Concrete building quality monitoring system and method

Abstract

The invention relates to the technical field of quality management, in particular to a concrete building quality monitoring system and method. The problems that in the prior art, concrete building quality monitoring mainly depends on manual sampling inspection and regular inspection modes, and potential quality hazards in a concrete building structure are difficult to find and solve are solved. The optimal sensor deployment scheme is generated through the sensor deployment unit in the cloud computing platform, the sensors are deployed according to the deployment scheme, the data acquisition module acquires key parameters of the concrete building through the sensors, and the data transmission module transmits the acquired data. And the cloud computing platform performs safety assessment on the acquired data through an internal quality assessment unit so as to obtain intelligent monitoring and management of the concrete building. Potential safety hazards of the concrete building on the construction site can be found in time so that measures can be taken for repairing in advance, meanwhile, the maintenance cost and the management cost of the concrete building can be reduced, and the management efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of quality management, and particularly relates to a concrete building quality monitoring system and method. Background Art

[0002] Concrete is one of the most commonly used materials in construction engineering. The quality of concrete is directly related to the safety, reliability, and durability of buildings. Traditional methods for monitoring the quality of concrete buildings mainly rely on manual sampling inspections and regular inspections, which have problems such as insufficient monitoring and slow response, making it difficult to detect and solve quality hidden dangers in the concrete building structure in a timely manner.

[0003] Therefore, it is urgent to develop a concrete building quality monitoring system and method to overcome the shortcomings in the prior art. Summary of the Invention

[0004] To solve the problem in the prior art that the quality monitoring of concrete buildings mainly relies on manual sampling inspections and regular inspections, resulting in difficulties in discovering and solving quality hidden dangers in the concrete building structure, the present invention provides such a concrete building quality monitoring system, including: a data acquisition module, a data transmission module, and a cloud computing platform;

[0005] The data acquisition module, according to the optimal sensor deployment plan generated by the cloud computing platform, installs sensors with different functions at key nodes of the concrete building during the construction process, for converting the temperature, humidity, stress, and deformation data of the concrete building into electrical signals in real time, and transmitting the electrical signals to the cloud computing platform through the data transmission module;

[0006] The data transmission module is used to obtain the real-time sensor monitoring data of the data acquisition module, and transmit the real-time sensor monitoring data to the cloud computing platform through a 5G network;

[0007] The cloud computing platform is used to obtain the real-time sensor monitoring data transmitted by the data transmission module, analyze and process the real-time sensor monitoring data, and issue an alarm when it is predicted that there are problems with the quality of the concrete building.

[0008] Further, the cloud computing platform specifically includes a sensor deployment unit, a data storage unit, a data processing unit, a quality assessment unit, a threshold adjustment unit, an equipment maintenance unit, a data visualization unit, and an anomaly diagnosis unit;

[0009] The sensor deployment unit generates an optimal sensor deployment plan according to the building structure design drawings, and provides technical support for the data acquisition module to install sensors with different functions at key nodes of the concrete building during the construction process;

[0010] The data storage unit is used to store the real-time sensor monitoring data of the concrete building, and the real-time sensor monitoring data includes temperature and humidity, stress and deformation;

[0011] The data processing unit is used to preprocess the real-time sensor monitoring data in the data storage unit, standardize the preprocessed real-time sensor monitoring data, and convert it into a unified dimension and data range; and according to the characteristics of the concrete building, provide corresponding weights for different real-time sensor monitoring data, and fuse the real-time sensor monitoring data according to the weights to generate a comprehensive monitoring index;

[0012] The quality assessment unit is built with a quality and safety assessment model for the concrete building, and is used to perform quality assessment on the comprehensive monitoring index, and issue a warning after triggering the alarm threshold;

[0013] The threshold adjustment unit is used to dynamically adjust the alarm threshold in the quality assessment unit;

[0014] The equipment maintenance unit is used to monitor the status of the sensor devices in the data acquisition module and notify relevant personnel for maintenance. The status of the sensor devices includes data false alarms and device failures of the monitoring devices;

[0015] The data visualization unit is used to visually display the prediction results of the real-time sensor monitoring data;

[0016] The anomaly diagnosis unit is used to automatically analyze the anomaly causes by using a diagnosis method based on knowledge graph and rule reasoning and a diagnosis method based on case library and similarity calculation when the quality assessment unit predicts that the quality of the concrete building is abnormal.

[0017] Furthermore, during the building preparation stage, the sensor deployment unit obtains the design drawings of the concrete building, establishes a three-dimensional BIM model according to the drawings, and marks the structural characteristics of the building, including key positions and stress conditions; based on the structural characteristics of the concrete building, uses a sensor deployment algorithm to determine the optimal sensor deployment plan.

[0018] Among them, the specific calculation formula of the sensor deployment algorithm is:

[0019] minF de (x) = w 1 f 1 (x) + w 2 f 2 (x) + … + w n f n (x) + w k (αf l (x) + βf s(x)

[0020] Among them, F de (x) is the objective function, representing the optimization objective of the sensor deployment scheme; f 1 (x), f 2 (x)…f n (x) are the various factors affecting sensor deployment, and the factors include sensor type, quantity, and location; w 2 , w 2 …w n are the weight coefficients of the various factors affecting sensor deployment; f l (x) is the matching degree between sensor deployment and the characteristics of the concrete building structure, specifically the matching degree between the sensor deployment location and the key positions of the building; f s (x) is the matching degree between sensor deployment and the characteristics of the concrete building structure, specifically the matching degree between the sensor deployment location and the stress condition of the building; w k is the weight coefficient of the matching degree between sensor deployment and the characteristics of the concrete building structure; α and β are the weight coefficients of f l (x) and f s (x).

