A mirror-finished fair-faced concrete quality detection method and system

Abstract

This invention relates to the field of building material quality testing technology, and particularly to a method and system for quality testing of mirror-finish fair-faced concrete. The method includes the following steps: S100, three-dimensional detection initiation; S200, multi-indicator collaborative detection in spatial dimensions; S300, dynamic attenuation analysis in the temporal dimension; S400, environmental adaptability calibration; S500, digital topology modeling; S600, comprehensive quality assessment. This invention also discloses a quality testing system for mirror-finish fair-faced concrete, including a three-dimensional holographic detection module; a dynamic tracking and environmental calibration module; a digital modeling engine; a comprehensive assessment platform; and a data interface module. This invention achieves quality misjudgment due to omissions in the quality assessment by simultaneously capturing multi-dimensional features such as flatness, gloss, color uniformity, and surface and near-surface air bubbles through four-dimensional fusion of "space-time-digital-environment," thus avoiding misjudgments caused by missing indicators.

Description

Technical Field

[0001] This invention relates to the field of building material quality testing technology, and in particular to a method and system for testing the quality of mirror-finish fair-faced concrete. Background Technology

[0002] Mirror-finish concrete, with its bright, mirror-like surface and delicate texture, is widely used in modern architectural decoration.

[0003] However, existing testing methods have significant limitations: First, the indicators are isolated, failing to consider the spatial correlation between smoothness and gloss, and the mutual influence between bubble distribution and color uniformity; second, they lack a dynamic tracking mechanism, failing to reflect the decay of concrete surface quality over time; third, the interference of environmental factors (temperature, humidity, etc.) on the test results is not calibrated, leading to insufficient data accuracy; and fourth, the digitization level of the test data is low, making it difficult to integrate with modern engineering management tools such as Building Information Modeling (BIM). Therefore, there is an urgent need for a technical solution that can achieve multi-dimensional, full-cycle, and high-precision testing. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a quality inspection method and system for mirror-finish fair-faced concrete.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A quality inspection method for mirror-finish fair-faced concrete includes the following steps:

[0007] S100, 3D Detection Startup: Determine the area to be detected, establish a 3D coordinate system with the lower left corner of the detection area as the origin, and simultaneously deploy spatial detection equipment, time tracking module, digital modeling tools and environmental sensors to complete the spatiotemporal reference calibration of the equipment.

[0008] S200, multi-indicator collaborative detection in spatial dimensions, as detailed below:

[0009] S201. Surface data is collected simultaneously using a laser flatness meter and a 60° gloss meter. The detection area is divided into 10cm×10cm grids, with 3 sampling points taken in each grid. A heat map of the global distribution of flatness and gloss is generated using the Kriging interpolation algorithm.

[0010] S202. Use a hyperspectral camera to collect surface reflectance spectra, and combine Image-proplus6.0 software to extract the reflectance characteristics of the red light band and calculate the color uniformity index.

[0011] S203. Based on the principle of topology graph construction, bubbles with a diameter > 0.5 mm are used as nodes and the distance between bubble centers is used as edges to establish a bubble distribution topology graph, and the node degree, average path length and clustering coefficient are statistically analyzed.

[0012] S204. Use an ultrasonic flaw detector to detect hidden bubbles within a 5mm range below the surface, and record the spatial relationship between the hidden bubbles and the surface bubbles.

[0013] S300, Dynamic decay analysis in time dimension: Spatial dimension detection was repeated at 1 day, 7 days, 28 days, 90 days and 180 days after pouring, the decay coefficient and decay acceleration of each index were calculated, and a secondary decay prediction model including time variables was established.

[0014] S400 Environmental Adaptability Calibration: The temperature, humidity and light intensity during the test are collected in real time by environmental sensors, and the gloss and color indexes are calibrated and corrected based on the preset environmental influence coefficient matrix.

[0015] S500, Digital Dimension Topological Modeling: Maps spatial detection data into 3D point clouds, generates digital twin models of a 0.5mm resolution base layer and a 2mm resolution application layer by referring to the normal distribution transformation map construction method, optimizes the bubble node topology structure through the repeated node semantic inference method, and realizes index mapping and fast switching of models with different resolutions;

[0016] S600, Comprehensive Quality Assessment: Based on spatial distribution heatmaps, time decay curves, environmental calibration values, and digital twin models, the weight of each indicator is calculated using the analytic hierarchy process (AHP), and the quality level and targeted optimization suggestions are output.

[0017] Preferably, the color uniformity index is calculated using the following formula: ,in, Let be the red band reflectance of the i-th sampling point. The average reflectance of all sampling points, where n is the total number of sampling points, and the U value ranges from 0 to 1. The closer it is to 1, the more uniform the color.

