A steel structure construction quality monitoring system based on drones and 3D scanning

Abstract

The present invention relates to the field of construction engineering technology, and specifically to a steel structure construction quality monitoring system based on unmanned aerial vehicles and three-dimensional scanning, including a data acquisition module, a data preprocessing module, a construction defect identification and risk assessment module, and a quality assessment module, wherein: the data acquisition module collects quality monitoring data of the construction site in real time; the data preprocessing module preprocesses the collected quality monitoring data; the construction defect identification and risk assessment module identifies steel structure defects during the construction process and assesses the severity of the steel structure defects; the quality assessment module performs a comprehensive quality assessment on the preprocessed quality monitoring data and provides corresponding repair suggestions. The present invention can quickly identify key problem areas, optimize construction resource allocation strategies, and provide comprehensive and accurate data support for steel structure quality assessment in complex construction environments.

Description

Technical Field

[0001] The present invention relates to the technical field of construction engineering, and in particular to a steel structure construction quality monitoring system based on unmanned aerial vehicle and three-dimensional scanning. Background Art

[0002] As an important form of construction engineering, steel structure has the advantages of high strength, good toughness and short construction period. It is widely used in bridges, high-rise buildings, large-span factories and other fields. However, due to the complexity of steel structure and the variability of construction environment, quality problems such as welding defects, structural dislocation, deformation and cracks may occur during the construction process. If these defects are not discovered and repaired in time, it will not only affect the progress of the project, but also may cause safety hazards. Therefore, real-time monitoring and evaluation of the quality of steel structure construction, especially in complex construction scenarios, is of great significance.

[0003] In the existing technology, the quality monitoring methods of steel structure construction mainly rely on manual inspection and static equipment detection. These methods usually require on-site construction personnel to identify and record structural defects through visual inspection or simple measuring tools, which are inefficient and have large errors. At the same time, traditional detection equipment is mostly single-function static equipment, such as total stations, static scanners, etc., which cannot integrate and analyze the dynamic changes and multi-source data of the construction site. In addition, the existing defect assessment and repair methods are mostly based on experience, lack of scientific and quantitative evaluation standards and efficient repair decision support, resulting in limited construction quality monitoring effects.

[0004] The present invention aims to overcome the above-mentioned shortcomings and provide a steel structure construction quality monitoring system based on drones and three-dimensional scanning, which not only improves the accuracy and efficiency of construction quality monitoring, but also effectively reduces the errors and resource waste of manual inspection, and significantly improves the safety and reliability of steel structure construction. Summary of the invention

[0005] The present invention provides a steel structure construction quality monitoring system based on unmanned aerial vehicle and three-dimensional scanning.

[0006] A steel structure construction quality monitoring system based on drone and three-dimensional scanning includes a data acquisition module, a data preprocessing module, a construction defect identification and risk assessment module and a quality assessment module, wherein;

[0007] The data acquisition module collects quality monitoring data of the construction site in real time, including three-dimensional data, image data, environmental monitoring data and geometric information of the steel structure of the construction site;

[0008] The data preprocessing module preprocesses the collected quality monitoring data, including format conversion, denoising, missing value filling and standardization;

[0009] The construction defect identification and risk assessment module identifies steel structure defects during the construction process based on the three-dimensional data and image data of the construction site, including welding defects, misalignment, deformation, and cracks, and assesses the severity of the steel structure defects, including:

[0010] Preliminary data processing: Preliminary processing of the collected 3D data and image data of the construction site, including local contrast enhancement and edge enhancement processing;

[0011] Construction defect identification: Based on the preliminarily processed 3D data and image data, the convolutional neural network (CNN) model is used to identify various defects in steel structures, including welding defects, misalignment, deformation, and cracks.

[0012] Defect severity assessment: Based on the various defects of the steel structure identified, the severity of each defect is assessed;

[0013] The quality assessment module combines the results of steel structure defect identification and severity assessment, conducts a comprehensive quality assessment on the pre-processed quality monitoring data, generates a construction quality report, including the detected defect type, location, severity, and results of the comprehensive quality assessment, and provides corresponding repair suggestions based on the construction quality report.

[0014] Optionally, the data acquisition module includes:

[0015] 3D data acquisition: Use drones to use laser scanning to collect 3D data of the construction site in real time, generate a 3D model, and provide spatial position and shape information of the steel structure;

[0016] Image data acquisition: Use camera equipment (drone-mounted camera or ground camera) to capture image data of the construction site in real time;

[0017] Environmental monitoring data collection: Deploy environmental sensors (temperature and humidity sensors, barometers, anemometers) to monitor the environmental parameters of the construction site in real time, including temperature, humidity, air pressure, and wind speed;

[0018] Steel structure geometric information collection: Through laser scanning equipment, the geometric information of the steel structure can be obtained in real time, and the size, shape and relative position of each component of the steel structure can be monitored.