[0021] Furthermore, the sensor deployment algorithm is calculated through a genetic algorithm combined with an ant colony algorithm to obtain an optimal sensor deployment scheme, specifically as follows:

[0022] S101: Randomly generate an initial population of sensor deployment schemes based on the characteristics of the building structure, where each individual in the population represents a sensor deployment scheme;

[0023] S102: Calculate the fitness minF de (x) for each individual; and select parent and offspring according to the fitness;

[0024] S103: Perform crossover and mutation operations on the parent and offspring to generate new offspring individuals and update the population; evaluate the fitness of the updated population, calculate the fitness value of each individual, and determine whether there is an individual in the population that meets the fitness threshold; if yes, output the optimal solution in the population and stop the calculation; if not, then continue to execute step S104;

[0025] S104: Divide the current population into several subpopulations, where the individual characteristics within the subpopulations are relatively similar, while the characteristics between the subpopulations are quite different;

[0026] S105: Independently execute the ant colony algorithm in each of the sub-populations, simulating the process of the ant colony searching in parallel in different regions; and convert the excellent ant search paths during the search process into corresponding sensor deployment schemes, and take the converted sensor deployment schemes as new individuals and add them to the corresponding sub-populations;

[0027] S106: Independently perform crossover and mutation operations on each of the sub-populations to optimize the individual quality of each of the sub-populations;

[0028] S107: Copy the optimal individual in each sub-population to the global optimal solution set, and determine whether the global termination condition is satisfied; if yes, output the optimal individual in the global optimal solution set as the optimal sensor deployment scheme; if no, take the global optimal solution set as the new population, and continue to execute step S104; where the global termination condition includes whether the preset maximum global iteration number is reached and whether there is an individual in the global optimal solution set whose fitness value is higher than the fitness threshold.

[0029] Further, the data storage unit also stores a knowledge graph, a rule base, and a case base; wherein, the knowledge graph stores the semantic associations between common causes of concrete buildings, which are composed of the concepts, attributes, and relationships of common causes of concrete buildings; the rule base summarizes the diagnostic rules for various typical abnormal types based on the abnormal cases of concrete building quality monitoring, and forms the rule base from the diagnostic rules; the case base is composed of historical abnormal cases of concrete buildings, and is used to diagnose the quality abnormality of concrete buildings in combination with similarity calculation.

[0030] Further, the quality evaluation unit analyzes the real-time sensor monitoring data of concrete quality based on a machine learning algorithm, and the specific process is as follows:

[0031] S201: Build a quality and safety evaluation model for concrete buildings based on a machine learning algorithm;

[0032] The calculation formula of the quality and safety evaluation model for concrete buildings is:

[0033]

[0034] where, F ass (x) represents the prediction result of the quality and safety state of the concrete building, k represents the category of the quality and safety state of the concrete building, n represents the number of training samples, i represents the current training sample, represents the support vector coefficient of sample i for category k, represents the label of sample i under category k, K(X i , X) represents the similarity between sample i and the sample X to be predicted, b krepresents the judgment threshold for category k, γ is used to control the influence degree of additional factors on the final result, w 1 ,…w n is used to reflect the relative importance of different factors to the quality and safety of concrete buildings, f 1 (x),…f n (x) represents the factors that additionally affect the quality and safety of concrete buildings, arg max k means to find the category k that makes the whole formula take the maximum value;

[0035] S202: Construct a training data set with the design parameters and test data of the concrete building as features and the safety of the concrete building structure as labels;

[0036] S203: Use cross-validation to train and optimize the concrete building quality and safety assessment model to improve the generalization ability of the concrete building quality and safety assessment model;

[0037] S204: Apply the comprehensive monitoring index to the trained concrete building quality and safety assessment model to judge the safety status of the concrete building quality;

[0038] S205: When the concrete building quality and safety assessment model judges that there are potential safety hazards in the concrete building quality, issue an alarm in time to prompt the construction personnel to handle it.

[0039] Furthermore, the threshold adjustment unit is used to dynamically adjust the alarm threshold for judging the assessment result of the concrete building quality and safety in the quality assessment unit;

[0040] According to the design parameters of the concrete building, the design parameters include material strength and load, and initially set the alarm threshold for real-time sensor monitoring data;

[0041] Conduct statistical analysis on the real-time sensor monitoring data, calculate the mean and standard deviation of the monitoring parameters, and dynamically adjust the alarm threshold according to the statistical analysis results.

[0042] Furthermore, when the concrete building quality and safety assessment model detects an anomaly, first match the corresponding diagnostic rules in the rule base according to the anomaly type to obtain the preliminary cause of the anomaly, and then identify the anomaly cause with the highest matching degree in the knowledge graph according to the preliminary cause of the anomaly;

[0043] At the same time, when conducting fault diagnosis based on knowledge graph and rule reasoning, the anomaly diagnosis unit calculates the similarity between the current anomaly type and the anomaly types in historical cases, and matches the historical case with the highest similarity;

[0044] The abnormal diagnosis unit combines the diagnosis results of the diagnosis method based on knowledge graph and rule reasoning and the diagnosis method based on case base and similarity calculation, and outputs the final diagnosis conclusion of the abnormal cause.