[0018] Preferably, the formula for calculating the attenuation coefficient k is: The formula for calculating the decay acceleration 'a' is: in, Let be the index value at time t. These are the initial indicator values ​​for an infant at 1 day old. , t1 and t2 are the decay coefficients, respectively. A negative k indicates that the index decays and a negative a indicates that the decay rate accelerates.

[0019] Preferably, the environmental impact coefficient matrix is ​​a 3×3 matrix. ,in Temperature influence coefficient, Humidity influence coefficient Light influence coefficient, gloss calibration value ,in To measure gloss, , , These represent the deviation values ​​between the measured environment and the standard environment, respectively.

[0020] Preferably, in the digital twin model, the base layer contains the original data of all sampling points, and the application layer aggregates the base layer data into 2mm×2mm×0.5mm voxel units using a voxel fusion algorithm. Each voxel unit stores the average flatness, average gloss, color uniformity, and bubble density parameters, which are then indexed. The two-layer model is linked and called, where x, y, and z are the coordinate values ​​of the voxel unit in the three-dimensional coordinate system, and x, y, and z correspond to the length direction, height direction, and normal depth direction of the detection area, respectively.

[0021] Preferably, in the bubble distribution topology graph, the semantic inference of repeated nodes adopts the adjacent node semantic intersection method.

[0022] This invention also proposes a quality inspection system for mirror-finish fair-faced concrete, comprising:

[0023] The 3D holographic inspection module consists of an integrated laser flatness meter, a 60° gloss meter, a hyperspectral camera, an ultrasonic flaw detector, and a 3D scanner. Each component is fixed on the same mobile platform by a rigid bracket and is synchronously triggered and data aggregated by a unified data acquisition controller to collect multi-indicator data in the spatial dimension.

[0024] Dynamic tracking and environmental calibration module: Installed near the detection area, and connected to the 3D holographic detection module via Ethernet TCP / IP protocol. The dynamic tracking and environmental calibration module includes temperature and humidity sensors, light sensors, timestamp units and calibration algorithm units, which are used to record time series and environmental parameters, and to calibrate gloss and color indices in real time.

[0025] Digital Modeling Engine: Deployed on a local server, it receives raw data uploaded by the 3D holographic detection module and the dynamic tracking and environmental calibration module through a high-speed data interface. The digital modeling engine has built-in normal distribution transformation algorithm and topology graph optimization module to convert point cloud data into multi-resolution digital twin models, and supports model compression and lightweight transmission.

[0026] Comprehensive evaluation platform: Adopting a B / S architecture, it is deployed in the central processing unit and connected to the digital modeling engine through a network interface, including but not limited to RJ45 network port and 4G / 5G network. The comprehensive evaluation platform includes an AHP weight calculation module, a quality level judgment module and a visualization display module, which supports three-dimensional visualization of detection data and historical data backtracking analysis.

[0027] Data interface module: Integrated into the comprehensive evaluation platform, it supports data interaction with the BIM platform and construction management system, enabling real-time sharing of test results.

[0028] Preferably, the sampling frequency of the three-dimensional holographic detection module is not less than 10Hz and the scanning speed is not less than 0.5m² / min, ensuring that all indicators of the 10m² detection area are collected within 2 hours.

[0029] Preferably, the visualization module of the comprehensive evaluation platform supports dynamic rendering of bubble topology diagrams, simulation of attenuation curve prediction, and cross-sectional analysis of digital twin models, and can achieve precise positioning and detailed viewing of the detection area through touch operation.

[0030] The present invention has the following beneficial effects:

[0031] 1. By integrating four dimensions of "space-time-digital-environment", it simultaneously captures multi-dimensional features such as flatness, gloss, color uniformity, and surface and near-surface bubbles, avoiding quality misjudgments due to omission of indicators; it covers the entire age period after concrete pouring, from short-term to long-term, tracking the natural evolution of quality over time, rather than relying solely on a single test; it specifically eliminates the interference of external factors such as temperature, humidity, and light on optical indicators, ensuring the consistency of test results under different scenarios.

[0032] 2. Utilize spatial collaborative detection technology to establish a spatial correlation model between flatness and gloss (such as the correspondence between flatness deviation and gloss decay in a certain area); through bubble topology analysis, treat bubbles as "nodes" and bubble spacing as "edges" to quantify the impact of bubble distribution aggregation on the overall texture; combined with a digital twin model, intuitively display the distribution and correlation of various indicators in three-dimensional space to help engineers locate the root cause of quality defects (such as the correspondence between template splicing seams and bubble aggregation areas).

