A construction project quality monitoring method and system based on big data

Abstract

The present invention relates to the field of quality monitoring technology, specifically to a method and system for monitoring the quality of construction projects based on big data, comprising the following steps: obtaining the pressure, temperature and wind speed data of the building structure in real time through sensors, adding timestamp information to the data, and performing time alignment on multiple types of sensor data through time synchronization technology. In the present invention, by synchronously monitoring the construction and real-time updating of data sets, detailed monitoring of the quality of construction projects is provided, and by dynamically analyzing the changes in node strength and connection mode, it effectively adapts to changes in the environment and load, enhances the adaptability of structural configuration, and utilizes the automatic update mechanism of conditional probability tables and dynamic probability models to achieve rapid prediction and response to potential problems in construction projects, improve prediction accuracy and decision-making efficiency, and in addition, by applying multi-core learning methods, it is possible to more accurately analyze and predict structural behavior in high-dimensional space, thereby improving the safety of construction projects.

Description

Technical Field

[0001] The present invention relates to the technical field of quality monitoring, and in particular to a construction project quality monitoring method and system based on big data. Background Art

[0002] The field of quality monitoring technology covers a range of methods and systems designed to ensure that a product, service or process meets predetermined quality standards. In the field of construction engineering, this usually involves real-time monitoring of the quality of the building structure through various monitoring tools and technologies in order to detect problems and make adjustments in a timely manner. These technologies may include the use of sensors to collect data from the structure (such as temperature, humidity, crack development, etc.) and send the data to a central processing system for analysis. In addition, quality monitoring also involves the inspection of materials, construction processes and final results to ensure that all links meet regulatory and standard requirements. This not only improves the safety and durability of the building, but also helps to improve the overall efficiency and cost-effectiveness of the construction project.

[0003] Among them, the construction project quality monitoring method refers to the use of scientific and technological means (such as sensors, data analysis and predictive models) to track and evaluate various quality parameters in the construction process. The main purpose of this monitoring is to ensure that the construction project meets the design specifications, prevent potential structural problems, and ensure the long-term stability and safety of the building. Through these methods, engineers and project managers can obtain detailed information about building materials, structures and construction processes in real time, and then quickly take corrective measures when necessary to prevent the expansion of problems or future quality defects.

[0004] Existing technologies are deficient in data integration and real-time response, especially when monitoring large-scale construction projects, where data processing often faces synchronization and timeliness issues. This makes it difficult for existing systems to achieve timely data analysis and structural adjustments for rapidly changing environmental factors and structural loads. The lack of efficient dynamic data analysis methods, such as multi-kernel learning and dynamic probability models, limits the adaptability and prediction accuracy of construction projects under complex conditions. These technical deficiencies result in insufficient performance of monitoring systems in real-time data processing and structural behavior prediction, increasing potential safety risks and maintenance costs. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a construction project quality monitoring method and system based on big data.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a construction project quality monitoring method based on big data, comprising the following steps:

[0007] S1: Obtain the pressure, temperature and wind speed data of the building structure in real time through sensors, add timestamp information to the data, and use time synchronization technology to time align multiple types of sensor data to obtain a synchronized monitoring data set;

[0008] S2: Analyze the data changes in the synchronous monitoring data set, adjust the node strength according to the load changes, adjust the structure of the connection nodes according to the environmental factors including temperature and wind speed, and use statistical analysis methods to evaluate the mutual influence between differentiated nodes, reconfigure the priorities between nodes, dynamically generate a structural configuration that matches the current data, and generate an optimized network structure;

[0009] S3: using the optimized network structure, monitoring current construction project data, analyzing the interaction between nodes, updating the probability relationship between nodes according to the interaction data, automatically integrating the data into a conditional probability table, updating the data through the automatic update mechanism of the conditional probability table, and establishing a dynamic probability model;

[0010] S4: Probabilistic inference is performed based on the dynamic probability model to predict the future stress state in the structure, analyze the propagation direction of stress in the structure, and obtain stress prediction and propagation analysis results by calculating the key propagation path;

[0011] S5: Processing the stress prediction and propagation analysis results, analyzing data in a high-dimensional space using a multi-kernel learning method, adjusting parameters of multiple kernels to match data features, and generating structural behavior analysis results after kernel parameter adjustment.

[0012] The synchronous monitoring data set specifically includes pressure data, temperature data, and wind speed data; the optimized network structure includes node strength, connection mode, and node priority; the dynamic probability model includes the probability relationship between nodes, the conditional probability table, and the data update mechanism; the stress prediction and propagation analysis results include the future stress state, stress propagation direction, and key propagation path; the structural behavior analysis results include the application of learning methods, data analysis dimensions, and kernel parameter adjustment.