[0019] Optionally, the data preprocessing module includes:

[0020] Format conversion: convert quality monitoring data from different sources into a unified standard format, including 3D data into point cloud data format, image data into standard image file format (JPEG or PNG), environmental monitoring data into time series data format, and geometric information of steel structures into BIM format;

[0021] Denoising: Denoising the collected 3D data, image data and environmental monitoring data. Use Gaussian filter to denoise the 3D data, median filter to denoise the image data, and Kalman filter to denoise the environmental monitoring data.

[0022] Missing value filling: Use linear interpolation to fill missing values ​​in the collected quality monitoring data;

[0023] Standardization processing: Standardize the collected quality monitoring data.

[0024] Optionally, the preliminary data processing includes:

[0025] Local contrast enhancement: Enhance the visibility of image details by enhancing the contrast of local areas;

[0026] Edge enhancement processing: By using the Laplace operator for edge detection, the edge information in the image data is enhanced and the boundaries of the steel structure components are highlighted.

[0027] Optionally, the convolutional neural network (CNN) model includes:

[0028] Multi-channel input data fusion: The 3D point cloud data and 2D image data are processed separately through a multi-channel structure to extract the spatial features of the 3D point cloud data and the texture and brightness features of the 2D image data. The extracted spatial features, texture and brightness features are then fused to generate comprehensive features containing spatial and texture information.

[0029] Deep residual feature extraction: Based on the fused comprehensive features, the residual network is introduced to extract deep and complex features;

[0030] Feature weighting based on attention mechanism: Introduce channel attention mechanism, calculate the importance of each channel feature through global average pooling, and dynamically generate channel weights;

[0031] Defect type classification: The steel structure defects are classified through the fully connected layer, and the probability distribution of the defect type is output, which represents the prediction results of various types of defects (welding defects, dislocations, deformations, cracks).

[0032] Optionally, the defect severity assessment includes:

[0033] Defect data extraction and classification: Extract basic information of each defect from the results of construction defect identification, including defect type (welding defect, dislocation, deformation, crack), defect location and geometric features (area, length, width);

[0034] Geometric feature quantification: Based on the geometric features of each defect, the physical scale of the defect is quantified and the geometric score of the defect is generated. ;

[0035] Position weight calculation: combined with the identified position of the defect , evaluate the potential impact of defects on structural stability and safety, and generate location weighted scores ;

[0036] Comprehensive severity score: Based on the extracted geometric score and position weight score , calculate the composite severity score of the defect .

[0037] Optionally, the quality assessment module includes:

[0038] Comprehensive analysis of defect information: Summarize and analyze the results of steel structure defect identification and severity assessment to determine the impact and severity ranking of each defect type;

[0039] Comprehensive structural quality assessment: Combined with the results of comprehensive analysis of defect information, comprehensive analysis is performed on the pre-processed quality monitoring data to calculate the overall quality score of the steel structure;

[0040] Repair suggestion generation: Generate targeted repair suggestions based on the results of comprehensive structural quality assessment.

[0041] Optionally, the comprehensive analysis of defect information includes:

[0042] Defect type impact calculation: Quantify the impact of each defect type and calculate the comprehensive impact score of each type based on the severity score and the importance score of the defect location ;

[0043] Defect severity ranking: comprehensive impact score based on defect type Sort all defect types and determine severity ranking .

[0044] Optionally, the structural comprehensive quality assessment includes:

[0045] Integrated defect impact data: for steel structures by region Group the areas The result of comprehensive analysis of defect information within (comprehensive impact score and severity ranking ) is integrated with the pre-processed quality monitoring data to generate the regional Comprehensive data of

[0046] Regional quality score calculation: Based on the integrated comprehensive data, the quality status of each region is quantitatively analyzed and the regional quality score is calculated. ;

[0047] Overall quality score calculation: The quality score of each area is weighted according to the area weight to generate the overall quality score of the steel structure. .