[0033] 3. Build multi-resolution digital twin models that retain the detail and precision of high-resolution models (such as microbubbles) while meeting engineering collaboration needs through a lightweight application layer, enabling flexible switching between "details" and "global" perspectives; the model can be seamlessly integrated with Building Information Modeling (BIM) to link inspection data to specific components, supporting full lifecycle quality traceability from construction acceptance, operation and maintenance to renovation; based on time-dimensional decay analysis, a quality evolution curve is generated, providing forward-looking guidance for later maintenance strategies (such as surface curing and defect repair), avoiding passive responses to quality problems.

[0034] 4. The automated testing process reduces manual intervention and operational errors, while improving testing efficiency and meeting the timeliness requirements of project acceptance; the built-in intelligent evaluation logic, combined with preset quality level standards, automatically outputs clear quality conclusions and targeted optimization suggestions (such as adjustment of formwork sealing process, optimization of vibration parameters, etc.), directly guiding on-site construction improvements; it can adapt to the testing needs of different formwork solutions (such as acrylic sheets, formwork paint), accurately identify the quality differences of different processes, and provide an objective basis for project selection. Attached Figure Description

[0035] Figure 1 This is a flowchart illustrating a quality inspection method for mirror-finish fair-faced concrete proposed in this invention.

[0036] Figure 2 This is a logical structure block diagram of a quality inspection system for mirror-finish fair-faced concrete proposed in this invention. Detailed Implementation

[0037] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0038] Reference Figure 1 A quality inspection method for mirror-finish fair-faced concrete includes the following steps:

[0039] S100, 3D Detection Startup: Determine the area to be detected, establish a 3D coordinate system with the lower left corner of the area as the origin (X-axis along the length direction, Y-axis along the height direction, Z-axis perpendicular to the surface), and simultaneously deploy spatial detection equipment, time tracking module, digital modeling tools and environmental sensors to complete the spatiotemporal reference calibration of the equipment.

[0040] Specifically, define the scope of inspection: delineate the inspection area according to the project requirements (such as the entire surface of the frame column or a specific area of ​​the wall), mark the boundary of the area with ink lines (error ≤ 5mm), and record the actual dimensions of the area (length × height).

[0041] Pretreatment: Remove surface dust, water stains and other interfering substances before testing. Mark obvious protrusions (such as concrete nodules) (circle them with a red marker) for separate analysis later.

[0042] Origin setting: The lower left corner of the detection area (the intersection of the ground and the boundary of the area) is set as the origin (0,0,0).

[0043] Shaft system definition:

[0044] X-axis: Along the length of the detection area (horizontally to the right), in meters (m);

[0045] Y-axis: along the height direction of the detection area (vertically upward), in meters (m);

[0046] Z-axis: Perpendicular to the concrete surface (pointing outwards is positive), unit is millimeters (mm, used to characterize the surface normal depth, such as bubble depth, flatness deviation).

[0047] Coordinate marking: Set coordinate reference points (using stainless steel markers) every 1m at the boundary of the area, and calibrate them with a total station (accuracy ±0.5mm) to ensure the orthogonality of the axis system (deviation ≤0.1°).

[0048] S200, multi-indicator collaborative detection in spatial dimensions, as detailed below:

[0049] S201. Surface data is simultaneously collected using a laser flatness meter (accuracy ±0.01mm) and a 60° gloss meter. The detection area is divided into 10cm×10cm grids, with 3 sampling points taken from each grid. A heat map of the overall distribution of flatness and gloss is generated using the Kriging interpolation algorithm. Specifically, a standard calibration block (known size 100mm×100mm×5mm) is placed in the center of the detection area. The laser flatness meter and the 3D scanner are used for measurement. The deviation must be ≤0.02mm; otherwise, the equipment position is adjusted. A hyperspectral camera is used to photograph a standard color chart (X-RiteColorChecker), and the white balance and spectral baseline are calibrated using Image-proplus 6.0.

[0050] The actual operation is as follows:

[0051] Grid generation:

[0052] Use a measuring tape and ink line to mark a 10cm×10cm grid line in the inspection area (error ≤1mm), and mark each grid vertex with a white marker (diameter ≤5mm).

[0053] Sampling point settings:

[0054] Three sampling points are taken within each grid: one point at the center and two points at the 1 / 4 mark of the diagonal (coordinates are (x+2.5cm, y+2.5cm) and (x+7.5cm, y+7.5cm) at the bottom left corner of the grid).

