Quality detection method and system for mirror-surface bare concrete

Abstract

The invention relates to the technical field of quality detection of building materials, in particular to a quality detection method and system for mirror-surface bare concrete. The quality detection method for the mirror-surface bare concrete comprises the following steps: S100, starting three-dimensional detection; s200, carrying out spatial dimension multi-index cooperative detection; s300, time dimension dynamic attenuation analysis is carried out; s400, carrying out environmental adaptability calibration; s500, carrying out digital dimension topological modeling; and S600, carrying out comprehensive quality evaluation. The invention further discloses a system for detecting the quality of the mirror-surface bare concrete. The system comprises a three-dimensional holographic detection module; a dynamic tracking and environment calibration module; a digital modeling engine; a comprehensive evaluation platform; and a data interface module. According to the method, through four-dimensional fusion of space-time-number-environment, multi-dimensional features such as flatness, glossiness, color uniformity and surface and near-surface bubbles are synchronously captured, and quality misjudgment caused by index omission is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of building material quality detection, and in particular to a quality detection method and system for mirror-surface plain concrete. Background Art

[0002] Mirror-finished concrete is widely used in modern architectural decoration due to its bright mirror-like surface and delicate texture.

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

[0004] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a quality detection method and system for mirror-finished plain concrete.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions: A quality inspection method for mirror-finished plain concrete comprises the following steps: S100, 3D detection start: 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 modules, digital modeling tools, and environmental sensors to complete the equipment's time and space reference calibration; S200, spatial dimension multi-index collaborative detection, details are as follows: S201. Use a laser flatness meter and a 60° gloss meter to synchronously collect surface data. Divide the test area into 10 cm × 10 cm grids, take three sampling points in each grid, and generate a heat map of the global distribution of flatness and gloss using the Kriging interpolation algorithm. S202. Use a hyperspectral camera to collect the surface reflectance spectrum, use Image-proplus6.0 software to extract the red light band reflectance characteristics, and calculate the color uniformity index; S203. Based on the principle of topological graph construction, bubbles with a diameter greater than 0.5 mm are used as nodes and the distance between bubble centers as edges to establish a bubble distribution topological graph, and calculate the node degree, average path length, and clustering coefficient. S204. Detect hidden bubbles within 5 mm below the surface using an ultrasonic flaw detector, and record the spatial position correlation between the hidden bubbles and the surface bubbles; S300, dynamic attenuation analysis in time dimension: Repeat the spatial dimension test 1 day, 7 days, 28 days, 90 days, and 180 days after pouring, calculate the attenuation coefficient and attenuation acceleration of each indicator, and establish a quadratic attenuation prediction model that includes the time variable; S400, Environmental Adaptability Calibration: The environmental sensor collects the temperature, humidity and light intensity during the test in real time, and calibrates and corrects the gloss and color indexes based on the preset environmental impact coefficient matrix; S500, Digital Dimension Topological Modeling: Maps spatial detection data into a 3D point cloud. Based on the normal distribution transformation map construction method, a digital twin model with a 0.5mm resolution base layer and a 2mm resolution application layer is generated. The bubble node topology is optimized through repeated node semantic inference, enabling index mapping and rapid switching between models of different resolutions. S600, comprehensive quality assessment: Based on spatial distribution heat maps, time decay curves, environmental calibration values ​​and digital twin models, the hierarchical analysis method is used to calculate the weights of each indicator, and output quality grades and targeted optimization suggestions.

[0006] Preferably, the color uniformity index is calculated by the following formula: ,in, is the red light band reflectivity of the i-th sampling point, The average reflectance of all sampling points, n is the total number of sampling points, and the U value range is 0-1. The closer to 1, the more uniform the color.

[0007] Preferably, the calculation formula of the attenuation coefficient k is: , the calculation formula of the attenuation acceleration a is: in, is the index value at time t, is the initial index value of 1 day old, 、 are the attenuation coefficients at time t1 and t2 respectively. A negative k indicates that the indicator is attenuating, and a negative a indicates that the attenuation speed is accelerating.

