the analytic hierarchy process. The second weight coefficient set is obtained through eigenvalue calculation and consistency verification. The first weight coefficient set and the second weight coefficient set are merged to generate a comprehensive weight set.

[0076] Step S335: Verify the stability of the comprehensive weight set, adjust the fusion parameters and observe the magnitude of weight changes; store the verified weight set in the weight database for the construction compliance assessment system to call and perform weighted calculations, and output the comprehensive evaluation result of construction compliance.

[0077] The optimization solutions provided in steps S333 to S335, through the synergy of objective data analysis and subjective experience ranking, form an important weight allocation scheme. This improved important weight allocation scheme possesses both mathematical rigor and conforms to practical engineering operations, making it widely applicable to various construction projects. Specifically, step S333 uses eigenvalue decomposition and semi-metric analysis based on multi-source monitoring data to screen key evaluation indicators, providing a clean and crucial data foundation for subsequent weight allocation. Step S334 uses a ranking algorithm to generate a first set of weight coefficients, and simultaneously uses the analytic hierarchy process (AHP) to generate a second set of weight coefficients, merging the two to generate a comprehensive weight set, achieving mutual verification and supplementation between objective data and subjective experience. Step S335 verifies the stability of the merged weights, ensuring the reliability of the weight output. This achieves the goal of constructing a scientific, stable, and practically aligned indicator weight system, significantly improving the accuracy and credibility of the comprehensive evaluation results for construction compliance. It solves the problem of strong subjectivity and susceptibility to bias caused by relying solely on expert subjective experience for weight allocation, preventing distortion of evaluation results and neglect of key risk indicators due to the inability to dynamically reflect actual data relationships.

[0078] like Figure 4As shown, preferably, in step S4, the compliance data characteristics of the comprehensive compliance coefficient of construction information are analyzed through the cloud platform. Based on these compliance data characteristics, the corresponding key construction data is obtained. The computing resource requirements of the key construction data are calculated, and the network transmission delay distance between different server nodes in the cloud platform is obtained. Combining the remaining processing capacity of the service nodes with the computing resource requirements, the optimal processing node is selected from running and idle service nodes. This further includes: Step S41 involves using a resource monitoring agent deployed on each server node of the cloud platform to collect dynamic operating parameters of each node in real time, generating a multi-dimensional resource vector for each node, and forming a real-time resource profile of all nodes on the platform. For example, the dynamic parameters of each node include CPU core utilization, available memory capacity, disk I / O throughput, GPU memory usage, and network bandwidth usage.

[0079] Step S42 involves performing feature analysis on the key construction data to be processed, identifying the corresponding data type, data scale, and required processing algorithm. Based on the algorithm complexity model and data volume, the required computing resources for the data processing task are calculated. Optimal processing nodes are then selected based on the obtained real-time resource profile and the required computing resources. The required resources include estimated CPU computing power, memory usage, storage space, and processing latency requirements.

[0080] Step S43: After determining the optimal processing node, dynamic scheduling of computing resources is performed through the container orchestration engine. A monitoring agent continuously tracks the node's resource usage and task execution progress. When insufficient node resources or task execution anomalies are detected, a container migration mechanism is triggered in real time to reschedule the task to another suitable node. In this step, Docker container technology can be used to package key construction data processing tasks and their runtime environment into independent container images, and the container instances can be deployed to the selected optimal processing node using the Kubernetes scheduler.

[0081] Step S44: After the task is completed, release the allocated container resources and update the node resource information in the resource status database to complete the entire resource scheduling lifecycle management.

[0082] The optimization solutions provided in steps S41 to S43 are based on a cloud platform for dynamic scheduling of key construction data. Through closed-loop management of real-time resource monitoring, demand forecasting, containerized scheduling, and elastic migration, resource allocation for key construction data processing tasks is achieved. Step S41 creates a real-time resource profile of all platform nodes. Step S42 calculates task resource requirements, optimizing resource utilization and improving task efficiency by matching task requirements with node capabilities, avoiding resource idleness or overload. Step S43 dynamically schedules resources, using a container orchestration engine to deploy tasks to the optimal nodes, significantly reducing task queuing time and resource contention, and improving overall processing efficiency. Step S44 immediately releases resources and updates the database upon task completion, preventing scheduling errors caused by lagging resource status, ensuring that long-running key construction tasks are not affected by single points of failure, and guaranteeing data processing continuity.