[0055] Synchronous data acquisition:

[0056] Start the electric slide rail, and the laser flatness meter and gloss meter move along the X-axis (speed 5cm / s), automatically stopping for 0.5s to collect data at each sampling point;

[0057] Record the 3D coordinates (X,Y,Z), flatness value (Z-direction deviation, mm), and gloss value (GU) of each point, and save them as CSV format (fields: X,Y,Z, flatness, gloss, timestamp).

[0058] Heatmap generation:

[0059] Import the data into ArcGIS or MATLAB and use the Kriging interpolation algorithm (select Gaussian model for the variogram, search radius 5cm).

[0060] Generate a global heatmap of flatness (color level: blue ≤ 1 mm, red ≥ 3 mm) and gloss (color level: red ≥ 40 GU, blue ≤ 20 GU) with a resolution of 1 mm × 1 mm.

[0061] S202. Use a hyperspectral camera (400-700nm band, spectral resolution 1nm) to collect surface reflectance spectra, and combine Image-proplus6.0 software to extract reflectance characteristics in the red band (630-670nm) and calculate the color uniformity index.

[0062] The specific steps are as follows:

[0063] Hyperspectral image acquisition: Shoot in 50cm×50cm sections, with each section overlapping by 10cm (to avoid stitching errors), and an exposure time of 50ms (adjusted according to lighting conditions to ensure no overexposure); after every 3 areas are shot, calibrate the spectral baseline once using a standard white board (99% reflectivity).

[0064] Reflectance extraction: Import Image-proplus6.0, crop out the detection area (remove background), and convert the image into reflectance data (400-700nm); extract the reflectance of the red light band (630-670nm): for each 10cm×10cm grid, calculate the average reflectance (R_i) of 3 sampling points (n≥300, covering the entire area).

[0065] Color uniformity index is calculated using the following formula: ,in, Let be the reflectance of the red light band (630-670nm) at the i-th sampling point. The average reflectance of all sampling points, where n is the total number of sampling points (n≥300), and the U value ranges from 0 to 1, with the closer to 1 indicating a more uniform color.

[0066] S203. Based on the principle of topology graph construction, bubbles with a diameter > 0.5 mm are used as nodes and the distance between bubble centers is used as edges to establish a bubble distribution topology graph, and the node degree, average path length and clustering coefficient are statistically analyzed.

[0067] Specifically, this is achieved through the following operations:

[0068] Bubble recognition: 3D scanner point cloud data is imported into Geomagic Studio, and surface bubbles (diameter > 0.5mm) are identified through the "curvature mutation" algorithm. The center coordinates (X,Y,Z), diameter (d), and depth (z, Z axis value) of each bubble are recorded.

[0069] Manual verification: 10% of the bubbles identified by the algorithm are randomly checked using a vernier caliper (accuracy 0.01mm), and the deviation must be ≤0.1mm.

[0070] Topology graph construction: Node definition: Each bubble is a node, with attributes including diameter, depth, and coordinates;

[0071] Edge definition: When the distance between the centers of two bubbles is less than 5cm, an edge is created (the weight is the distance value).

[0072] Tools: Use Python's networkx library to build the topology graph and save it as a .graphml file.

[0073] Topology parameter calculation:

[0074] Node degree: The number of edges connected to a single bubble (reflecting the density of surrounding bubbles);

[0075] Average path length: The average of the shortest paths between all node pairs (reflecting the dispersion of bubble distribution);

[0076] Clustering coefficient: The ratio of the actual number of edges between a node's neighbors to the possible number of edges (reflecting the degree of clustering).

[0077] Semantic inference of repeated nodes:

[0078] If duplicate bubble nodes exist at the same location (e.g., due to scanning errors), the semantic intersection of the preceding and following nodes is taken. For example: preceding node "diameter 2-3mm, depth 0.5mm" + following node "diameter 2-3mm, depth 1mm" → target semantic "diameter 2-3mm, depth 0.5-1mm".

[0079] S204. Use an ultrasonic flaw detector (detection depth 0-50mm) to detect hidden bubbles within a 5mm range below the surface, and record the spatial relationship between the hidden bubbles and the surface bubbles.

[0080] Specifically, ultrasonic testing is used: Probe selection: 5MHz straight probe (10mm in diameter), coupling agent is machine oil (to ensure no air bubbles); Detection range: 0-5mm below the surface (Z-axis -0.1mm to -5mm), stop and test at 3 sampling points in every 10cm×10cm grid, and record the center coordinates (X,Y,Z) and diameter of the hidden bubbles.