[0045] Further, the data visualization unit displays the real-time sensor monitoring data on the BIM model. When there is a problem at one of the monitoring points, the position of the BIM model where the monitoring point is located is replaced with a bright color, so that the management personnel can quickly confirm the monitoring point that needs to be maintained; at the same time, the data visualization module displays the optimal sensor deployment plan generated by the sensor deployment unit through the BIM model, and the engineering construction personnel can deploy sensors according to the displayed optimal sensor deployment plan.

[0046] Further, a concrete building quality monitoring method includes:

[0047] S1: In the preparation stage of the concrete building, input the engineering construction drawing data into the cloud computing platform. The sensor deployment unit in the cloud computing platform obtains the drawing data, establishes a BIM model and obtains the key nodes of the concrete building structure; the sensor deployment unit generates an optimal sensor deployment plan according to the key nodes and the drawing data;

[0048] S2: In the construction stage of the concrete building, the engineering construction personnel deploy sensors with different functions according to the optimal sensor deployment plan;

[0049] S3: The data acquisition module obtains the real-time sensor monitoring data collected by the started sensors, transmits it to the cloud computing platform through the data transmission module, and stores it in the data storage unit of the cloud computing platform;

[0050] S4: The data processing unit obtains the original real-time sensor monitoring data stored in the data storage unit, preprocesses the original real-time sensor monitoring data, and unifies the dimension and data range of the real-time sensor monitoring data; and fuses different sensor monitoring data according to different weights to obtain a comprehensive monitoring index;

[0051] S5: The quality assessment unit conducts a quality and safety assessment on the key nodes of the concrete building through the comprehensive monitoring index, and issues a warning when the prediction result exceeds the alarm threshold;

[0052] S6: When the concrete building quality and safety assessment model determines that the quality of the concrete building is abnormal, the data visualization unit marks the problematic nodes in the BIM model with bright colors to prompt the engineering construction personnel that the nodes need to be maintained;

[0053] S7: When the abnormal diagnosis unit automatically analyzes the cause of the abnormality by using the diagnosis method based on the knowledge graph and rule reasoning and the diagnosis method based on the case base and similarity calculation, and outputs the abnormal diagnosis conclusion;

[0054] S8: After the construction worker completes the node maintenance according to the abnormal diagnosis conclusion, the quality evaluation unit re-obtains the real-time sensor monitoring data of the node for safety evaluation. After the evaluation is normal, the cloud computing platform withdraws the warning and continues to monitor.

[0055] Beneficial effects:

[0056] 1. The present invention collects various key influencing parameters including temperature, humidity, deformation, etc. in the concrete building structure through sensors, monitors the overall condition of the concrete building structure in real time, and constructs a concrete building quality and safety evaluation model based on the SVM algorithm to analyze the detection data in time and predict potential safety hazards. Once a potential safety problem occurs, an alarm will be issued to remind the management personnel to pay attention. The present invention can not only master the operation condition of the concrete building structure in real time, but also timely discover safety hazards, providing a strong guarantee for the safe operation of the concrete building.

[0057] 2. The present invention generates an optimal sensor deployment plan by using the BIM technology and combining the characteristics of the concrete building structure, and determines the optimal sensor deployment position and type; the optimal sensor deployment plan can not only collect representative and accurate monitoring data, but also minimize the number of sensors to the greatest extent, which helps to improve the economy and operability of the quality monitoring of the concrete building project. Description of the drawings

[0058] Figure 1 It is the system structure diagram of the present invention.

[0059] Figure 2 It is the flow chart for solving the sensor deployment algorithm of the present invention.

[0060] Figure 3 It is the flow chart for analyzing the monitoring data of the concrete structure of the present invention.

[0061] Figure 4 It is the flow chart of the concrete quality monitoring method of the present invention. Detailed implementation manners

[0062] The present invention will be further described below with reference to the drawings and embodiments.

[0063] As shown in the attached Figure 1As shown in the figure, in order to solve the problems that the monitoring of traditional concrete buildings relies on manual sampling inspection and regular inspection, resulting in insufficient monitoring, lagging response, and difficulty in timely discovering and solving quality hidden dangers in the structure of concrete buildings, the present invention provides a concrete building quality monitoring system. Refer to the appendix Figure 1 , which includes a data acquisition module, a data transmission module, and a cloud computing platform. The cloud computing platform further includes a sensor deployment unit, a data storage unit, a data processing unit, a quality assessment unit, a threshold adjustment unit, an equipment maintenance unit, a data visualization unit, and an anomaly diagnosis unit.

[0064] In this embodiment, according to the optimal sensor deployment plan generated by the cloud computing platform, during the construction process of the building, sensors with different functions are installed at the key nodes of the concrete building to convert the temperature, humidity, stress, and deformation data of the concrete building into electrical signals in real time, and transmit the electrical signals to the cloud computing platform through the data transmission module.

[0065] The data acquisition module includes sensors such as temperature and humidity sensors, stress sensors, and deformation sensors. During the construction of the concrete building, they are installed inside the concrete building. By converting the temperature, humidity, stress, deformation and other data of the concrete building into electrical signals, and transmitting the electrical signals to the cloud computing platform through the data transmission module. Among them, the data acquisition module adopts the automatic acquisition method and collects relevant data once every fixed time period; the cloud computing platform can set the transmission frequency time of the sensors and adjust the acquisition time interval to cope with emergencies.