[0083] For example, in steps S42 and S43, the Kubernetes scheduler's decision is controlled based on the resource requirements (such as the memory usage of nodes and GPU memory) estimated by the algorithm complexity model, ensuring that the container instances meet the requirements of building construction, so that the cloud platform for building construction projects can simultaneously adapt to heterogeneous tasks such as computing-intensive and storage-intensive tasks, realize elastic scaling across resource dimensions, and support diverse scenarios of complex construction data processing.

[0084] As a further improvement, in step S42, the optimal processing node is selected based on the obtained real-time resource profile and computing resource demand capacity, which further includes: Step S421: Set constraints for node selection. The constraints include: the remaining memory of the node is greater than the memory usage of the task; the available CPU computing power of the node is greater than the estimated CPU computing power of the task; and the network transmission delay is lower than the maximum allowable delay of the task.

[0085] Step S422: The candidate nodes selected through the constraints are comprehensively evaluated using a weighted scoring algorithm. Based on the matching degree between the node's remaining computing resources and the task's required resources, the network transmission delay from the node to the data source, and the node's current load balancing degree, the comprehensive adaptation score of each candidate node is calculated using preset weight coefficients. The node with the highest comprehensive adaptation score is selected as the optimal processing node.

[0086] The optimization solutions provided in steps S421 and S422 use hard constraints to initially screen nodes, ensuring the feasibility of selecting the optimal processing node. Then, a weighted comprehensive evaluation is performed based on multi-dimensional matching degrees to optimize resource allocation. This optimization process combines rigid requirements with flexible judgments, forming a progressive screening logic from qualified to optimal. Step S421, as a basic guarantee layer, excludes nodes that cannot meet the basic requirements of the task, avoiding task failure due to insufficient resources. Step S422, based on step S421, performs weighted scoring to achieve further optimization. This process of constraint-then-scoring ensures that the selected nodes meet the minimum resource requirements of the task, while multi-dimensional evaluation prevents nodes that only meet the minimum requirements but have poor actual performance from being selected, improving the reliability of construction task execution. In step S422, the pursuit is no longer for the optimal single indicator, but for finding a comprehensive optimal solution. For example, a node is not selected simply because it has the most idle CPU, but its network status and current load are evaluated simultaneously to avoid creating new performance bottlenecks and achieve a balance between individual task performance and overall cluster performance.

[0087] like Figure 5 As shown, preferably, in step S5, the load balancing deviation of the optimal processing node is calculated, and a dynamic allocation scheme for computing resources is implemented based on the load balancing deviation. The corresponding construction information processing template is retrieved from the construction storage database, and the construction data of the construction project is tagged according to the template and transmitted to the designated construction supervision terminal. This further includes: Step S51: The resource utilization data of each node is collected in real time by the monitoring agent deployed on the cloud platform node. The load balance deviation of the cluster is calculated based on the collected data. When the deviation exceeds the preset threshold, the computing tasks on the load node are scheduled to the low load node, and additional virtualized computing resources are allocated to the overload node.

[0088] Step S52: Prioritize the construction data processing tasks input to the cloud platform using the task scheduler; based on task priority and resource requirement characteristics, establish a task scheduling strategy model using a reinforcement learning algorithm based on deep Q-networks, detect the matching degree between tasks and nodes, make corresponding optimization scheduling decisions, and allocate computationally intensive tasks to high-performance servers.