[0081] Spatial correlation analysis: Calculate the straight-line distance between the hidden bubble and the nearest surface bubble. If it is ≤2cm, it is determined to be a "surface-hidden bubble cluster" (a causal relationship exists); the formula for calculating the proportion of hidden bubbles is: ;

[0082] S300, Dynamic Attenuation Analysis in Time Dimension: Spatial dimension detection was repeated 1 day (24±2h after pouring), 7 days (168±4h), 28 days (672±8h), 90 days (2160±24h), and 180 days (4320±48h) after pouring, and the attenuation coefficient and attenuation acceleration of each index were calculated to establish a secondary attenuation prediction model that includes time variables;

[0083] The formula for calculating the attenuation coefficient k is: The formula for calculating the decay acceleration 'a' is: in, Let be the index value at time t. These are the initial indicator values ​​for an infant at 1 day old. , t1 and t2 are the decay coefficients, respectively. A negative k indicates that the index decays and a negative a indicates that the decay rate accelerates.

[0084] The form of the secondary decay prediction model: , , Fitting coefficients) Fitting method: Use the least squares method to fit the 1-day, 7-day, and 28-day data, and verify with 90-day data (deviation must be ≤5%); Application: Predict 180-day index values ​​and determine whether they meet long-term use requirements.

[0085] S400 Environmental Adaptability Calibration: Temperature (0-50℃, accuracy ±0.5℃), humidity (20%-95%RH, accuracy ±5%RH), and light intensity (0-10000lux, accuracy ±100lux) are collected in real-time by environmental sensors during testing. It should be noted that temperature T, humidity H, and light intensity L are recorded simultaneously, and the average value is stored every 5 minutes (to avoid instantaneous fluctuations). Gloss and color indicators are calibrated and corrected based on a preset environmental influence coefficient matrix.

[0086] The environmental impact coefficient matrix is ​​a 3×3 matrix. ,in The temperature effect coefficient (-0.005 / ℃) Humidity effect coefficient (-0.002 / %RH) The light effect factor (0.001 / lux) is the gloss calibration value. ,in To measure gloss, , , These represent the deviation values ​​between the measured environment and the standard environment, respectively.

[0087] The standard environment is: 25℃ (Temperature T), 50%RH (Humidity H), and 5000 lux (Light Intensity L).

[0088] The deviation value is: .

[0089] S500, Digital Dimension Topological Modeling: Maps spatial detection data into a 3D point cloud (point spacing 0.1mm), and generates digital twin models of a 0.5mm resolution base layer and a 2mm resolution application layer by referring to the normal distribution transformation map construction method. The topological structure of bubble nodes is optimized by the semantic inference method of repeated nodes, realizing index mapping and fast switching of models with different resolutions.

[0090] Specifically, the 3D point cloud generation process is as follows:

[0091] 3D scanner data stitching: Use Cyclone software to align multi-station scan data (stitching error ≤ 0.3mm) and remove noise points (retain points with confidence ≥ 95%).

[0092] Point cloud attribute mapping: Associating flatness, gloss, reflectivity, bubble parameters, etc., with corresponding 3D coordinate points to generate point clouds with attributes (.las format).

[0093] In the digital twin model, the base layer contains the original data of all sampling points, retains all point cloud data, and includes the original index value of each 0.5mm×0.5mm×0.1mm unit; the base layer is used for high-precision defect analysis (such as microbubble localization).

[0094] The application layer aggregates the base layer data into 2mm×2mm×0.5mm voxel units using a voxel fusion algorithm. Each unit stores the following data: average flatness (average Z-deviation of all points within the unit); average gloss (average GU value after calibration); color uniformity (U-value within the unit); and bubble density (number of bubbles within the unit / unit volume). The application layer is used for lightweight display and engineering collaboration. (Index value is used for...) Implement the association and call between the two-layer model. Here, / / represents integer division, index value. The unit is mm. x, y, and z are the coordinates of the voxel unit in the three-dimensional coordinate system, corresponding to the length direction, height direction, and normal depth direction perpendicular to the surface of the detection area, respectively.

[0095] S600 Comprehensive Quality Assessment: Based on spatial distribution heatmaps, time decay curves, environmental calibration values, and digital twin models, the Analytic Hierarchy Process (AHP) is used to calculate the weights of each indicator (flatness 0.25, gloss 0.3, color 0.2, bubble parameter 0.25), outputting quality grades (excellent, good, qualified, unqualified) and targeted optimization suggestions.

[0096] Defect areas are located using a digital twin model, and the causes are analyzed by combining topological maps and attenuation curves. For example, if bubbles are clustered in the central area (clustering coefficient > 0.7) and the proportion of hidden bubbles is high, it is recommended to optimize the sealing of the template splice joints (widen the sealing tape) and extend the vibration time by 5-10 seconds; while if the gloss decays rapidly (k < -0.02 / day), it is recommended to increase the number of times the surface curing agent is applied (from 2 to 3 coats).