[0013] As a further solution of the present invention, the step of acquiring the synchronous monitoring data set is specifically:

[0014] S111: collecting pressure data, temperature data and wind speed data through sensors installed on key structures of the building project, adding a timestamp to each data point, and obtaining a time-stamped original data set;

[0015] S112: Applying a time alignment algorithm to process the time-tagged original data set, aligning the pressure, temperature, and wind speed data on the same time axis by adjusting the data timestamps, and obtaining a time-aligned sensor data set;

[0016] S113: Analyze the time-aligned sensor data set and calculate the monitoring index of all sensor data at each time point using the formula:

[0017] ;

[0018] Get comprehensive monitoring data set ;

[0019] in, Represents the pressure data value at a single time point, Represents the temperature data value at a single point in time, Represents the wind speed data value at a single time point, is the weight coefficient of pressure data, which is used to adjust the influence of pressure data in the comprehensive monitoring indicators. is the weight coefficient of temperature data, which is used to adjust the influence of temperature data in the comprehensive monitoring indicators. is the weight coefficient of wind speed data, which is used to adjust the influence of wind speed data in the comprehensive monitoring indicators;

[0020] S114: Integrate and format the comprehensive monitoring data set to generate a synchronous monitoring data set.

[0021] As a further solution of the present invention, the step of obtaining the optimized network structure is specifically as follows:

[0022] S211: collecting the node load data in the synchronous monitoring data set, recording the load condition of each node at a differentiated time point, and generating time series load data of each node;

[0023] S212: Analyze the load fluctuation pattern of each node from the time series load data of each node, calculate the standard deviation and mean of the node load, and generate node load statistics;

[0024] S213: Based on the node load statistics, dynamically adjust the processing capacity and connection strength of each node according to its load condition, using the formula:

[0025] ;

[0026] Generate adjusted node configuration;

[0027] in, Representative Node Adjusted processing power and connection strength, Representative Node The real-time load, represents the maximum load among all nodes during the observation period, Representative Node Raw processing power, This is to avoid the small adjustment factor introduced when the denominator is zero and increase the stability of the calculation. Adjustment factor used to enhance node processing capability;

[0028] S214: Apply the adjusted node configuration to the entire network structure, dynamically match load changes according to real-time data, optimize the connection and processing capabilities of each node in the network, and generate an optimized network structure.

[0029] As a further solution of the present invention, the step of acquiring the dynamic probability model is specifically as follows:

[0030] S311: using the optimized network structure to monitor data of the current construction project, analyzing interactions between nodes, and obtaining a preliminary node interaction data set;

[0031] S312: updating the probability relationship between nodes according to the node interaction data in the preliminary node interaction data set, calculating and adjusting the statistical probability value of the node relationship, and generating an updated node probability relationship matrix;

[0032] S313: Automatically integrate the updated node probability relationship matrix into the conditional probability table using the formula:

[0033] ;

[0034] Update the conditional probability table, dynamically predict node behavior and system status based on real-time data, and generate a dynamic probability model;

[0035] in, It represents the conditional probability of event A occurring under the condition that event B occurs, reflecting the predictability of event A occurring when event B is known to occur. represents the marginal probability of event A, represents the marginal probability of event B, It represents the conditional probability of event B occurring under the condition that event A occurs, which indicates the probability of event B occurring when event A has already occurred.

[0036] As a further solution of the present invention, the steps of obtaining the stress prediction and propagation analysis results are specifically as follows:

[0037] S411: performing probability inference based on the dynamic probability model, predicting the future stress state in the structure using a statistical analysis method, and obtaining prediction data of the future stress state;

[0038] S412: Analyze the propagation direction of stress in the structure based on the predicted data of the future stress state, and generate an analysis result of the stress propagation direction with reference to material properties and structural connections;

[0039] S413: According to the stress propagation direction analysis result, the key propagation path is calculated using the formula:

[0040] ;

[0041] Generate calculation results of key propagation paths;

[0042] in, represents the critical strength of stress propagation, represents the probability contribution of node i, is the average of the probabilities, is the standard deviation of the probability, Represents the slave node The distance to the next node, is the total number of nodes in the path;

[0043] S414: Using the critical propagation path calculation results, the structural model and the probability data are integrated to generate stress prediction and propagation analysis results.

[0044] As a further solution of the present invention, the steps of obtaining the structural behavior analysis results after the kernel parameters are adjusted are specifically:

[0045] S511: analyzing the stress prediction and propagation analysis results, processing the data in a high-dimensional space using a multi-kernel learning method, assigning preliminary kernel parameters to each data point, and generating a structural behavior analysis;

[0046] S512: Based on the structural behavior analysis result, adjust the parameters of each core to ensure that each parameter matches the characteristics of the data, adjust the weights through linear algebra and optimization algorithms, and generate a parameter adjustment plan;

[0047] S513: Apply the parameter adjustment scheme to recalculate the structural behavior using the formula:

[0048] ;

[0049] Reflecting the adjusted structural behavior through a weighted Gaussian function, generating structural behavior analysis results after kernel parameter adjustment;

[0050] in, Represents the response of the structural behavior, reflecting the prediction result obtained by the multi-kernel learning algorithm through the weighted summation of the differentiated kernel function. It is The tuning parameter of each core represents the weight of the core in the overall structural behavior. is the input variable, representing the data characteristics, It is The mean parameter of the kernel function describes the center position of the kernel corresponding to the data distribution. It is The standard deviation parameter of the kernel function indicates the degree of dispersion of the data distribution.