[0066] Specifically, the temperature and humidity sensors are deployed on the concrete surface to monitor the surface temperature and humidity changes of the concrete structure; the stress sensors are deployed on the steel bars to monitor the stress changes of the steel bars and reflect the stress state of the concrete structure; the deformation sensors are deployed at the key positions of the concrete structure, such as the column foot, beam end, etc., to monitor the deformation displacement of the structure; at the same time, according to the stress characteristics of the concrete structure, additional sensor deployment points are added in the stress concentration area and deformation area.

[0067] In this embodiment, the data transmission module is used to obtain the data of the data acquisition module and transmit the collected data to the cloud computing platform through the 5G network.

[0068] Specifically, the data transmission frequency of the data transmission module is consistent with the data acquisition frequency of the data acquisition module, that is, the collected data is monitored and transmitted at fixed times and will change with the change of the acquisition interval; at the same time, the data transmission module also has the function of automatic caching. When the network is interrupted, the collected data can be temporarily stored and automatically uploaded after the network is restored.

[0069] In this embodiment, a cloud computing platform is used to obtain the concrete building monitoring data transmitted by the data transmission module, analyze and process the data, and issue an alarm when it is predicted that there are problems with the concrete building structure.

[0070] Specifically, the cloud computing platform includes a sensor deployment unit, a data storage unit, a data processing unit, a quality assessment unit, a threshold adjustment unit, a device maintenance unit, a data visualization unit, and an anomaly diagnosis unit. Among them, the data storage unit communicates with other units through interfaces and stores the data in other units and modules in a hierarchical manner.

[0071] As shown in the appendix Figure 2 In this embodiment, the sensor deployment unit generates a sensor deployment plan according to the building structure design drawings, and provides technical support for the data acquisition module to install sensors with different functions at the key nodes of the concrete building during the building construction process.

[0072] Specifically, in the building preparation stage, the sensor deployment unit obtains the concrete building design drawings, establishes a three-dimensional BIM model according to the design drawings, and marks the structural characteristics of the building, including key positions and stress conditions, etc.; based on the structural characteristics of the concrete building, the sensor deployment algorithm is used to determine the optimal sensor deployment plan, including sensor type, quantity, location, and cost, etc.

[0073] Specifically, the specific calculation formula of the sensor deployment algorithm is:

[0074] minF de (x) = w 1 f 1 (x) + w 2 f 2 (x) + … + w n f n (x) + w k (αf t (x) + βf s (x)

[0075] Among them, F de (x) is the objective function, representing the optimization objective of the sensor deployment plan; f 1 (x), f 2 (x) … f n (x) are the various factors affecting sensor deployment, including sensor type, quantity, location, and deployment cost, etc.; w 2 , w 2 … w n are the weight coefficients of the various factors affecting sensor deployment; f t(x) represents the factors of the building structure characteristics, specifically the matching degree between the sensor deployment location and the key locations of the building; f s (x) represents the factors of the building's characteristic structure, specifically the matching degree between the sensor deployment location and the stress conditions of the building; w k are the weight coefficients of the building characteristic factors; α and β are the weight coefficients of f t (x) and f s (x).

[0076] Specifically, the genetic algorithm is combined with the ant colony algorithm and clustering analysis to calculate and solve the sensor deployment algorithm to obtain the optimal sensor deployment plan. Refer to the appendix Figure 2 , and the specific steps are as follows:

[0077] S101: Randomly generate an initial population of sensor deployment plans based on the building structure characteristics, where each individual in the population represents a sensor deployment plan.

[0078] S102: Calculate the fitness minF de (x) for each individual; and select the parent and offspring according to the fitness.

[0079] S103: Perform crossover and mutation operations on the parent and offspring to generate new offspring individuals and update the population; evaluate the fitness of the updated population, calculate the fitness value of each individual, and determine whether there is an individual in the population that meets the fitness threshold; if yes, output the optimal solution in the population and stop the calculation; if no, then continue to execute step S104.

[0080] S104: Divide the current population into several sub-populations, where the individual characteristics within the sub-populations are relatively similar, while the characteristics between the sub-populations are quite different.

[0081] S105: Independently execute the ant colony algorithm in each sub-population to simulate the process of the ant colony searching in different regions in parallel; and convert the excellent ant search paths during the search process into corresponding sensor deployment plans, and add the converted sensor deployment plans as new individuals to the corresponding sub-populations.

[0082] S106: Independently perform crossover and mutation operations on each sub-population to optimize the individual quality of each sub-population.

[0083] S107: Copy the optimal individual in each sub-population to the global optimal solution set, and determine whether the global termination condition is met; if yes, output the optimal individual in the global optimal solution set as the optimal sensor deployment plan; if no, then use the global optimal solution set as the new population and continue to execute step S104; where the global termination condition includes whether the preset maximum global iteration times are reached and whether there is an individual in the global optimal solution set with a fitness value higher than the fitness threshold.

[0084] Specifically, the new individuals obtained by introducing the ant colony algorithm after the genetic algorithm updates the population are to enhance the global search ability of the algorithm, improve population diversity, avoid premature convergence, and thus obtain better solutions.

[0085] In this embodiment, the data storage unit is used to store the monitoring data of the concrete building, including temperature and humidity, stress, deformation, etc.

[0086] Specifically, the data storage unit adopts a hierarchical database structure to store the original data layer, the preprocessed data layer, and the analysis result layer in layers. Among them, the original data is the original sensor data obtained from the data acquisition module, the preprocessed data is the sensor data preprocessed by the data processing unit, and the analysis result data is the concrete building quality safety assessment result obtained by the quality assessment unit after analyzing the comprehensive monitoring indicators. At the same time, in order to protect the privacy of the data, the data storage unit encrypts and stores sensitive data including the design parameters of the building, and only authorized users can access it.