[0097] In addition, the quality grade judgment criteria are as follows: Excellent requires flatness ≤2mm, gloss ≥35GU, color uniformity index ≥0.9, maximum diameter of surface bubbles ≤3mm and hidden bubble ratio <5%; Good requires flatness ≤3mm, gloss ≥25GU, color uniformity index ≥0.8, maximum diameter of surface bubbles ≤5mm and hidden bubble ratio <10%; Qualified requires flatness ≤4mm, gloss ≥15GU, color uniformity index ≥0.7, maximum diameter of surface bubbles ≤8mm and hidden bubble ratio <15%; failure to meet the qualified criteria is considered unqualified.

[0098] Reference Figure 2 The present invention also proposes a quality inspection system for mirror-finish fair-faced concrete, comprising:

[0099] The 3D holographic inspection module consists of an integrated laser flatness meter, a 60° gloss meter, a hyperspectral camera, an ultrasonic flaw detector, and a 3D scanner. Each component is fixed on the same mobile platform by a rigid bracket and is synchronously triggered and data aggregated by a unified data acquisition controller to collect multi-indicator data in the spatial dimension.

[0100] Dynamic tracking and environmental calibration module: Installed near the detection area, and connected to the 3D holographic detection module via Ethernet TCP / IP protocol. The dynamic tracking and environmental calibration module includes temperature and humidity sensors, light sensors, timestamp units and calibration algorithm units, which are used to record time series and environmental parameters, and to calibrate gloss and color indices in real time.

[0101] Digital Modeling Engine: Deployed on a local server, it receives raw data uploaded by the 3D holographic detection module and the dynamic tracking and environmental calibration module through a high-speed data interface. The digital modeling engine has built-in normal distribution transformation algorithm and topology graph optimization module to convert point cloud data into multi-resolution digital twin models, and supports model compression and lightweight transmission.

[0102] Comprehensive evaluation platform: Adopting a B / S architecture, it is deployed in the central processing unit and connected to the digital modeling engine through a network interface, including but not limited to RJ45 network port and 4G / 5G network. The comprehensive evaluation platform includes an AHP weight calculation module, a quality level judgment module and a visualization display module, which supports three-dimensional visualization of detection data and historical data backtracking analysis.

[0103] Data interface module: Integrated into the comprehensive evaluation platform, it supports data interaction with the BIM platform and construction management system, enabling real-time sharing of test results.

[0104] The 3D holographic detection module has a sampling frequency of no less than 10Hz and a scanning speed of no less than 0.5m² / min, ensuring that all indicators of a 10m² detection area are collected within 2 hours. The visualization module of the comprehensive evaluation platform supports dynamic rendering of bubble topology diagrams, attenuation curve prediction simulation, and digital twin model sectioning analysis. The detection area can be accurately located and viewed in detail through touch operation.

[0105] Furthermore, this invention selects six C40 mirror-finished fair-faced concrete frame columns (KZ1-KZ6, dimensions 1.6m × 1.6m, height 5.505m) on the concourse level of an intercity railway station. KZ1-KZ3 employ a "steel formwork + 1.5mm acrylic sheet" formwork scheme, while KZ4-KZ6 employ a "steel formwork + HD-6 formwork paint" scheme. The following embodiments and comparative examples are designed:

[0106] Example 1: Steel formwork + acrylic sheet (method of this invention)

[0107] Inspection object: KZ1 frame column (1.5mm acrylic sheet inside steel formwork, laser-cut and spliced).

[0108] Detection method:

[0109] Spatial dimensions: Laser flatness meter + 60° gloss meter (10cm×10cm grid, 3 points / grid), hyperspectral camera for color analysis (630-670nm band), ultrasonic flaw detector for detecting hidden bubbles;

[0110] Time dimension: Four time points for detection: 1 day, 28 days, 90 days, and 180 days;

[0111] Environmental calibration: During testing, the temperature is 25℃ and the humidity is 50%, according to the formula. Calibrate gloss, among which Temperature influence coefficient, Humidity influence coefficient The light influence coefficient is denoted as , where To measure gloss, , , These represent the deviation values ​​between the measured environment and the standard environment, respectively.