[0048] Optionally, the repair suggestion generation includes:

[0049] Determine fix priority: Calculate a fix priority score for each defect based on the area quality score, severity ranking, and defect severity score ;

[0050] Recommended repair method: Based on defect type and the geometric characteristics of the defect , combined with the preset repair method library, generate the corresponding repair plan, including the repair method, required materials and tools;

[0051] Generate repair plan: Based on the repair priority score and recommended repair method, generate an overall repair plan, including a priority list of defect repairs, a list of materials and equipment required for each repair task, a repair duration forecast, and allocate time by repair task.

[0052] Beneficial effects of the present invention:

[0053] The present invention, by integrating a data acquisition module, a data preprocessing module, a construction defect identification and risk assessment module and a quality assessment module, can collect and process a variety of quality monitoring data of the construction site in real time, covering three-dimensional data, image data, environmental monitoring data and steel structure geometry information. The data preprocessing module significantly improves the consistency and quality of the data through format conversion, denoising, missing value filling and standardization processing. The construction defect identification module combines deep learning models and multi-source data fusion technology to accurately identify and evaluate steel structure defects, providing efficient and accurate technical support for construction quality monitoring.

[0054] The present invention integrates the results of comprehensive analysis of defect information with normalized weighted quality monitoring data, scientifically calculates regional and overall quality scores, dynamically quantifies the impact degree and severity ranking of defects, and combines the results of comprehensive impact scores and severity rankings to quickly identify key problem areas, optimize construction resource allocation strategies, and provide comprehensive and accurate data support for steel structure quality assessment in complex construction environments, effectively improving the scientificity and reliability of construction management.

[0055] The present invention, through the repair suggestion generation module, intelligently generates targeted repair suggestions according to the quality assessment results, including defect repair priority sorting, repair method recommendation and repair plan formulation. By combining the defect type, geometric characteristics and regional weight, the repair priority is accurately quantified, and the most suitable repair plan is recommended, covering the required materials, equipment and time allocation, which significantly improves the efficiency and accuracy of the repair work, reduces resource waste and construction risks, and provides comprehensive protection for the long-term stability and safety of steel structure construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1 A schematic diagram of system function modules according to an embodiment of the present invention;

[0058] Figure 2 Schematic diagram of a construction defect identification and risk assessment module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0060] It should be noted that the references to "one embodiment", "embodiment", "exemplary embodiments", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0061] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0062] like Figure 1-Figure 2 As shown, a steel structure construction quality monitoring system based on drone and three-dimensional scanning includes a data acquisition module, a data preprocessing module, a construction defect identification and risk assessment module and a quality assessment module, wherein;

[0063] The data acquisition module collects quality monitoring data of the construction site in real time, including 3D data, image data, environmental monitoring data and geometric information of the steel structure;

[0064] The data preprocessing module preprocesses the collected quality monitoring data, including format conversion, denoising, missing value filling and standardization;

[0065] The construction defect identification and risk assessment module identifies steel structure defects during construction, including welding defects, misalignment, deformation, and cracks, based on the 3D data and image data of the construction site, and assesses the severity of the steel structure defects, including:

[0066] Preliminary data processing: Preliminary processing of the collected 3D data and image data of the construction site, including local contrast enhancement and edge enhancement processing;

[0067] Construction defect identification: Based on the preliminarily processed 3D data and image data, the convolutional neural network (CNN) model is used to identify various defects in steel structures, including welding defects, misalignment, deformation, and cracks.

[0068] Defect severity assessment: Based on the various defects of the steel structure identified, the severity of each defect is assessed;

[0069] The quality assessment module combines the results of steel structure defect identification and severity assessment to conduct a comprehensive quality assessment on the pre-processed quality monitoring data, generate a construction quality report, including the type, location, severity, and comprehensive quality assessment results of the detected defects, and provide corresponding repair suggestions based on the construction quality report;

[0070] Through the above content, various data from the construction site are collected and processed in real time, various defects in steel structure construction are accurately identified and evaluated, and the efficiency and accuracy of construction quality monitoring are comprehensively improved. Through data preprocessing, defect identification and risk assessment, and multi-level analysis of comprehensive quality assessment, detailed construction quality reports can be generated in real time, and scientific and reasonable repair suggestions can be provided, thereby effectively reducing construction risks, improving construction safety, reducing errors and costs of manual inspection, and ensuring high quality and high reliability of steel structure construction.