[0087] Specifically, the data storage unit also stores a knowledge graph, a rule base, and a case base; among them, the knowledge graph stores the semantic associations between common causes of concrete buildings, which are composed of the concepts, attributes, and relationships of common causes of concrete buildings; the rule base summarizes the diagnostic rules for various typical abnormal types based on the abnormal cases of concrete building quality monitoring, and forms a rule base from the diagnostic rules. Each rule in the rule base contains parameter abnormal conditions and corresponding failure causes; the case base is composed of historical abnormal cases of concrete buildings and is used to diagnose the quality abnormality of concrete buildings in combination with similarity calculation. During abnormal diagnosis, by calculating the similarity between the current case and the historical cases, the most similar case can be quickly matched as a reference and supplement to the diagnostic result, where the similarity calculation comprehensively considers multiple feature dimensions such as the abnormal mode of monitoring parameters, the structure type, and the environmental conditions.

[0088] In this embodiment, the data processing unit is used to preprocess the monitoring data in the data storage unit and generate comprehensive monitoring indicators.

[0089] The data processing unit acquires the sensor data collected by the data acquisition module and preprocesses the sensor data, including data cleaning, missing value processing, outlier monitoring, and denoising, to remove outliers and noise interference in the monitoring data and improve the integrity and continuity of the data. Then, it standardizes the preprocessed sensor data to convert different sensor data into a unified dimension and data range. According to the characteristics of the concrete building and the monitoring requirements, the data processing unit selects different weights for different sensor data, fuses the sensor data, and obtains a comprehensive monitoring index. The comprehensive monitoring index is used as the input feature of the concrete building quality safety assessment model in the quality assessment unit, enabling the comprehensive monitoring index to reflect the overall condition of the concrete building quality.

[0090] In this embodiment, a concrete building quality safety assessment model is deployed inside the quality assessment unit to conduct a safety assessment on the comprehensive monitoring index. When the prediction result exceeds the alarm threshold, a warning is issued and maintenance measures are proposed.

[0091] Specifically, the quality assessment unit analyzes the real-time sensor monitoring data of the concrete building based on a machine learning algorithm, referring to Appendix Figure 3 , and the specific process is as follows:

[0092] S201: Based on a machine learning algorithm, construct a concrete building quality safety assessment model.

[0093] The calculation formula of the concrete building quality safety assessment model is:

[0094]

[0095] Among them, F ass (x) represents the prediction result of the concrete building quality safety status; k represents the category of the concrete building quality safety status, including categories such as safe and potential hazards; n represents the number of training samples; i represents the current training sample; represents the support vector coefficient of sample i for category k, which is used to reflect the importance of each sample in model training; represents the label of sample i under category k, that is, the safety status of the sample; K(X i ,X) is the kernel function, which is used to represent the similarity between sample i and the sample X to be predicted; b k represents the judgment threshold of category k, which defines the judgment boundary between different safety status categories and provides a basis for the final safety status judgment; γ is used to control the influence degree of additional factors on the final result; w 1 ,…w n is used to reflect the relative importance of different factors to the concrete building quality safety; f 1 (x),…f n(x) represents the factors affecting the quality and safety of concrete buildings, including factors such as temperature and humidity, stress, and deformation; argmax k means to find the category k that maximizes the entire formula, that is, among all possible categories k, find the one that makes the entire formula reach the maximum value; the quality and safety assessment model of concrete buildings comprehensively evaluates the safety status of the quality of concrete buildings by integrating the real-time sensor monitoring data of concrete buildings, improving the accuracy and reliability of the quality and safety assessment of concrete buildings.

[0096] S202: Using the design parameters and detection data of concrete buildings as features, including the comprehensive monitoring indicators of the data processing unit, etc., and using whether the concrete building structure is safe as a label, construct a training data set.

[0097] S203: Use cross-validation to train and optimize the quality and safety assessment model of concrete buildings, improve the generalization ability of the quality and safety assessment model of concrete buildings. During the training process, by adjusting the algorithm parameters of the quality and safety assessment model of concrete buildings, continuously optimize the performance of the quality and safety assessment model of concrete buildings, and finally obtain a quality and safety assessment model of concrete buildings with better performance.

[0098] S204: Apply the comprehensive monitoring indicators to the trained quality and safety assessment model of concrete buildings to predict the safety status of the quality of concrete buildings.

[0099] S205: When the quality and safety assessment model of concrete buildings determines that there are potential safety hazards in the building structure, issue a warning message in a timely manner to prompt relevant personnel to handle it.

[0100] In this embodiment, the threshold adjustment unit is used to dynamically adjust the threshold warning in the safety assessment unit.

[0101] Specifically, before the construction of the concrete building starts, according to the design parameters of the concrete building, including parameters such as concrete strength, steel bar strength, and bearing load, initially set the alarm thresholds for various real-time sensor monitoring data; after the construction starts, continuously collect the real-time data of various real-time sensor monitoring indicators. By statistically analyzing these real-time sensor monitoring data, calculate the average value and standard deviation of each indicator to understand the actual distribution of each real-time sensor monitoring data; then, according to the statistical analysis results, dynamically adjust the alarm thresholds of each real-time sensor monitoring data in the quality assessment unit.