[0112] Digital modeling: A 0.5mm resolution digital twin model was generated, and the bubble topology was optimized. The results are shown in the table below:

[0113] Table 1: Test Results of Various Parameters of C40 Mirror-finished Fair-faced Concrete Frame Columns in Example 1

[0114]

[0115] Example 2: Steel formwork + HD-6 formwork paint (method of this invention)

[0116] Test object: KZ4 frame column (steel formwork inside coated with HD-6 formwork paint, applied by roller).

[0117] Detection method: Same as in Example 1.

[0118] The test results are shown in the table below:

[0119] Table 2: Test Results of Various Parameters of C40 Mirror-finished Fair-faced Concrete Frame Columns in Example 2

[0120]

[0121] The comparative example is the KZ2 frame column (with the same template scheme) in the same area as in Example 1. The detection method is traditional manual inspection (flatness measurement with straightedge and feeler gauge, single-point gloss meter), without environmental calibration, time tracking and hidden bubble detection.

[0122] Table 3: Test Results of Various Parameters of Comparative Example C40 Mirror-finished Fair-faced Concrete Frame Columns

[0123]

[0124] Based on the above embodiments and comparative examples, it is easy to see that in terms of detection comprehensiveness: Embodiments 1 and 2 detect 6 core indicators using the method of the present invention, covering explicit / latent defects, time decay, and environmental influences; while the comparative example only detects 2 indicators, missing key parameters such as color uniformity and latent bubbles, resulting in a significant bias in the evaluation.

[0125] Regarding data accuracy: In Example 1, environmental calibration kept the gloss error within ±1 GU and the flatness measurement deviation ≤0.03 mm; the comparative example, due to lack of calibration and low sampling density, had an error of 6.6%-25%, which could not reflect the true quality status.

[0126] Regarding the effectiveness in the time dimension: Examples 1 and 2 established a decay model through 180-day tracking, predicting that the gloss deviation after 180 days would be ≤1%; the comparative example did not have time tracking, so it was impossible to assess long-term performance, and there was a risk of quality problems in the later stage (such as the exposure of hidden bubbles).

[0127] Regarding the adaptability of different template schemes: Example 1 (acrylic sheet) is superior to Example 2 (HD-6 template paint) in terms of flatness, gloss and bubble control, verifying that the present invention can accurately distinguish the quality differences of different construction schemes; the test data of Example 2 provides a basis for template scheme optimization (such as increasing the coating thickness of HD-6 template paint to reduce bubbles).

[0128] In terms of efficiency and cost: The detection time of the method of the present invention (Example 1) is 1.5 hours per column, which saves 40% of the time compared with the traditional method (comparative example); the digital twin model can be repeatedly called to avoid repeated detection, and the long-term cost is reduced by more than 30%.

[0129] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for quality inspection of mirror-finish fair-faced concrete, characterized in that, Includes the following steps: S100, 3D Detection Startup: Determine the area to be detected, establish a 3D coordinate system with the lower left corner of the detection area as the origin, and simultaneously deploy spatial detection equipment, time tracking module, digital modeling tools and environmental sensors to complete the spatiotemporal reference calibration of the equipment. S200, multi-indicator collaborative detection in spatial dimensions, as detailed below: S201. Surface data is collected simultaneously using a laser flatness meter and a 60° gloss meter. The detection area is divided into 10cm×10cm grids, with 3 sampling points taken in each grid. A heat map of the global distribution of flatness and gloss is generated using the Kriging interpolation algorithm. S202. Use a hyperspectral camera to collect surface reflectance spectra, and combine Image-proplus6.0 software to extract the reflectance characteristics of the red light band and calculate the color uniformity index. S203. Based on the principle of topology graph construction, bubbles with a diameter > 0.5 mm are used as nodes and the distance between bubble centers is used as edges to establish a bubble distribution topology graph, and the node degree, average path length and clustering coefficient are statistically analyzed. S204. Use an ultrasonic flaw detector to detect hidden bubbles within a 5mm range below the surface, and record the spatial relationship between the hidden bubbles and the surface bubbles. S300, Dynamic decay analysis in time dimension: Spatial dimension detection was repeated at 1 day, 7 days, 28 days, 90 days and 180 days after pouring, the decay coefficient and decay acceleration of each index were calculated, and a secondary decay prediction model including time variables was established. S400 Environmental Adaptability Calibration: The temperature, humidity and light intensity during the test are collected in real time by environmental sensors, and the gloss and color indexes are calibrated and corrected based on the preset environmental influence coefficient matrix. S500, Digital Dimension Topological Modeling: Maps spatial detection data into 3D point clouds, generates digital twin models of a 0.5mm resolution base layer and a 2mm resolution application layer by referring to the normal distribution transformation map construction method, optimizes the bubble node topology structure through the repeated node semantic inference method, and realizes index mapping and fast switching of models with different resolutions; S600, Comprehensive Quality Assessment: Based on spatial distribution heatmaps, time decay curves, environmental calibration values, and digital twin models, the weight of each indicator is calculated using the analytic hierarchy process (AHP), and the quality level and targeted optimization suggestions are output.