[0089] Step S53 involves retrieving the construction information processing template from the database via the metadata definition module. This template defines data structuring specifications and processing rules, parses the raw construction data according to the template specifications, extracts key information, adds type tags, and generates a standardized construction information data package. This package is then asynchronously transmitted to the designated construction supervision terminal via a message queue. In step S53, key information refers to core data elements extracted from the raw data stream that have direct decision-making support value for construction management. These are typically cleaned and parsed signals, such as numerical status signals (progress deviation rate, concrete strength, today's safe working hours), event alarm signals (deep foundation pit displacement exceeding limit alarm signal, main material inventory early warning signal), compliance judgment signals (welding operation compliance, results of random checks on high-altitude workers' certification), etc. Type tags are used to classify key information; they do not contain specific numerical values ​​but indicate the management dimension, professional field, or processing priority to which the key information belongs. Examples could be: labels categorized by management dimensions (progress, cost, quality, safety, resources), labels categorized by professional fields (civil engineering, steel structure, electromechanical, curtain wall, decoration), labels categorized by processing priority (urgent, important, general, reference), and labels categorized by data nature (measured data, approval process, alarm events, statistical analysis).

[0090] The optimization solutions provided in steps S51 to S53 achieve efficient processing and reliable transmission of high-concurrency, high-computational-resource-consuming construction data through the synergistic effect of load balancing monitoring, intelligent task scheduling, and data standardization. They optimize three aspects: dynamic resource allocation, intelligent task decision-making, and data standardization, ensuring system stability and balanced resource utilization, and improving the task processing efficiency and intelligent scheduling level of construction projects. Step S51 monitors load balancing deviation, providing a global perspective on resource distribution for the system. Dynamic scheduling is triggered when uneven load is detected. Step S52 uses a task scheduler to perform refined resource allocation based on priority and resource demand characteristics. A reinforcement learning algorithm based on a deep Q-network comprehensively considers multiple factors such as task priority, resource demand characteristics, and node status, continuously learning and optimizing the scheduling strategy to achieve the best match between tasks and nodes. Step S53 generates standardized construction information data packets through templated data parsing and tokenization. Combined with the asynchronous transmission mechanism of a message queue, this ensures both the consistency of data processing quality and the reliability of data transmission, meeting the dual requirements of the monitoring terminal for data accuracy and timeliness.

[0091] For the method steps disclosed in the above embodiments, the method steps are described as a series of actions for the purpose of simplicity. However, those skilled in the art should understand that the embodiments of this application are not limited to the described order of actions, because according to the embodiments of this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily necessary for the embodiments of this application.

[0092] like Figure 6 As shown, this application embodiment, based on the cloud platform-based construction project construction data management method, also provides a corresponding cloud platform-based construction project construction data management system to implement the cloud platform-based construction project construction data management method, including: The first construction project data grouping generation module acquires construction data of building projects, constructs a multi-source construction element dataset, extracts multi-dimensional features from the multi-source construction element dataset, generates a corresponding construction project feature list, and groups the construction data of building projects based on the construction project feature list using a clustering calculation method to form the first construction project data group. The second construction project data grouping generation module determines the feature matching degree between the target construction project data and the reference construction project data. When the feature matching degree reaches the preset fusion threshold, the target construction project data is fused into the reference construction project data to form the second construction project data group. The construction information comprehensive compliance coefficient generation module compares and analyzes the first construction project data group with the preset construction benchmark indicators, calculates the construction compliance coefficient based on the second construction project data group, and calculates the construction information comprehensive compliance coefficient based on the construction compliance coefficient through a weighted statistical algorithm. The optimal processing node selection module analyzes the compliance data characteristics of the comprehensive compliance coefficient of construction information through the cloud platform, obtains the corresponding key construction data based on the compliance data characteristics, calculates the computing resource demand capacity of the key construction data, obtains the network transmission delay distance between different server nodes in the cloud platform, and selects the optimal processing node from running service nodes and idle service nodes by combining the remaining processing capacity of the service node and the computing resource demand capacity. The load balancing reliability calculation module calculates the load balancing deviation of the optimal processing node, implements a dynamic allocation scheme for computing resources based on the load balancing deviation, retrieves the corresponding construction information processing template from the construction storage database, marks the construction data of the construction project according to the template, and transmits it to the designated construction supervision terminal.