[0102] In this embodiment, the equipment maintenance unit is used to monitor the status of the sensor equipment in the data acquisition module, including monitoring problems such as data false alarms and equipment damage of the equipment, and notifying relevant personnel for maintenance.

[0103] Specifically, in order to address possible deviations that may occur during long-term use and ensure the continuous accuracy of data, the present invention is provided with a device maintenance unit. The device maintenance unit continuously monitors the operating states of various sensor devices in the data acquisition module, including but not limited to: data transmission quality, device failure latency, and data anomalies. Among them, the data transmission quality is to monitor the integrity and timeliness of real-time sensor monitoring data and detect whether there are problems such as data loss and latency; device failure is to monitor in real time whether the sensor device fails, including problems such as sensor damage and abnormal power supply; data anomaly is to analyze whether there are abnormal values in the real-time sensor monitoring data and promptly discover possible sensor failures or calibration problems.

[0104] In this embodiment, the anomaly diagnosis unit is used to automatically analyze the anomaly causes when the quality assessment unit predicts the anomaly of the concrete building quality, adopting the diagnosis method based on knowledge graph and rule reasoning and the diagnosis method based on case base and similarity calculation.

[0105] Specifically, when the quality assessment model detects an anomaly, first match the corresponding diagnosis rules in the rule base according to the anomaly type to obtain the preliminary anomaly cause, and then further query other possible causes in the knowledge graph according to the preliminary anomaly cause to identify the anomaly cause with the highest matching degree.

[0106] Meanwhile, when performing the fault diagnosis based on knowledge graph and rule reasoning, the anomaly diagnosis unit calculates the similarity between the current anomaly type and the anomaly types in historical cases, and matches the historical case with the highest similarity. Among them, the monitoring parameter anomaly characteristics and the final diagnosis reasons of various typical anomalies are recorded in the case base; and the similarity calculation comprehensively considers multiple feature dimensions such as the anomaly pattern of monitoring parameters, structure type, and environmental conditions.

[0107] The anomaly diagnosis unit combines the diagnosis results of the diagnosis method based on knowledge graph and rule reasoning and the diagnosis method based on case base and similarity calculation, and outputs the final anomaly cause diagnosis conclusion.

[0108] In this embodiment, the data visualization unit is used to visually display the real-time sensor monitoring data and the prediction results.

[0109] The data visualization unit can visually display various real-time sensor monitoring data obtained from the data acquisition module on the BIM model through different charts and graphs. When a problem occurs at a monitoring point, the position of the BIM model where the monitoring point is located is replaced with a bright color, enabling managers to quickly confirm the monitoring points that need to be maintained. At the same time, the data visualization module can visually display the optimal sensor deployment plan generated by the sensor deployment unit to the engineering construction personnel in the form of a BIM model. The engineering construction personnel can, based on the visualization effect, understand the specific installation positions and installation steps of the sensors, and the data visualization module uses this to guide the actual sensor deployment work.

[0110] In this embodiment, a method for monitoring the quality of concrete buildings refers to the appendix Figure 4 , including:

[0111] S1: In the preparation stage of the concrete building, input the engineering construction drawing data into the cloud computing platform. The sensor deployment unit in the cloud computing platform obtains the drawing data, establishes a BIM model, and obtains the key nodes of the concrete building structure. The sensor deployment unit gives a sensor deployment plan based on the key nodes and the drawing data.

[0112] S2: In the construction stage of the concrete building, the engineering construction personnel deploy sensors at the key nodes according to the sensor deployment plan.

[0113] S3: The data acquisition module obtains the real-time sensor monitoring data collected by the sensors that start to run, transmits it to the cloud computing platform through the data transmission module, and stores it in the data storage unit of the cloud computing platform.

[0114] S4: The data processing unit obtains the original real-time sensor monitoring data stored in the data storage unit, preprocesses the original real-time sensor monitoring data, unifies the dimension and data range of the data; and selects different weights for different real-time sensor monitoring data to obtain a comprehensive monitoring index.

[0115] S5: The quality assessment unit conducts a safety assessment on the key nodes of the concrete building through the comprehensive monitoring index, and issues a warning when the prediction result exceeds the preset threshold.

[0116] S6: The data visualization unit obtains the warning information, modifies the color of the node on the BIM model to red warning, and the engineering construction personnel maintain the node according to the node information displayed by the data visualization unit and the solution recommended by the quality assessment unit.

[0117] S7: When the anomaly diagnosis unit automatically analyzes the cause of the anomaly using the diagnosis method based on knowledge graph and rule reasoning and the diagnosis method based on case library and similarity calculation, and outputs the anomaly diagnosis conclusion;

[0118] S8: After the engineering construction personnel complete the node maintenance according to the abnormal diagnosis conclusion, the quality assessment unit re-obtains the real-time sensor monitoring data of the node for safety assessment. After the assessment is normal, the cloud computing platform withdraws the warning and continues to monitor.

[0119] The above-described embodiments merely represent the preferred embodiments of the present invention, and the description thereof is relatively specific and detailed. However, it should not be construed as a limitation on the scope of the patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several variations, improvements and substitutions can be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the appended claims.