[0051] A construction project quality monitoring system based on big data, the construction project quality monitoring system based on big data is used to execute the above-mentioned construction project quality monitoring method based on big data, and the system comprises:

[0052] The data acquisition module obtains the pressure, temperature and wind speed data of the building structure in real time through sensors, adds timestamp information, and uses time synchronization technology to time align multi-type sensor data to generate a synchronous monitoring data set;

[0053] The network optimization module adjusts the node strength value, updates the edge connection status, evaluates the node influence value, reconfigures the node priority, and generates an optimized network structure based on the pressure change value, temperature change value, and wind speed change value in the synchronous monitoring data set;

[0054] The probability modeling module monitors the current pressure value, temperature value and wind speed value based on the optimized network structure, analyzes the interaction between nodes, updates the probability relationship between nodes, integrates the data into the conditional probability table, and establishes a dynamic probability model;

[0055] The prediction and analysis module performs probability inference based on the dynamic probability model, predicts future pressure state values, analyzes stress propagation direction values, calculates key propagation path values, obtains stress prediction and propagation analysis results, processes stress prediction and propagation analysis results, uses multi-kernel learning methods to analyze pressure values, temperature values ​​and wind speed values, adjusts kernel parameter values, and generates structural behavior analysis results.

[0056] Compared with the prior art, the advantages and positive effects of the present invention are:

[0057] In the present invention, by constructing and updating the synchronous monitoring data set in real time, a detailed monitoring of the quality of the construction project is provided. By dynamically analyzing the changes in node strength and connection mode, the changes in the environment and load are effectively adapted to enhance the adaptability of the structural configuration. By using the automatic update mechanism of the conditional probability table and the dynamic probability model, the rapid prediction and response to potential problems of the construction project are achieved, and the prediction accuracy and decision-making efficiency are improved. In addition, by applying the multi-kernel learning method, the structural behavior can be analyzed and predicted more accurately in high-dimensional space, thereby improving the safety of the construction project. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0059] Figure 2 A flowchart of the steps for acquiring a synchronous monitoring data set of the present invention;

[0060] Figure 3 A flowchart of the steps for obtaining the network structure optimized by the present invention;

[0061] Figure 4 A flowchart of the steps for obtaining the dynamic probability model of the present invention;

[0062] Figure 5 A flow chart of the steps for obtaining stress prediction and propagation analysis results of the present invention;

[0063] Figure 6 The flowchart is a step of obtaining the structural behavior analysis results after the kernel parameters are adjusted according to the present invention. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0065] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0066] Example 1: Please refer to Figure 1 The present invention provides a technical solution: a construction project quality monitoring method based on big data, comprising the following steps:

[0067] S1: Obtain the pressure, temperature and wind speed data of the building structure in real time through sensors, add timestamp information to the data, and use time synchronization technology to time align multiple types of sensor data to obtain a synchronized monitoring data set;

[0068] S2: Analyze the data changes in the synchronous monitoring data set, adjust the node strength according to the load changes, adjust the structure of the connected nodes according to environmental factors including temperature and wind speed, and use statistical analysis methods to evaluate the mutual influence between differentiated nodes, reconfigure the priorities between nodes, dynamically generate a structural configuration that matches the current data, and generate an optimized network structure;

[0069] S3: Use the optimized network structure to monitor the current construction project data, analyze the interaction between nodes, update the probability relationship between nodes based on the interaction data, automatically integrate the data into the conditional probability table, update the data through the automatic update mechanism of the conditional probability table, and establish a dynamic probability model;

[0070] S4: Probabilistic inference is performed based on the dynamic probability model to predict the future stress state in the structure, analyze the propagation direction of stress in the structure, and obtain stress prediction and propagation analysis results by calculating the key propagation path;

[0071] S5: Process the stress prediction and propagation analysis results, use the multi-kernel learning method to analyze the data in high-dimensional space, adjust the parameters of multiple kernels to match the data characteristics, and generate the structural behavior analysis results after the kernel parameters are adjusted.

[0072] The synchronous monitoring data set specifically includes pressure data, temperature data, and wind speed data. The optimized network structure includes node strength, connection method, and node priority. The dynamic probability model includes the probability relationship between nodes, conditional probability table, and data update mechanism. The stress prediction and propagation analysis results include future stress state, stress propagation direction, and key propagation path. The structural behavior analysis results include the application of learning methods, data analysis dimensions, and kernel parameter adjustment.

[0073] See also Figure 2 ,The specific steps for obtaining the synchronous monitoring data set are:

[0074] S111: collecting pressure data, temperature data and wind speed data through sensors installed on key structures of the building project, adding a timestamp to each data point, and obtaining a time-stamped original data set;

[0075] Pressure data, temperature data and wind speed data are collected by sensors installed on key structures of building projects. Timestamps are added to each data point to ensure that the data reflects the collection time. Timestamps are used to ensure the consistency of multiple data points in the time series. Timestamps are calibrated through data synchronization algorithms to obtain time-stamped original data sets. The original data sets include multiple sensor data marked with timestamps, and the data reflect the physical state of the building structure at differentiated time points.