[0071] The data acquisition module includes:

[0072] 3D data acquisition: Use drones to use laser scanning to collect 3D data of the construction site in real time, generate a 3D model, and provide spatial position and shape information of the steel structure;

[0073] Image data acquisition: Use camera equipment (drone-mounted camera or ground camera) to capture image data of the construction site in real time;

[0074] Environmental monitoring data collection: Deploy environmental sensors (temperature and humidity sensors, barometers, anemometers) to monitor the environmental parameters of the construction site in real time, including temperature, humidity, air pressure, and wind speed;

[0075] Steel structure geometric information collection: Use laser scanning equipment to obtain the geometric information of the steel structure in real time and monitor the size, shape and relative position of each component of the steel structure;

[0076] Through the above content, the three-dimensional data, image data, environmental monitoring data and steel structure geometric information of the construction site can be collected in real time and comprehensively. This all-round and multi-level data collection method not only greatly improves the accuracy and real-time performance of data collection, but also can accurately monitor every detail of the construction process, discover potential problems in advance, avoid blind spots in manual inspection, and ensure the reliability and safety of construction quality.

[0077] The data preprocessing module includes:

[0078] Format conversion: convert quality monitoring data from different sources into a unified standard format, including 3D data into point cloud data format, image data into standard image file format (JPEG or PNG), environmental monitoring data into time series data format, and geometric information of steel structures into BIM format;

[0079] Denoising: Denoising the collected 3D data, image data and environmental monitoring data. Use Gaussian filter to denoise the 3D data, median filter to denoise the image data, and Kalman filter to denoise the environmental monitoring data. Specifically, it includes:

[0080] (1) 3D data denoising: ;

[0081] in, is the denoised data, is the original data, is the mean value of the noise, is the standard deviation;

[0082] (2) Image data denoising: ;

[0083] in, is the pixel value after denoising, is the neighborhood pixel value in the image;

[0084] (3) Environmental monitoring data denoising: ;

[0085] in, is the state after filtering, is the actual observed value, is the Kalman gain, is the observation matrix, is the estimated state at the previous moment;

[0086] Missing value filling: Linear interpolation is used to fill missing values ​​in the collected quality monitoring data, expressed as:

[0087] ;

[0088] in, is the fill value for missing values, and is the time point of known data, and is the corresponding known value, is the time variable;

[0089] Standardization: The collected quality monitoring data is standardized so that data from different sources can be compared on the same scale, expressed as:

[0090] ;

[0091] in, is the original data, is the mean of the data, is the standard deviation of the data, is the standardized data;

[0092] Through the above content, the quality and consistency of data can be effectively improved. Format conversion ensures that information from different data sources can be processed uniformly. De-noising can remove interference signals in the data and improve the accuracy and reliability of the data. Missing value filling effectively makes up for the gaps that may occur during the data collection process and avoids errors and analysis biases caused by incomplete data. Standardization further ensures the consistency of the data in terms of dimensions.

[0093] Initial data processing includes:

[0094] Local contrast enhancement: By enhancing the contrast of local areas, the visibility of image details is enhanced, helping to better identify small structural differences at the construction site, expressed as:

[0095] ;

[0096] in, is the original image at pixel position The value of is the mean value of the local area, is the standard deviation of the local area, is the contrast enhancement factor, is the image after local contrast enhancement at pixel position The new pixel value of

[0097] Edge enhancement processing: By using the Laplace operator for edge detection, the edge information in the image data is enhanced, and the boundaries of the steel structure components are highlighted, thereby improving the accuracy of defect recognition, which is expressed as:

[0098] ;

[0099] in, is the edge strength, is the original image at pixel position The value of and They are images exist and The second-order derivative in the direction highlights the high-frequency changes in the image, that is, the edge part;

[0100] Through the above content, the detail visibility and boundary clarity of the image are effectively improved. Local contrast enhancement can highlight the tiny brightness differences on the surface of the steel structure, making potential problems such as welding defects and cracks more obvious. Edge enhancement highlights the structural boundary information and improves the accuracy and resolution of defect positioning. It can significantly improve data quality and feature expression capabilities, and is particularly suitable for steel structure defect detection in complex construction scenarios, reducing the possibility of missed detection and false detection, and providing key support for high-precision quality monitoring.