Claims

1. A concrete building quality monitoring system, characterized in that: Including data acquisition module, data transmission module and cloud computing platform; The data acquisition module, according to the optimal sensor deployment plan generated by the cloud computing platform, installs sensors with different functions at key nodes of the concrete building during the construction process, so as to convert the temperature, humidity, stress and deformation data of the concrete building into electrical signals in real time, and transmits the electrical signals to the cloud computing platform through the data transmission module; The data transmission module is used to obtain the real-time sensor monitoring data of the data acquisition module, and transmit the real-time sensor monitoring data to the cloud computing platform through the 5G network; The cloud computing platform is used to obtain the real-time sensor monitoring data transmitted by the data transmission module, analyze and process the real-time sensor monitoring data, and issue an alarm when problems with the quality of the concrete building are predicted.

2. A concrete building quality monitoring system according to claim 1, characterized in that: The cloud computing platform specifically includes a sensor deployment unit, a data storage unit, a data processing unit, a quality assessment unit, a threshold adjustment unit, an equipment maintenance unit, a data visualization unit and an abnormality diagnosis unit; the sensor deployment unit generates an optimal sensor deployment plan based on the building structure design drawings, and provides technical support for the data acquisition module to install sensors with different functions at key nodes of concrete buildings during the construction process; The data storage unit is used to store real-time sensor monitoring data of the concrete building, wherein the real-time sensor monitoring data includes temperature and humidity, stress and deformation; The data processing unit is used to preprocess the real-time sensor monitoring data in the data storage unit, and standardize the preprocessed real-time sensor monitoring data to convert it into a unified dimension and data range; and according to the characteristics of the concrete building, provide corresponding weights for different real-time sensor monitoring data, and fuse the real-time sensor monitoring data according to the weights to generate a comprehensive monitoring index; The quality assessment unit has a built-in concrete building quality safety assessment model, which is used to perform quality assessment on the comprehensive monitoring indicators and issue a warning after triggering an alarm threshold; The threshold adjustment unit is used to dynamically adjust the alarm threshold in the quality assessment unit; The equipment maintenance unit is used to monitor the status of the sensor equipment in the data acquisition module and notify relevant personnel to perform maintenance, wherein the status of the sensor equipment includes data misreporting and equipment damage of the monitoring equipment; The data visualization unit is used to visualize the prediction results of real-time sensor monitoring data; the abnormality diagnosis unit is used to automatically analyze the cause of the abnormality when the quality assessment unit predicts that the concrete building quality is abnormal, using a diagnosis method based on a knowledge graph and rule reasoning and a diagnosis method based on a case library and similarity calculation.

3. A concrete building quality monitoring system according to claim 1, characterized in that: The sensor deployment unit obtains the concrete building design drawings during the construction preparation stage, establishes a three-dimensional BIM model based on the drawings, and marks the structural characteristics of the building, including key locations and stress conditions; based on the structural characteristics of the concrete building, uses the sensor deployment algorithm to determine the optimal sensor deployment plan; The specific calculation formula of the sensor deployment algorithm is: minF de (x)=w1f1(x)+w2f2(x)+…+w n f n (x)+w k (αf l (x)+βf s (x) Among them, F de (x) is the objective function, which represents the optimization goal of the sensor deployment plan; f1(x), f2(x)…f n (x) are various factors affecting sensor deployment, including sensor type, quantity and location; w2, w2…w n is the weight coefficient of each factor affecting sensor deployment; f l (x) is the matching degree between the sensor deployment and the structural characteristics of the concrete building, specifically, the matching degree between the sensor deployment location and the key location of the building; f s (x) is the matching degree between the sensor deployment and the structural characteristics of the concrete building, specifically, the matching degree between the sensor deployment position and the stress condition of the building; w k is the weight coefficient of the matching degree between sensor deployment and concrete building structure characteristics; α and β are f l (x) and f s The weight coefficient of (x).

4. A concrete building quality monitoring system according to claim 3, characterized in that: The sensor deployment algorithm is calculated by combining the genetic algorithm with the ant colony algorithm to obtain the optimal sensor deployment solution, which is specifically: S101: Randomly generate an initial sensor deployment scheme population based on the structural characteristics of the building, wherein each individual in the population represents a sensor deployment scheme; S102: Calculate the fitness minF for each individual de (x); and selecting the parent offspring according to the fitness; S103: performing crossover and mutation operations on the parent-generation offspring to generate new offspring individuals and update the population; evaluating the fitness of the updated population, calculating the fitness value of each individual, and determining whether there is an individual in the population that meets the fitness threshold; If yes, output the optimal solution in the population and stop the calculation; if no, continue to execute step S104; S104: dividing the current population into a plurality of sub-populations, wherein individual characteristics within the sub-populations are relatively similar, while the characteristics between the sub-populations are relatively different; S105: independently executing the ant colony algorithm in each of the sub-populations to simulate the process of parallel search of ants in different areas; converting the excellent ant search paths in the search process into corresponding sensor deployment plans, and adding the converted sensor deployment plans as new individuals to the corresponding sub-populations; S106: independently performing crossover and mutation operations on each of the sub-populations to optimize the individual quality of each of the sub-populations; S107: Copy the best individual in each sub-population to the global optimal solution set, and determine whether the global termination condition is met; if yes, output the best individual in the global optimal solution set as the optimal sensor deployment solution; if no, use the global optimal solution set as the new population, and continue to execute step S104: wherein the global termination condition includes whether a preset maximum number of global iterations is reached and whether there is an individual with a fitness value higher than the fitness threshold in the global optimal solution set.