[0076] S112: applying a time alignment algorithm to process the time-tagged original data set, aligning the pressure, temperature, and wind speed data on the same time axis by adjusting the data timestamps, and obtaining a time-aligned sensor data set;

[0077] A time alignment algorithm is applied to process the time-tagged original data set. The pressure, temperature, and wind speed data are synchronized through data calibration and time alignment technology to ensure that the data can be accurately compared and analyzed at the same time point. This step is crucial because time alignment is the basis for ensuring the accuracy of subsequent data analysis. The obtained time-aligned sensor data set includes three differentiated types of sensor data, which are compared and analyzed on the same time axis, greatly enhancing the reliability of data analysis, thereby obtaining a time-aligned sensor data set.

[0078] S113: Analyze the time-aligned sensor data set and calculate the monitoring indicators of all sensor data at each time point using the formula:

[0079] ;

[0080] Get comprehensive monitoring data set ;

[0081] in, Represents the pressure data value at a single time point, Represents the temperature data value at a single point in time, Represents the wind speed data value at a single time point, is the weight coefficient of pressure data, which is used to adjust the influence of pressure data in the comprehensive monitoring indicators. is the weight coefficient of temperature data, which is used to adjust the influence of temperature data in the comprehensive monitoring indicators. is the weight coefficient of wind speed data, which is used to adjust the influence of wind speed data in the comprehensive monitoring indicators;

[0082] formula:

[0083] ;

[0084] The benefit of the formula is that it balances the data influence of multiple sensors by reasonably setting the weight coefficient and using square root operations to obtain a comprehensive indicator, which helps reduce the impact of any single sensor failure on the overall data analysis. Detailed explanation of the formula and the formula calculation derivation process:

[0085] First, set multiple weight parameters , , The weights are 1.0, 0.8, and 1.2 respectively. The weights reflect the importance of multiple sensor data in the monitoring system. Set the pressure value at a certain time point kPa, temperature value °C, wind speed m / s. The calculation steps are as follows:

[0086] 1. ;

[0087] 2. The calculated index is ;

[0088] The results show that the comprehensive monitoring index is 29.15, which can reflect the comprehensive physical status of the building structure at the same time point, which is helpful for risk assessment and structural health monitoring.

[0089] S114: Integrate and format the comprehensive monitoring data set to generate a synchronous monitoring data set.

[0090] The comprehensive monitoring data set is integrated and finally formatted to ensure that all necessary monitoring parameters and time information are included. After strict data formatting and integration, the comprehensive monitoring data set includes all monitoring parameters collected from multiple sensors and arranged in chronological order. These data are organized in a structured data set. Such a structured arrangement makes subsequent data analysis and storage more efficient and systematic. In this way, the required information can be retrieved from batch data to generate a synchronous monitoring data set.

[0091] See also Figure 3 , the specific steps for obtaining the optimized network structure are:

[0092] S211: collecting node load data in the synchronous monitoring data set, recording the load status of each node at different time points, and generating time series load data of each node;

[0093] By recording the load conditions of each node at differentiated time points, systematically collecting node load data in the synchronous monitoring data set, and generating time series load data for each node is the first step in network structure optimization. This process involves installing high-precision sensors on network nodes and regularly monitoring and recording data. The data collection capability and accuracy of the sensors directly affect the accuracy and reliability of subsequent data analysis. The data of each node is encoded into a time series format and stored in a central database. These data are then used to analyze the load change pattern of the node and determine the overall health status of the network.

[0094] S212: analyzing the load fluctuation pattern of each node from the time series load data of each node, calculating the standard deviation and mean of the node load, and generating node load statistics;

[0095] Starting from the time series load data, use the data analysis library in MATLAB or Python to perform trend analysis and pattern recognition on the load data of each node. The key is to use the provided standard deviation and mean calculation functions to measure the stability and volatility of the node load changes. This analysis helps to discover potential network bottlenecks and load imbalance problems, thereby providing a basis for network reconfiguration. The performance data of each node will be comprehensively evaluated based on the previous load records to generate load statistics.

[0096] S213: Based on the node load statistics, dynamically adjust the processing capacity and connection strength of each node according to its load situation, using the formula:

[0097] ;

[0098] Generate adjusted node configuration;

[0099] in, Representative Node Adjusted processing power and connection strength, Representative Node The real-time load, represents the maximum load among all nodes during the observation period, Representative Node Raw processing power, This is to avoid the small adjustment factor introduced when the denominator is zero and increase the stability of the calculation. Adjustment factor used to enhance node processing capability;

[0100] formula:

[0101] ;

[0102] The formula is useful because it dynamically adjusts the coefficients and , can adjust node strength according to real-time node load changes, match network load fluctuations, and improve network responsiveness and stability.

[0103] Detailed explanation of the formula and the process of formula calculation and derivation:

[0104] Assume the real-time load of node i is Unit, maximum load during observation period Unit, node raw processing power Unit, adjustment factor and , then the adjusted node strength is calculated as follows:

[0105] ;

[0106] The results show that by adjusting the calculation formula, the strength of node i is adjusted to about 29.03 units, matching its real-time load changes and ensuring the operation of the network.