[0101] The Convolutional Neural Network (CNN) model consists of:

[0102] Multi-channel input data fusion: The 3D point cloud data and 2D image data are processed separately through a multi-channel structure, the spatial features of the 3D point cloud data and the texture and brightness features of the 2D image data are extracted, and the extracted spatial features, texture and brightness features are fused to generate a comprehensive feature containing spatial and texture information, which is expressed as:

[0103] ;

[0104] ;

[0105] ;

[0106] in, is the input 3D point cloud data, is the input image data, , Respectively The weights corresponding to the layers, , is the bias term, is the ReLU activation function, Represents the feature fusion operation, merging multi-channel features. For comprehensive characteristics;

[0107] Deep residual feature extraction: Based on the fused comprehensive features, the residual network is introduced to extract deep complex features. The residual structure directly transfers the input features to the output layer through jump connections, and extracts high-dimensional features in the convolution operation to avoid the problems of gradient disappearance and network degradation. It is expressed as:

[0108] ;

[0109] in, is the comprehensive feature after fusion of multi-channel input, The output features after residual structure optimization retain the fusion features and enhance the expression ability of complex defects. , is a two-layer convolution kernel. , is the bias term;

[0110] Feature weighting based on attention mechanism: Introduce channel attention mechanism, calculate the importance of each channel feature through global average pooling, and dynamically generate channel weights, which are expressed as:

[0111] ;

[0112] ;

[0113] in, is the output feature after residual structure optimization, is the global average pooling operation, , is the weight matrix of the attention mechanism, is the channel weight, is the weighted feature;

[0114] ;

[0115] in, For the The channel in the layer is at position The pixel value of is the height of the feature map, is the width of the feature map, For the The global average pooling value of the channels;

[0116] Defect type classification: The steel structure defects are classified through the fully connected layer, and the probability distribution of the defect type is output, which represents the prediction results of various defects (welding defects, dislocations, deformations, cracks), expressed as:

[0117] ;

[0118] in, is the probability distribution of defect types, indicating the possibility of predicted defects. is the feature weighted by the attention mechanism, which is the input feature of the classification. is the classification weight matrix, is the bias vector for classification;

[0119] Through the above content, it is possible to efficiently process multi-source data at the construction site, extract key features of steel structure defects, make full use of the attention mechanism to dynamically adjust channel weights, highlight important information related to defects, and at the same time enhance deep feature extraction capabilities through the residual structure to avoid gradient vanishing and information loss. It shows higher recognition accuracy and robustness in complex construction scenarios, greatly improving the efficiency and reliability of steel structure construction quality monitoring.

[0120] Defect severity assessment includes:

[0121] Defect data extraction and classification: Extract basic information of each defect from the results of construction defect identification, including defect type (welding defect, dislocation, deformation, crack), defect location and geometric features (area, length, width), expressed as:

[0122] ;

[0123] in, For the The type of defect, For the The location of the defect, For the The geometric characteristics of the defect, including area ,length and width , is the total number of defects identified;

[0124] Geometric feature quantification: Based on the geometric features of each defect, the physical scale of the defect is quantified and the geometric score of the defect is generated. , expressed as:

[0125] ;

[0126] in, , , are the weight coefficients of the corresponding geometric features, It is a quantified geometric score, which indicates the impact of defect size on the overall structure;

[0127] Position weight calculation: combined with the identified position of the defect , evaluate the potential impact of defects on structural stability and safety, and generate location weighted scores , expressed as:

[0128] ;

[0129] in, Score the structural importance of the defect location, Score the overall location of the steel structure;

[0130] Comprehensive severity score: Based on the extracted geometric score and position weight score , calculate the composite severity score of the defect , expressed as:

[0131] ;

[0132] in, and is the balance coefficient, For the The composite severity score of the defects;

[0133] Structural importance score of the defect location The settings include:

[0134] (1) Determine the key characteristics of the location: According to the design and construction requirements of the steel structure, define the key characteristics of the defect location, including bearing capacity (the maximum load that the location can bear), connection node density (the number and density of connection points between the location and surrounding components), stress distribution level (stress level at this location under structural stress), geometric importance (Whether the location is a key structural component);

[0135] (2) Define the feature importance weight: Assign a weight to each key feature to reflect its importance to the overall structural safety, expressed as:

[0136] ;

[0137] in, is the bearing capacity weight, is the connection node density weight, is the stress distribution level weight, is the geometric importance weight;

[0138] (3) Calculate the characteristic score: Based on the collected quality monitoring data, calculate the score of each characteristic, expressed as:

[0139] ;

[0140] in, For carrying capacity rating, is the actual bearing capacity at the defect location, is the design bearing capacity of the defect location;

[0141] ;

[0142] in, Score the density of connected nodes, is the actual number of connected nodes at the defect location, is the maximum number of connected nodes in the structure;

[0143] ;

[0144] in, Score the stress distribution level, is the actual stress level at the defect location, is the design stress limit of the defect location;

[0145] = ;

[0146] in, Score geometric importance;