2. The method for quality inspection of mirror-finish fair-faced concrete according to claim 1, characterized in that, The color uniformity index is calculated using the following formula: ,in, Let be the red band reflectance of the i-th sampling point. The average reflectance of all sampling points, where n is the total number of sampling points, and the U value ranges from 0 to 1. The closer it is to 1, the more uniform the color.

3. The quality inspection method for mirror-finish fair-faced concrete according to claim 1, characterized in that, The formula for calculating the attenuation coefficient k is: The formula for calculating the decay acceleration 'a' is: in, Let be the index value at time t. These are the initial indicator values ​​for an infant at 1 day old. , t1 and t2 are the decay coefficients, respectively. A negative k indicates that the index decays and a negative a indicates that the decay rate accelerates.

4. The quality inspection method for mirror-finish fair-faced concrete according to claim 1, characterized in that, The environmental impact coefficient matrix is ​​a 3×3 matrix. ,in Temperature influence coefficient, Humidity influence coefficient Light influence coefficient, gloss calibration value ,in To measure gloss, , , These represent the deviation values ​​between the measured environment and the standard environment, respectively.

5. The quality inspection method for mirror-finish fair-faced concrete according to claim 1, characterized in that, In the digital twin model, the base layer contains the original data of all sampling points. The application layer aggregates the base layer data into 2mm×2mm×0.5mm voxel units using a voxel fusion algorithm. Each voxel unit stores the average flatness, average gloss, color uniformity, and bubble density parameters, which are then indexed. The two-layer model is linked and called, where x, y, and z are the coordinate values ​​of the voxel unit in the three-dimensional coordinate system, in millimeters. x, y, and z correspond to the length direction, height direction, and normal depth direction perpendicular to the surface of the detection area, respectively.

6. The quality inspection method for mirror-finish fair-faced concrete according to claim 1, characterized in that, In the bubble distribution topology graph, the semantic inference of repeated nodes is performed using the semantic intersection method of adjacent nodes.

7. A quality inspection system for mirror-finished fair-faced concrete for implementing the method of any one of claims 1-6, characterized in that, include: The 3D holographic inspection module consists of an integrated laser flatness meter, a 60° gloss meter, a hyperspectral camera, an ultrasonic flaw detector, and a 3D scanner. Each component is fixed on the same mobile platform by a rigid bracket and is synchronously triggered and data aggregated by a unified data acquisition controller to collect multi-indicator data in the spatial dimension. Dynamic tracking and environmental calibration module: Installed near the detection area, and connected to the 3D holographic detection module via Ethernet TCP / IP protocol. The dynamic tracking and environmental calibration module includes temperature and humidity sensors, light sensors, timestamp units and calibration algorithm units, which are used to record time series and environmental parameters, and to calibrate gloss and color indices in real time. Digital Modeling Engine: Deployed on a local server, it receives raw data uploaded by the 3D holographic detection module and the dynamic tracking and environmental calibration module through a high-speed data interface. The digital modeling engine has built-in normal distribution transformation algorithm and topology graph optimization module to convert point cloud data into multi-resolution digital twin models, and supports model compression and lightweight transmission. Comprehensive evaluation platform: Adopting a B / S architecture, it is deployed in the central processing unit and connected to the digital modeling engine through a network interface, including but not limited to RJ45 network port and 4G / 5G network. The comprehensive evaluation platform includes an AHP weight calculation module, a quality level judgment module and a visualization display module, which supports three-dimensional visualization of detection data and historical data backtracking analysis. Data interface module: Integrated into the comprehensive evaluation platform, it supports data interaction with the BIM platform and construction management system, enabling real-time sharing of test results.

8. The quality inspection system for mirror-finish fair-faced concrete according to claim 7, characterized in that, The sampling frequency of the three-dimensional holographic detection module is no less than 10Hz, and the scanning speed is no less than 0.5m² / min, ensuring that all indicators of the 10m² detection area are collected within 2 hours.

9. The quality inspection system for mirror-finish fair-faced concrete according to claim 7, characterized in that, The visualization module of the comprehensive evaluation platform supports dynamic rendering of bubble topology diagrams, simulation of attenuation curve prediction, and cross-sectional analysis of digital twin models. It can achieve precise positioning and detailed viewing of the detection area through touch operation.

Images