[0107] S214: Apply the adjusted node configuration to the entire network structure, dynamically match load changes according to real-time data, optimize the connection and processing capabilities of each node in the network, and generate an optimized network structure.

[0108] Apply the calculated node configuration to the entire network structure. In a high-load environment, it is crucial to dynamically adjust the processing power and connection strength of each node to optimize performance. Through real-time data dynamic matching technology, the network can automatically adjust its structure and resource allocation according to the current load data. The key in this process is to perform data flow and balanced load processing. The generated optimized network structure should record the performance indicators and connection status of each node before and after adjustment in order to evaluate the effect and adjust the structure.

[0109] See also Figure 4 , the specific steps for obtaining the dynamic probability model are:

[0110] S311: Use the optimized network structure to monitor the data of the current construction project, analyze the interaction between nodes, and obtain a preliminary node interaction data set;

[0111] The data of the current construction project is monitored using an optimized network structure. A preliminary node interaction data set is obtained by analyzing the interactions between nodes. In this process, after obtaining the node data, the real-time monitoring system removes outliers and noise through data cleaning and preliminary processing to ensure the reliability and accuracy of the data. The data between nodes is statistically analyzed to evaluate the interaction intensity and frequency between nodes, thereby providing a basis for the next step of probability relationship update and ensuring that the acquired node interaction data set can truly reflect the dynamic relationship between nodes.

[0112] S312: updating the probability relationship between nodes according to the node interaction data in the preliminary node interaction data set, calculating and adjusting the statistical probability value of the node relationship, and generating an updated node probability relationship matrix;

[0113] The data of the current construction project is monitored using an optimized network structure. A preliminary node interaction data set is obtained by analyzing the interactions between nodes. In this process, after obtaining the node data, the real-time monitoring system removes outliers and noise through data cleaning and preliminary processing to ensure the reliability and accuracy of the data. The data between nodes is statistically analyzed to evaluate the interaction intensity and frequency between nodes, thereby providing a basis for the next step of probability relationship update and ensuring that the acquired node interaction data set can truly reflect the dynamic relationship between nodes.

[0114] S313: Automatically integrate the updated node probability relationship matrix into the conditional probability table using the formula:

[0115] ;

[0116] Update the conditional probability table, dynamically predict node behavior and system status based on real-time data, and generate a dynamic probability model;

[0117] in, It represents the conditional probability of event A occurring under the condition that event B occurs, reflecting the predictability of event A occurring when event B is known to occur. represents the marginal probability of event A, represents the marginal probability of event B, It represents the conditional probability of event B occurring under the condition that event A occurs, which indicates the probability of event B occurring when event A has already occurred.

[0118] formula:

[0119] ;

[0120] The benefit of the formula is that by introducing square root operations and multiplication, the nonlinearity of the calculation is increased, thereby improving the model's matching and prediction accuracy for complex interactive behaviors between nodes.

[0121] Detailed explanation of the formula and the process of formula calculation and derivation:

[0122] Assume that nodes A and B have 100 interactions during the monitoring period, and the probability of A acting independently is is 0.6, the probability of B being an independent activity is 0.7, the conditional probability of A given B The calculated value is 0.8, which is substituted into the formula:

[0123] ;

[0124] The results show that the probability of A occurring under the condition that B occurs is about 63%. The association with the step results is that this probability value directly affects the update of the conditional probability table and affects the accuracy and response of the dynamic probability model.

[0125] See also Figure 5 , the specific steps for obtaining stress prediction and propagation analysis results are as follows:

[0126] S411: Probabilistic inference is performed based on the dynamic probability model, and the future stress state in the structure is predicted using statistical analysis methods to obtain prediction data of the future stress state;

[0127] To perform probability inference based on the dynamic probability model, first, it is necessary to collect previous data and real-time monitoring data related to structural stress, and use the data to establish and calibrate the model. This involves data from physical sensors, including stress sensors and displacement sensors. These sensors are installed at key locations of the structure to obtain real-time stress change data. Next, statistical analysis methods are used to process these data, including regression analysis and Bayesian inference, to estimate the stress state of the structure at a certain moment in the future. This process requires data processing and model calculations. Through analysis, the stress state in the future time period is predicted. The generated prediction data is a quantitative representation of the future structural safety.

[0128] S412: Analyze the propagation direction of stress in the structure from the predicted data of future stress state, refer to material properties and structural connections, and generate analysis results of stress propagation direction;

[0129] To analyze the direction of stress propagation in the structure from the predicted data of future stress states, it is necessary to refer to the mechanical properties of the material and the connection method of the structure. The elastic modulus and yield strength of differentiated materials will affect the distribution and propagation of stress in the structure. The structural connection method, including welding and bolting, will also have a significant impact on the stress propagation path. By simulating these factors, the stress propagation path can be simulated in the computer model. Through finite element analysis, the generated analysis results can show how stress is transferred from one part to another, which is crucial for preventing structural failure and conducting maintenance planning.