[0093] The implementation methods of the system described above are merely illustrative. For example, the various functional modules, units, or subsystems within the system may or may not be physically separate, or they may or may not be physical units; that is, they may be located in the same place or distributed across multiple different systems and their subsystems or modules. Those skilled in the art can select some or all of the functional modules, units, or subsystems to achieve the objectives of the embodiments of this application according to actual needs. Those skilled in the art can understand and implement the above-described situations without any creative effort.

[0094] In the description of the embodiments of this application, the reference to terms such as "an embodiment," "example," "specific example," etc., means that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the embodiments of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0095] All features disclosed in the embodiments of this application, or all steps in the disclosed methods or processes, may be combined in any way, except for mutually exclusive features and / or steps. Any feature disclosed in the specification of the embodiments of this application, unless specifically stated otherwise, may be replaced by other equivalent or similar alternative features. That is, unless specifically stated otherwise, each feature is merely one example of a series of equivalent or similar features. Throughout the specification, the same reference numerals indicate the same elements.

[0096] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification of embodiments (including the corresponding claims, abstract, and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, each feature disclosed in this specification of embodiments (including the corresponding claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0097] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of the specification of the embodiments of this application.

[0098] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of this application, and are not intended to limit them. Although the embodiments of this application have been described in detail with reference to the foregoing specific embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein, and such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the specific embodiments of this application.

Claims

1. A method for managing construction data of building projects based on a cloud platform, characterized in that, include: Obtain construction data of building projects, construct a multi-source construction element dataset, extract multi-dimensional features from the multi-source construction element dataset, generate a corresponding construction project feature list, and group the construction data of building projects based on the construction project feature list using a clustering calculation method to form a first construction project data group. Determine the feature matching degree between the target construction project data and the reference construction project data. When the feature matching degree reaches a preset fusion threshold, merge the target construction project data into the reference construction project data to form a second construction project data group. The first construction project data group is compared and analyzed with the preset construction benchmark indicators. The construction compliance coefficient is calculated based on the second construction project data group. The comprehensive compliance coefficient of construction information is calculated based on the construction compliance coefficient through a weighted statistical algorithm. The cloud platform analyzes the compliance data characteristics of the comprehensive compliance coefficient of construction information, obtains the corresponding key construction data based on the compliance data characteristics, calculates the computing resource demand capacity of the key construction data, obtains the network transmission delay distance between different server nodes in the cloud platform, and selects the optimal processing node from running service nodes and idle service nodes by combining the remaining processing capacity of service nodes and the computing resource demand capacity. Calculate the load balancing deviation of the optimal processing node, implement a dynamic allocation scheme for computing resources based on the load balancing deviation, call the corresponding construction information processing template from the construction storage database, mark the construction data of the construction project according to the template, and transmit it to the designated construction supervision terminal.

2. The construction data management method for building engineering based on a cloud platform according to claim 1, characterized in that, The process of acquiring construction project data, constructing a multi-source construction element dataset, extracting multi-dimensional features from the multi-source construction element dataset to generate a corresponding construction project feature list, and grouping the construction project data using a clustering method based on the construction project feature list to form a first construction project data group, further includes: The weight coefficients of the multidimensional features in the multi-source construction element dataset are calculated. The weight coefficients are combined with the weighted feature fusion algorithm to generate the optimized feature vectors of each construction sub-project. The optimized feature vectors are then subjected to preliminary clustering calculations to obtain the initial clustering dataset and the set of sample numbers of each sub-class in the initial clustering dataset. Calculate the 25th percentile, 75th percentile, and mean point of the sample set, determine the corresponding high-density cluster judgment threshold and sparse cluster judgment threshold, and divide the initial clustering dataset into high-density clusters, sparse clusters, and core clusters. Obtain the cutting threshold for high-density clusters and the exclusion threshold for sparse clusters. Based on the cutting threshold and the exclusion threshold, merge the high-density clusters and the core clusters to generate the first construction project data group.