5. A concrete building quality monitoring system according to claim 2, characterized in that: The data storage unit also stores a knowledge graph, a rule base and a case base; wherein the knowledge graph stores semantic associations between common causes of concrete building diseases, and is composed of concepts, attributes and relationships of common causes of concrete building diseases; the rule base is based on abnormal cases of concrete building quality monitoring, and summarizes diagnostic rules for various typical abnormal types, and the rule base is formed by the diagnostic rules; the case base is composed of historical abnormal cases of concrete buildings, and is used to diagnose concrete building quality abnormalities in combination with similarity calculations.

6. A concrete building quality monitoring system according to claim 2, characterized in that: The quality assessment unit analyzes the real-time sensor monitoring data of concrete quality based on a machine learning algorithm. The specific process is as follows: S201: Construct a concrete building quality safety assessment model based on machine learning algorithm; The calculation formula of the concrete building quality safety assessment model is: Among them, F ass (x) represents the prediction result of the quality and safety status of concrete buildings, k represents the category of the quality and safety status of concrete buildings, n represents the number of training samples, i represents the current training sample, represents the support vector coefficient of sample i for category k, represents the label of sample i under category k, K(X i ,X) represents the similarity between sample i and the sample to be predicted X, b k represents the judgment threshold of category k, γ is used to control the influence of additional factors on the final result, w1,…w n It is used to reflect the relative importance of different factors to the quality and safety of concrete buildings, f1(x),…f n (x) represents additional factors affecting the quality and safety of concrete buildings, argmax k It means to find the category k that takes the maximum value in the entire formula; S202: construct a training data set using the design parameters and test data of concrete buildings as features and whether the concrete building structure is safe as a label; S203: training and tuning the concrete building quality and safety assessment model using cross-validation to improve the generalization ability of the concrete building quality and safety assessment model; S204: applying the comprehensive monitoring index to the trained concrete building quality safety assessment model to judge the safety status of the concrete building quality; S205: When the concrete building quality safety assessment model determines that there are potential safety hazards in the concrete building quality, an alarm is promptly issued to prompt the construction personnel to handle the problem.

7. A concrete building quality monitoring system according to claim 2, characterized in that: The threshold adjustment unit is used to dynamically adjust the alarm threshold used by the quality assessment unit to determine the concrete building quality safety assessment result; Preliminarily setting alarm thresholds for real-time sensor monitoring data based on design parameters of the concrete building, including material strength and load; Perform statistical analysis on real-time sensor monitoring data, calculate the mean and standard deviation of monitoring parameters, and dynamically adjust the alarm threshold based on the statistical analysis results.

8. A concrete building quality monitoring system according to claim 2, characterized in that: When the concrete building quality safety assessment model detects an abnormality, it first matches the corresponding diagnostic rules in the rule base according to the abnormality type to obtain the preliminary abnormality cause, and then identifies the abnormality cause with the highest matching degree in the knowledge graph based on the preliminary abnormality cause; At the same time, when performing fault diagnosis based on knowledge graph and rule reasoning, the abnormality diagnosis unit calculates the similarity between the current abnormality type and the abnormality type in the historical case, and matches the historical case with the highest similarity; The abnormality diagnosis unit combines the diagnosis results based on the knowledge graph and rule reasoning diagnosis method and the diagnosis results based on the case library and similarity calculation diagnosis method to output the final abnormality cause diagnosis conclusion.

9. A concrete building quality monitoring system according to claim 2, characterized in that: The data visualization unit displays the real-time sensor monitoring data on the BIM model. When a problem occurs at one of the monitoring points, the BIM model position of the monitoring point is replaced with a bright color, so that management personnel can quickly identify the monitoring point that needs maintenance; at the same time, the data visualization module displays the optimal sensor deployment plan generated by the sensor deployment unit through the BIM model, and construction personnel can deploy sensors according to the displayed optimal sensor deployment plan.

10. A method for monitoring the quality of concrete buildings, characterized in that: include: S1: In the concrete building preparation stage, the engineering construction drawing data is input into the cloud computing platform, and the sensor deployment unit in the cloud computing platform obtains the drawing data, establishes a BIM model and obtains key nodes of the concrete building structure; the sensor deployment unit generates an optimal sensor deployment plan based on the key nodes and the drawing data; S2: During the concrete building construction phase, construction personnel deploy sensors with different functions according to the optimal sensor deployment scheme; S3: the data acquisition module acquires the real-time sensor monitoring data collected by the sensor that has started to run, transmits the data to the cloud computing platform through the data transmission module, and the data storage unit of the cloud computing platform stores the data; S4: the data processing unit obtains the original real-time sensor monitoring data stored in the data storage unit, pre-processes the original real-time sensor monitoring data, unifies the dimension and data range of the real-time sensor monitoring data; and fuses different sensor monitoring data according to different weights to obtain a comprehensive monitoring index; S5: The quality assessment unit performs a quality safety assessment on the key nodes of the concrete building through the comprehensive monitoring indicators, and issues a warning when the prediction result exceeds the alarm threshold; S6: When the concrete building quality safety assessment model determines that the concrete building quality is abnormal, the data visualization unit marks the nodes with problems in bright colors in the BIM model, prompting the construction personnel that the nodes need maintenance; S7: When the abnormality diagnosis unit adopts the diagnosis method based on knowledge graph and rule reasoning and the diagnosis method based on case library and similarity calculation to automatically analyze the abnormality cause and output the abnormality diagnosis conclusion; S8: After the engineering construction personnel complete the node maintenance according to the abnormal diagnosis conclusion, the quality assessment unit reacquires the real-time sensor monitoring data of the node for safety assessment. After the assessment is normal, the cloud computing platform withdraws the warning and continues monitoring.

Images