[0130] S413: According to the stress propagation direction analysis results, calculate the critical propagation path using the formula:

[0131] ;

[0132] Generate calculation results of key propagation paths;

[0133] in, represents the critical strength of stress propagation, represents the probability contribution of node i, is the average of the probabilities, is the standard deviation of the probability, Represents the slave node The distance to the next node, is the total number of nodes in the path;

[0134] formula:

[0135] ;

[0136] Detailed explanation of the formula and the process of formula calculation and derivation:

[0137] Evaluate critical strength of stress propagation in structures , calculated based on the deviation, standard deviation and distance between nodes of multi-node probability contribution. The steps and parameter assignments are as follows:

[0138] Parameter assignment:

[0139] :node According to the recent structural detection data, for example, the probability contribution of node 1 is 0.30, node 2 is 0.45, and node 3 is 0.25;

[0140] : The average of the probabilities. Calculate all The average value of ;

[0141] : The standard deviation of probability. According to the probability contribution data, the standard deviation is calculated as follows: , insert the data and calculate ;

[0142] : From the node The distance to the next node (in meters). Obtained from field measurements, rice, rice, rice;

[0143] : The total number of nodes in the path, 3 in this example.

[0144] Calculation formula:

[0145] For each node, calculate .

[0146] Node 1: ;

[0147] Node 2: ;

[0148] Node 3: ;

[0149] Add up the above calculation results and divide by the total number of nodes : .

[0150] Numerical results represent the critical strength of stress propagation This indicates that under the current structural configuration and material performance conditions, the strength of stress propagation is relatively significant. Based on this key strength value, structural safety analysis can be performed to evaluate the performance of the structure under future loads and provide a basis for maintenance and reinforcement work.

[0151] S414: Use the critical propagation path calculation results, integrated structural models and probabilistic data to generate stress prediction and propagation analysis results.

[0152] Using the calculation results of the critical propagation path, combined with the structural model and probability data, the stress prediction and propagation analysis results are generated. This process involves combining the calculated stress propagation data with the actual structural model to simulate the behavior of the structure in the actual working environment. This requires the use of professional structural analysis software to integrate multiple data sources and apply physical and mathematical models to predict the stress state of the structure in the future time period. Through this comprehensive analysis, the stress concentration areas in the structure can be more accurately identified, providing a scientific basis for structural health monitoring and maintenance decisions in the future time period. The obtained stress prediction results will directly affect the structural safety assessment and the formulation of maintenance strategies.

[0153] See also Figure 6 , the steps to obtain the structural behavior analysis results after kernel parameter adjustment are as follows:

[0154] S511: Analyze stress prediction and propagation analysis results, use multi-kernel learning methods to process data in high-dimensional space, assign preliminary kernel parameters to each data point, and generate structural behavior analysis;

[0155] When dealing with stress prediction and propagation analysis results, it is crucial to use multi-kernel learning methods to process data in high-dimensional space. This process involves the initial assignment of kernel parameters, which correspond to the characteristics of high-dimensional data points. Through mathematical models, including kernel methods and support vector machines, the data is preliminarily classified and feature extracted. The choice of kernel function and parameter setting directly affect the performance and accuracy of the model. The bandwidth of the Gaussian kernel and the degree of the polynomial kernel need to be adjusted according to the distribution and characteristics of the data. This requires not only the implementation of the algorithm, but also preliminary analysis of the data, including data standardization and outlier detection. These steps ensure the accuracy of the mapping from high-dimensional space to kernel parameters and lay the foundation for generating structural behavior analysis results.

[0156] S512: Based on the results of the structural behavior analysis, adjust the parameters of each core to ensure that each parameter matches the characteristics of the data, adjust the weights through linear algebra and optimization algorithms, and generate a parameter adjustment plan;

[0157] Based on the preliminary kernel parameter analysis results, parameter adjustment is performed. This process is the core step in the multi-kernel learning method. The mathematical optimization techniques involved include gradient descent and genetic algorithms. These techniques can assist in adjusting the parameters of each kernel to match the data characteristics. When processing nonlinear data, kernel adjustment has a huge impact on the accuracy and generalization ability of the prediction model. The adjustment of weights is not only based on the performance of a single kernel, but also needs to refer to the complementarity between kernels to enhance the overall performance of the model. This requires analysis and testing of the mutual influence between differentiated kernels to ensure that each parameter adjustment is based on the optimization of overall performance rather than local optimality. The generated parameter adjustment scheme will directly affect the accuracy and reliability of the structural behavior analysis results.

[0158] S513: Apply the parameter adjustment scheme and recalculate the structural behavior using the formula:

[0159] ;

[0160] Reflecting the adjusted structural behavior through a weighted Gaussian function, generating structural behavior analysis results after kernel parameter adjustment;

[0161] in, Represents the response of the structural behavior, reflecting the prediction result obtained by the multi-kernel learning algorithm through the weighted summation of the differentiated kernel function. It is The tuning parameter of each core represents the weight of the core in the overall structural behavior. is the input variable, representing the data characteristics, It is The mean parameter of the kernel function describes the center position of the kernel corresponding to the data distribution. It is The standard deviation parameter of the kernel function indicates the degree of dispersion of the data distribution.