3. The construction data management method for building engineering based on a cloud platform according to claim 2, characterized in that, The process involves obtaining the cutting threshold for high-density clusters and the exclusion threshold for sparse clusters, and then merging the high-density clusters and the core clusters based on these thresholds to generate the first construction project data group. Specifically: Obtain the cutting threshold for high-density clusters and the exclusion threshold for sparse clusters. Perform segmentation processing on high-density clusters based on the cutting thresholds and remove sparse clusters based on the exclusion thresholds. The high-density clusters after segmentation are merged with the core clusters after removal. The centroid features after merging are extracted, and the construction data of the building project are grouped according to the centroid features to generate the first construction project data group.

4. The construction data management method for building engineering based on a cloud platform according to claim 1, characterized in that, The step of comparing and analyzing the first construction project data group with preset construction benchmark indicators, calculating the construction compliance coefficient based on the second construction project data group, and calculating the comprehensive compliance coefficient of construction information based on the construction compliance coefficient using a weighted statistical algorithm, further includes: Key construction process indicators for each sub-project are extracted from the data group of the second construction project. A mapping relationship table between actual indicators and benchmark indicators is established. Each mapping relationship in the mapping relationship table is quantitatively analyzed, and the absolute deviation and relative deviation rate of the key construction process indicators relative to the benchmark values ​​are calculated. The compliance coefficient for each mapping relationship is calculated based on the absolute deviation and the relative deviation rate, and a set of construction compliance coefficients is generated according to the influencing factors of different key construction process indicators. The importance weights of different elements in the construction compliance coefficient set are determined based on the analytic hierarchy process. The compliance coefficients are multiplied by their corresponding weight values ​​and then summed using a weighted summation formula to obtain the weighted total score. The weighted total score is converted into a standardized value in the range of 0-1 to generate a comprehensive compliance coefficient of construction information that characterizes the overall construction quality.

5. The cloud-based construction data management method for building projects according to claim 4, characterized in that, The method of determining the importance weights of different elements in the construction compliance coefficient set based on the analytic hierarchy process (AHP), and then multiplying the compliance coefficients by their corresponding weight values ​​using a weighted summation formula to obtain a weighted total score, further includes: By constructing a judgment matrix, the compliance coefficients of construction cycle, material consumption, labor efficiency and equipment utilization rate in the set are compared pairwise in terms of importance. The elements in the judgment matrix are assigned values ​​according to the 1-9 scale method. Find the largest eigenvalue of the judgment matrix and its corresponding normalized eigenvector, calculate the consistency ratio of the judgment matrix, and when the consistency ratio is less than 0.1, confirm that the normalized eigenvector is the indicator importance weight of the compliance coefficient.

6. The construction data management method for building engineering based on a cloud platform according to claim 5, characterized in that, The confirmation of the normalized feature vector as the indicator importance weight of the compliance coefficient is specifically as follows: after obtaining the indicator importance weight, the weighted summation formula is used to multiply each compliance coefficient by its corresponding indicator importance weight and then sum them to obtain the weighted total score.

7. The construction data management method for building engineering based on a cloud platform according to claim 1, characterized in that, The process of analyzing the compliance data characteristics of the comprehensive compliance coefficient of construction information through the cloud platform, obtaining corresponding key construction data based on the compliance data characteristics, calculating the computing resource requirements of the key construction data, obtaining the network transmission latency distance between different server nodes in the cloud platform, and selecting the optimal processing node from running and idle service nodes by combining the remaining processing capacity of the service nodes and the computing resource requirements, further includes: By deploying resource monitoring agents on each server node of the cloud platform, the dynamic operating parameters of each node are collected in real time, generating a multi-dimensional resource vector for each node, forming a real-time resource profile of all nodes on the platform. Feature analysis is performed on the key construction data that needs to be processed to identify the corresponding data type, data size, and required processing algorithm. Based on the algorithm complexity model and data volume, the computing resource requirements for the data processing task are calculated. Based on the obtained real-time resource profile and computing resource requirements, the optimal processing node is selected. After determining the optimal processing node, the container orchestration engine performs dynamic scheduling of computing resources. During task execution, the monitoring agent continuously tracks the resource usage and task execution progress of the node. When insufficient node resources or abnormal task execution is detected, the container migration mechanism is triggered in real time to reschedule the task to other suitable nodes. Once the task is completed, the allocated container resources are released, and the node resource information in the resource status database is updated, thus completing the entire resource scheduling lifecycle management.