[0147] (4) Comprehensive calculation of location importance score: Based on each feature score and the corresponding weight, the comprehensive importance score of the defect location is calculated, expressed as:

[0148] ;

[0149] Overall location score for steel structures The settings include:

[0150] (1) Divide the structural area: Divide the steel structure into several key areas (such as main beams, secondary beams, node connections, supporting components, etc.), each area is divided according to its function, force characteristics and design importance;

[0151] (2) Determine regional weights: for each region Assign a weight , indicating the importance of the region in the overall structure, expressed as:

[0152] ;

[0153] in, , , is the corresponding adjustment coefficient;

[0154] (3) Calculate the location score of each area: For each area , calculate its location score based on the collected data , expressed as:

[0155] ;

[0156] in, For Region The total number of critical points or defects in In the region The score of each key point or defect, is the weight of the corresponding key characteristic;

[0157] (4) Calculate the total location score of the steel structure: Comprehensive weight of each area and location score , calculate the total location score of the steel structure , expressed as:

[0158] ;

[0159] in, is the total number of areas of the steel structure, For the The weight of the region, For the Rating of the region;

[0160] Through the above content, the impact of each defect on the safety and stability of steel structures can be fully quantified. Through the dynamic calculation of characteristic scores and location weights based on actual data, the evaluation results can be ensured to be more scientific and accurate. In addition, this method combines the regional importance of the structure and multi-source data analysis, which can adapt to the design requirements and construction scenarios of different steel structures, and provide strong support for priority repair decisions and quality assessments, thereby significantly improving the efficiency and reliability of construction quality monitoring.

[0161] The quality assessment modules include:

[0162] Comprehensive analysis of defect information: Summarize and analyze the results of steel structure defect identification and severity assessment to determine the impact and severity ranking of each defect type;

[0163] Comprehensive structural quality assessment: Combined with the results of comprehensive analysis of defect information, comprehensive analysis is performed on the pre-processed quality monitoring data to calculate the overall quality score of the steel structure;

[0164] Repair suggestion generation: Generate targeted repair suggestions based on the results of comprehensive structural quality assessment;

[0165] Through the above content, the whole process from defect data summarization to overall quality quantitative assessment and then to repair decision support is covered. It combines multi-dimensional information such as defect type, severity and location importance to ensure the scientificity and accuracy of the assessment results. Through the generation of comprehensive quality scores and targeted repair suggestions, it not only improves the comprehensiveness and efficiency of steel structure construction quality monitoring, but also provides clear operational guidance for construction personnel, reducing resource waste and potential risks in the repair process.

[0166] Comprehensive analysis of defect information includes:

[0167] Defect type impact calculation: Quantify the impact of each defect type and calculate the comprehensive impact score of each type based on the severity score and the importance score of the defect location , expressed as:

[0168] ;

[0169] in, Defect type , For type The number of defects, For the Importance score of each defect location;

[0170] Defect severity ranking: comprehensive impact score based on defect type Sort all defect types and determine severity ranking , expressed as:

[0171] ;

[0172] in, Defect type The severity ranking, is the total number of defect types, For Press A function that sorts values ​​from high to low;

[0173] Through the above content, the type, location, geometric characteristics and severity score of the defects are fully integrated to ensure the integrity and operability of the data. By calculating the comprehensive impact score of each defect type, the potential threat of the defects to the overall structure is dynamically reflected, and a severity ranking is generated based on this. It provides clear priority guidance for construction quality monitoring and repair plans, can quickly identify key problem areas, help optimize resource allocation, improve construction management efficiency, and effectively reduce potential safety risks.

[0174] Comprehensive structural quality assessment includes:

[0175] Integrated defect impact data: for steel structures by region Group the areas The result of comprehensive analysis of defect information within (comprehensive impact score and severity ranking ) is integrated with the pre-processed quality monitoring data to generate the regional The comprehensive data is expressed as:

[0176]

[0177] in, For Region The comprehensive data of Defect type The comprehensive impact score of Rank the severity of the defect types, Comprehensive quality monitoring data for the region, is the total number of regions;

[0178] Regional quality score calculation: Based on the integrated comprehensive data, the quality status of each region is quantitatively analyzed and the regional quality score is calculated. , expressed as:

[0179] ;

[0180] in, For Region The number of defect types, is the maximum comprehensive impact score of all defect types, used for normalization processing, Defect type The corresponding comprehensive monitoring quality data is obtained by normalizing and weighting the monitoring quality data;

[0181] Overall quality score calculation: The quality score of each area is weighted according to the area weight to generate the overall quality score of the steel structure. , expressed as:

[0182] ;

[0183] in, For Region The weight is set according to the importance of structure and function. For Region Quality rating of

[0184] Through the above content, the results of comprehensive analysis of defect information and normalized weighted quality monitoring data are integrated to scientifically quantify the regional quality and overall quality status of steel structures. The comprehensive impact score and severity ranking are used to dynamically reflect the criticality of defects. Combined with the weight analysis of monitoring data, the accuracy and comprehensiveness of the evaluation results are ensured. It can not only identify potential weak areas of the structure, but also provide a scientific basis for the optimization of construction quality and the rational allocation of maintenance resources, thereby significantly improving the efficiency of construction quality monitoring and the reliability of decision-making.

[0185] Repair suggestion generation includes:

[0186] Determine fix priority: Calculate a fix priority score for each defect based on the area quality score, severity ranking, and defect severity score , expressed as:

[0187] ;

[0188] in, The defect area The weight of Rank the severity of the defect types, Rate the severity of the defect, Defects The remediation priority score of

[0189] Recommended repair method: Based on defect type and the geometric characteristics of the defect , combined with the preset repair method library, generate the corresponding repair plan, including the repair method, required materials and tools, specifically including:

[0190] (1) For welding defects, welding repair technology is recommended, and the amount of welding material required is calculated based on the defect length and width, expressed as:

[0191] ;

[0192] in, The demand for welding materials, is the density of the welding material, , are the length and width of the defect, respectively;

[0193] (2) For crack defects, recommend crack filling or reinforcement methods, and the required materials are calculated based on the crack area;

[0194] (3) Recommend correction and reinforcement techniques for dislocation and deformation defects, and provide necessary mechanical equipment and operating procedures;

[0195] Generate repair plan: Based on the repair priority score and recommended repair method, generate an overall repair plan, including a priority list of defect repairs, a list of materials and equipment required for each repair task, a repair duration forecast, and allocate time according to the repair task;

[0196] Through the above content, the results of the comprehensive quality assessment of the structure are comprehensively analyzed, the repair priority is scientifically determined, and specific repair plans are provided for different defect types, including the required materials, equipment and operating instructions. The generated repair plan covers priorities, resource allocation and repair time allocation to ensure the efficiency and accuracy of the repair work. It not only optimizes resource allocation and reduces unnecessary construction waste, but also significantly improves the timeliness and reliability of defect repair, thereby comprehensively ensuring the construction quality and long-term stability of the steel structure.

[0197] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0198] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A steel structure construction quality monitoring system based on drone and 3D scanning, characterized in that: It includes data acquisition module, data preprocessing module, construction defect identification and risk assessment module and quality assessment module, among which; The data acquisition module collects quality monitoring data of the construction site in real time, including three-dimensional data, image data, environmental monitoring data and geometric information of the steel structure at the construction site; The data preprocessing module preprocesses the collected quality monitoring data, including format conversion, denoising, missing value filling and standardization; The construction defect identification and risk assessment module identifies steel structure defects during the construction process based on the three-dimensional data and image data of the construction site, including welding defects, misalignment, deformation, and cracks, and assesses the severity of the steel structure defects, including: Preliminary data processing: Preliminary processing of the collected 3D data and image data of the construction site, including local contrast enhancement and edge enhancement processing; Construction defect identification: Based on the preliminarily processed 3D data and image data, various defects in the steel structure, including welding defects, misalignment, deformation, and cracks, are identified through a convolutional neural network model. The convolutional neural network model includes: Multi-channel input data fusion: The 3D point cloud data and 2D image data are processed separately through a multi-channel structure to extract the spatial features of the 3D point cloud data and the texture and brightness features of the 2D image data. The extracted spatial features, texture and brightness features are then fused to generate comprehensive features containing spatial and texture information. Deep residual feature extraction: Based on the fused comprehensive features, the residual network is introduced to extract deep and complex features; Feature weighting based on attention mechanism: Introduce channel attention mechanism, calculate the importance of each channel feature through global average pooling, and dynamically generate channel weights; Defect type classification: Classify steel structure defects through the fully connected layer, output the probability distribution of defect types, and indicate the prediction results of various defects; Defect severity assessment: Based on the various defects of the steel structure identified, the severity of each defect is assessed; The quality assessment module combines the results of steel structure defect identification and severity assessment, conducts a comprehensive quality assessment on the pre-processed quality monitoring data, generates a construction quality report, including the detected defect type, location, severity, and results of the comprehensive quality assessment, and provides corresponding repair suggestions based on the construction quality report.