[0162] formula:

[0163] ;

[0164] The benefit of the formula is that, through the weighted Gaussian function form, the formula explanation and formula calculation derivation process can be explained in detail:

[0165] There are three kernels set up, and their parameters are ,enter , the mean of the kernel , standard deviation The calculation process is:

[0166] ;

[0167] ;

[0168] ;

[0169] ;

[0170] ;

[0171] ;

[0172] The results show that the structural behavior response At a given input When , according to the set kernel parameters, the prediction result is , showing the contribution of multi-core parameters to the output, so that the parameters can be adjusted according to actual conditions to optimize model performance.

[0173] A construction project quality monitoring system based on big data, the construction project quality monitoring system based on big data is used to execute the above-mentioned construction project quality monitoring method based on big data, the system comprises:

[0174] The data acquisition module obtains the pressure, temperature and wind speed data of the building structure in real time through sensors, adds timestamp information, and uses time synchronization technology to time align multi-type sensor data to generate a synchronous monitoring data set;

[0175] The network optimization module adjusts the node strength value, updates the edge connection status, evaluates the node influence value, reconfigures the node priority, and generates an optimized network structure based on the pressure change value, temperature change value, and wind speed change value in the synchronous monitoring data set;

[0176] The probability modeling module is based on optimizing the network structure, monitoring the current pressure value, temperature value and wind speed value, analyzing the interaction between nodes, updating the probability relationship between nodes, integrating the data into the conditional probability table, and establishing a dynamic probability model;

[0177] The prediction and analysis module performs probability inference based on the dynamic probability model, predicts future pressure state values, analyzes stress propagation direction values, calculates key propagation path values, obtains stress prediction and propagation analysis results, processes stress prediction and propagation analysis results, uses multi-kernel learning methods to analyze pressure values, temperature values ​​and wind speed values, adjusts kernel parameter values, and generates structural behavior analysis results.

[0178] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A construction project quality monitoring method based on big data, characterized in that: The following steps are involved: The pressure, temperature and wind speed data of the building structure are obtained in real time through sensors, and timestamp information is added to the data. Time synchronization technology is used to align the time of multiple types of sensor data to obtain a synchronized monitoring data set. Analyze the data changes in the synchronous monitoring data set, adjust the node strength according to the load changes, adjust the structural mode of connecting nodes according to environmental factors including temperature and wind speed, and use statistical analysis methods to evaluate the mutual influence between differentiated nodes, reconfigure the priorities between nodes, dynamically generate a structural configuration that matches the current data, and generate an optimized network structure; Using the optimized network structure, monitoring current construction project data, analyzing the interaction between nodes, updating the probability relationship between nodes according to the interaction data, automatically integrating the data into a conditional probability table, updating the data through the automatic update mechanism of the conditional probability table, and establishing a dynamic probability model; Based on the dynamic probability model, probability inference is performed to predict the future stress state in the structure, analyze the propagation direction of stress in the structure, and obtain stress prediction and propagation analysis results by calculating the key propagation path; Processing the stress prediction and propagation analysis results, analyzing the data in a high-dimensional space using a multi-kernel learning method, adjusting parameters of multiple kernels to match data features, and generating structural behavior analysis results after kernel parameter adjustment; The steps for obtaining the optimized network structure are specifically as follows: Collecting the node load data in the synchronous monitoring data set, recording the load condition of each node at different time points, and generating time series load data of each node; Analyzing the load fluctuation pattern of each node from the time series load data of each node, calculating the standard deviation and mean of the node load, and generating node load statistics; Based on the node load statistics, the processing capacity and connection strength of each node are dynamically adjusted according to its load situation, using the formula: ； Generate adjusted node configuration; in, Representative Node Adjusted processing power and connection strength, Representative Node The real-time load, represents the maximum load among all nodes during the observation period, Representative Node Raw processing power, This is to avoid the small adjustment factor introduced when the denominator is zero and increase the stability of the calculation. Adjustment factor used to enhance node processing capability; The adjusted node configuration is applied to the entire network structure, load changes are dynamically matched according to real-time data, the connection and processing capabilities of each node in the network are optimized, and an optimized network structure is generated.

2. The construction engineering quality monitoring method based on big data according to claim 1 is characterized in that: The synchronous monitoring data set specifically includes pressure data, temperature data, and wind speed data; the optimized network structure includes node strength, connection mode, and node priority; the dynamic probability model includes the probability relationship between nodes, the conditional probability table, and the data update mechanism; the stress prediction and propagation analysis results include the future stress state, stress propagation direction, and key propagation path; the structural behavior analysis results include the application of learning methods, data analysis dimensions, and kernel parameter adjustment.