8. The construction data management method for building engineering based on a cloud platform according to claim 7, characterized in that, The optimal processing node selection based on the obtained real-time resource profile and computing resource demand capacity further includes: Set constraints for node selection, including that the remaining memory of the node is greater than the memory usage of the task, the available CPU computing power of the node is greater than the estimated CPU computing power of the task, and the network transmission latency is lower than the maximum allowable latency of the task. A weighted scoring algorithm is used to comprehensively evaluate the candidate nodes selected through constraints. Based on the matching degree between the node's remaining computing resources and the task's required resources, the network transmission latency from the node to the data source, and the node's current load balancing, a comprehensive adaptation score for each candidate node is calculated using preset weight coefficients. The node with the highest comprehensive adaptation score is selected as the optimal processing node.

9. The construction data management method for building engineering based on a cloud platform according to claim 1, characterized in that, The calculation of the load balancing deviation of the optimal processing node, the implementation of a dynamic allocation scheme for computing resources based on the load balancing deviation, the retrieval of the corresponding construction information processing template from the construction storage database, the marking of the construction data of the construction project according to the template, and the transmission to the designated construction supervision terminal, further includes: By collecting resource utilization data of each node in real time through monitoring agents deployed on cloud platform nodes, the load balance deviation of the cluster is calculated based on the collected data. When the deviation exceeds a preset threshold, the computing tasks on the overloaded nodes are scheduled to low-load nodes, and additional virtualized computing resources are allocated to the overloaded nodes. The task scheduler prioritizes the construction data processing tasks input to the cloud platform. Based on task priority and resource requirement characteristics, a task scheduling strategy model is established using a reinforcement learning algorithm based on deep Q-networks. The matching degree between tasks and nodes is detected, and corresponding optimization scheduling decisions are made to allocate computationally intensive tasks to high-performance servers. The construction information processing template is called from the database through the metadata definition module. The construction information processing template is used to define the data structuring specifications and processing rules, parse the original construction data according to the template specifications, extract key information and add type tags, generate standardized construction information data packets, and transmit them asynchronously to the designated construction supervision terminal through a message queue.

10. A cloud-based construction project data management system, used to implement the cloud-based construction project data management method according to any one of claims 1 to 9, characterized in that, include: The first construction project data grouping generation module acquires construction data of building projects, constructs a multi-source construction element dataset, extracts multi-dimensional features from the multi-source construction element dataset, generates a corresponding construction project feature list, and groups the construction data of building projects based on the construction project feature list using a clustering calculation method to form the first construction project data group. The second construction project data grouping generation module determines the feature matching degree between the target construction project data and the reference construction project data. When the feature matching degree reaches a preset fusion threshold, the target construction project data is fused into the reference construction project data to form the second construction project data group. The construction information comprehensive compliance coefficient generation module compares and analyzes the first construction project data group with the preset construction benchmark indicators, calculates the construction compliance coefficient based on the second construction project data group, and calculates the construction information comprehensive compliance coefficient based on the construction compliance coefficient through a weighted statistical algorithm. The optimal processing node selection module analyzes the compliance data characteristics of the comprehensive compliance coefficient of construction information through the cloud platform, obtains the corresponding key construction data based on the compliance data characteristics, calculates the computing resource requirement capacity of the key construction data, obtains the network transmission delay distance between different server nodes in the cloud platform, and selects the optimal processing node from running service nodes and idle service nodes by combining the remaining processing capacity of the service node and the computing resource requirement capacity. The load balancing reliability calculation module calculates the load balancing deviation of the optimal processing node, implements a dynamic allocation scheme for computing resources based on the load balancing deviation, retrieves the corresponding construction information processing template from the construction storage database, marks the construction data of the construction project according to the template, and transmits it to the designated construction supervision terminal.

Images