2. The steel structure construction quality monitoring system based on drone and 3D scanning according to claim 1 is characterized in that: The data acquisition module comprises: 3D data collection: Use drones to use laser scanning to collect 3D data of the construction site in real time; Image data acquisition: Use camera equipment to capture image data of the construction site in real time; Environmental monitoring data collection: Deploy environmental sensors to monitor the environmental parameters of the construction site in real time, including temperature, humidity, air pressure and wind speed; Steel structure geometric information collection: Through laser scanning equipment, the geometric information of the steel structure can be obtained in real time, and the size, shape and relative position of each component of the steel structure can be monitored.

3. The steel structure construction quality monitoring system based on drone and 3D scanning according to claim 1 is characterized in that: The data preprocessing module comprises: Format conversion: convert quality monitoring data from different sources into a unified standard format, including 3D data into point cloud data format, image data into standard image file format, environmental monitoring data into time series data format, and geometric information of steel structures into BIM format; Denoising: Denoising the collected 3D data, image data and environmental monitoring data. Use Gaussian filter to denoise the 3D data, median filter to denoise the image data, and Kalman filter to denoise the environmental monitoring data. Missing value filling: Use linear interpolation to fill missing values ​​in the collected quality monitoring data; Standardization processing: Standardize the collected quality monitoring data.

4. The steel structure construction quality monitoring system based on drone and 3D scanning according to claim 1 is characterized in that: The preliminary data processing includes: Local contrast enhancement: Enhance the visibility of image details by enhancing the contrast of local areas; Edge enhancement processing: By using the Laplace operator for edge detection, the edge information in the image data is enhanced and the boundaries of the steel structure components are highlighted.

5. The steel structure construction quality monitoring system based on drone and 3D scanning according to claim 1 is characterized in that: The defect severity assessment includes: Defect data extraction and classification: Extract basic information of each defect from the results of construction defect identification, including defect type, defect location and geometric features; Geometric feature quantification: Based on the geometric features of each defect, the physical scale of the defect is quantified and the geometric score of the defect is generated. ; Position weight calculation: combined with the identified position of the defect , evaluate the potential impact of defects on structural stability and safety, and generate location weighted scores ; Comprehensive severity score: Based on the extracted geometric score and position weight score , calculate the composite severity score of the defect .

6. The steel structure construction quality monitoring system based on drone and 3D scanning according to claim 5 is characterized in that: The quality assessment module includes: Comprehensive analysis of defect information: Summarize and analyze the results of steel structure defect identification and severity assessment to determine the impact and severity ranking of each defect type; Comprehensive structural quality assessment: Combined with the results of comprehensive analysis of defect information, comprehensive analysis is performed on the pre-processed quality monitoring data to calculate the overall quality score of the steel structure; Repair suggestion generation: Generate targeted repair suggestions based on the results of comprehensive structural quality assessment.

7. The steel structure construction quality monitoring system based on drone and 3D scanning according to claim 6 is characterized in that: The comprehensive analysis of defect information includes: Defect type impact calculation: Quantify the impact of each defect type and calculate the comprehensive impact score of each type based on the severity score and the importance score of the defect location ; Defect severity ranking: comprehensive impact score based on defect type Sort all defect types and determine severity ranking .

8. The steel structure construction quality monitoring system based on drone and 3D scanning according to claim 7 is characterized in that: The comprehensive structural quality assessment includes: Integrated defect impact data: for steel structures by region Group the areas The results of comprehensive analysis of defect information within the area are integrated with the pre-processed quality monitoring data to generate the area Comprehensive data of Regional quality score calculation: Based on the integrated comprehensive data, the quality status of each region is quantitatively analyzed and the regional quality score is calculated. ; Overall quality score calculation: The quality score of each area is weighted according to the area weight to generate the overall quality score of the steel structure. .

9. The steel structure construction quality monitoring system based on drone and 3D scanning according to claim 8 is characterized in that: The repair suggestion generation includes: Determine fix priority: Calculate a fix priority score for each defect based on the area quality score, severity ranking, and defect severity score ; Recommended repair method: Based on defect type and the geometric characteristics of the defect , combined with the preset repair method library, generate the corresponding repair plan, including the repair method, required materials and tools; Generate repair plan: Based on the repair priority score and recommended repair method, generate an overall repair plan, including a priority list of defect repairs, a list of materials and equipment required for each repair task, a repair duration forecast, and allocate time by repair task.

Images