3. The construction engineering quality monitoring method based on big data according to claim 2 is characterized in that: The steps for acquiring the synchronous monitoring data set are specifically as follows: The pressure data, temperature data and wind speed data are collected by sensors installed on the key structures of the building project, and a time stamp is added to each data point to obtain a time-stamped raw data set; Applying a time alignment algorithm to process the time-tagged original data set, aligning the pressure, temperature and wind speed data on the same time axis by adjusting the data timestamps, and obtaining a time-aligned sensor data set; The time-aligned sensor data set is analyzed to calculate the monitoring indicators of all sensor data at each time point using the formula: ； Get comprehensive monitoring data set ; in, Represents the pressure data value at a single time point, Represents the temperature data value at a single point in time, Represents the wind speed data value at a single time point, is the weight coefficient of pressure data, which is used to adjust the influence of pressure data in the comprehensive monitoring indicators. is the weight coefficient of temperature data, which is used to adjust the influence of temperature data in the comprehensive monitoring indicators. is the weight coefficient of wind speed data, which is used to adjust the influence of wind speed data in the comprehensive monitoring indicators; The comprehensive monitoring data set is integrated and formatted to generate a synchronous monitoring data set.

4. The construction engineering quality monitoring method based on big data according to claim 3 is characterized in that: The steps of obtaining the dynamic probability model are specifically as follows: Using the optimized network structure to monitor data of the current construction project, analyzing the interactions between nodes, and obtaining a preliminary node interaction data set; According to the node interaction data in the preliminary node interaction data set, the probability relationship between nodes is updated, the statistical probability value of the node relationship is calculated and adjusted, and an updated node probability relationship matrix is ​​generated; The updated node probability relationship matrix is ​​automatically integrated into the conditional probability table using the formula: ； Update the conditional probability table, dynamically predict node behavior and system status based on real-time data, and generate a dynamic probability model; in, It represents the conditional probability of event A occurring under the condition that event B occurs, reflecting the predictability of event A occurring when event B is known to occur. represents the marginal probability of event A, represents the marginal probability of event B, It represents the conditional probability of event B occurring under the condition that event A occurs, which indicates the probability of event B occurring when event A has already occurred.

5. The construction engineering quality monitoring method based on big data according to claim 4 is characterized in that: The steps for obtaining the stress prediction and propagation analysis results are specifically as follows: Probabilistic inference is performed based on the dynamic probability model, and the future stress state in the structure is predicted using a statistical analysis method to obtain prediction data of the future stress state; Analyzing the propagation direction of stress in the structure from the predicted data of the future stress state, and generating analysis results of the stress propagation direction with reference to material properties and structural connections; According to the stress propagation direction analysis results, the key propagation path is calculated using the formula: ； Generate calculation results of key propagation paths; in, represents the critical strength of stress propagation, represents the probability contribution of node i, is the average of the probabilities, is the standard deviation of the probability, Represents the slave node The distance to the next node, is the total number of nodes in the path; The critical propagation path calculation results are used to integrate the structural model and probability data to generate stress prediction and propagation analysis results.

6. The construction engineering quality monitoring method based on big data according to claim 5 is characterized in that: The steps for obtaining the structural behavior analysis results after the kernel parameters are adjusted are specifically as follows: analyzing the stress prediction and propagation analysis results, processing the data in a high-dimensional space using a multi-kernel learning method, assigning preliminary kernel parameters to each data point, and generating a structural behavior analysis; Based on the structural behavior analysis results, adjust the parameters of each core to ensure that each parameter matches the characteristics of the data, adjust the weights through linear algebra and optimization algorithms, and generate a parameter adjustment plan; Applying the parameter adjustment scheme described, the structural behavior is recalculated using the formula: ； Reflecting the adjusted structural behavior through a weighted Gaussian function, generating structural behavior analysis results after kernel parameter adjustment; in, Represents the response of the structural behavior, reflecting the prediction result obtained by the multi-kernel learning algorithm through the weighted summation of the differentiated kernel function. It is The tuning parameter of each core represents the weight of the core in the overall structural behavior. is the input variable, representing the data characteristics, It is The mean parameter of the kernel function describes the center position of the kernel corresponding to the data distribution. It is The standard deviation parameter of the kernel function indicates the degree of dispersion of the data distribution.

7. A construction engineering quality monitoring system based on big data, characterized in that: According to the method for monitoring construction project quality based on big data according to any one of claims 1 to 6, the system comprises: The data acquisition module obtains the pressure, temperature and wind speed data of the building structure in real time through sensors, adds timestamp information, and uses time synchronization technology to time align multi-type sensor data to generate a synchronous monitoring data set; The network optimization module adjusts the node strength value, updates the edge connection status, evaluates the node influence value, reconfigures the node priority, and generates an optimized network structure based on the pressure change value, temperature change value, and wind speed change value in the synchronous monitoring data set; The probability modeling module monitors the current pressure value, temperature value and wind speed value based on the optimized network structure, analyzes the interaction between nodes, updates the probability relationship between nodes, integrates the data into the conditional probability table, and establishes a dynamic probability model; The prediction and analysis module performs probability inference based on the dynamic probability model, predicts future pressure state values, analyzes stress propagation direction values, calculates key propagation path values, obtains stress prediction and propagation analysis results, processes stress prediction and propagation analysis results, uses multi-kernel learning methods to analyze pressure values, temperature values ​​and wind speed values, adjusts kernel parameter values, and generates structural behavior analysis results.

Images