[0008] Preferably, the environmental impact coefficient matrix is ​​a 3×3 matrix ,in is the temperature influence coefficient, is the humidity influence coefficient, is the lighting influence coefficient, gloss calibration value ,in is the measured glossiness, 、 、 are the deviation values ​​between the measured environment and the standard environment respectively.

[0009] 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 through the voxel fusion algorithm. Each voxel unit stores the flatness mean, gloss mean, color uniformity and bubble density parameters, and the index value is used to calculate the average value of the flatness, glossiness, color uniformity and bubble density. Realize the associated call of the two-layer model, 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 perpendicular to the surface of the detection area respectively.

[0010] Preferably, in the bubble distribution topology diagram, semantic inference of repeated nodes adopts the adjacent node semantic intersection method.

[0011] The present invention also proposes a mirror surface plain concrete quality detection system, comprising: 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 to the same mobile platform via a rigid bracket and synchronized with a unified data acquisition controller for data collection and multi-indicator data collection in spatial dimensions. 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 a temperature and humidity sensor, a light sensor, a timestamp unit, and a calibration algorithm unit. It is used to record time series and environmental parameters and perform real-time calibration of gloss and color indicators. 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 a built-in normal distribution transformation algorithm and topology optimization module to convert point cloud data into a multi-resolution digital twin model, supporting model compression and lightweight transmission. Comprehensive evaluation platform: This platform adopts a B / S architecture, is deployed in the central processing unit, and is connected to the digital modeling engine through a network interface, including but not limited to an RJ45 network port and a 4G / 5G network. The comprehensive evaluation platform includes an AHP weight calculation module, a quality grade determination module, and a visualization display module, supporting 3D visualization of test data and retrospective analysis of historical data. Data interface module: integrated into the comprehensive assessment platform, supports data interaction with the BIM platform and construction management system, and realizes real-time sharing of test results.

[0012] Preferably, the sampling frequency of the three-dimensional holographic detection module is not less than 10 Hz, and the scanning speed is not less than 0.5 m² / min, ensuring that all indicators of the 10 m² detection area are collected within 2 hours.

[0013] Preferably, the visualization display module of the comprehensive evaluation platform supports dynamic rendering of bubble topology maps, attenuation curve prediction simulation and digital twin model section analysis, and can achieve precise positioning and detailed viewing of the detection area through touch operation.

[0014] The present invention has the following beneficial effects: 1. Through the four-dimensional fusion of "space-time-digital-environment", multi-dimensional characteristics such as flatness, glossiness, color uniformity, surface and near-surface bubbles are simultaneously captured to avoid quality misjudgment due to missed indicators; throughout the entire age period from short-term to long-term after concrete pouring, the natural evolution of quality over time is tracked, rather than relying solely on a single test; targeted elimination of interference from external factors such as temperature, humidity, and light on optical indicators ensures the consistency of test results in different scenarios.

[0015] 2. Utilize spatial collaborative detection technology to establish a spatial correlation model between flatness and glossiness (e.g., the correspondence between flatness deviation and gloss attenuation in a certain area). Through bubble topology analysis, consider bubbles as "nodes" and bubble spacing as "edges" to quantify the impact of bubble distribution aggregation on overall texture. Combined with the digital twin model, the distribution correlation of various indicators in three-dimensional space is intuitively displayed to help engineers locate the root cause of quality defects (e.g., the correspondence between template joints and bubble aggregation areas).

[0016] 3. Build a multi-resolution digital twin model that retains the detail accuracy of the high-resolution model (such as tiny bubbles) while meeting engineering collaboration needs through a lightweight application layer, achieving flexible switching between "details and the whole picture"; the model can be seamlessly connected with the building information model (BIM), linking inspection data to specific components, and supporting quality traceability throughout the entire life cycle from construction acceptance, operation and maintenance to renovation; based on the attenuation analysis of the time dimension, a quality evolution curve is formed to provide forward-looking guidance for later maintenance strategies (such as surface maintenance and defect repair), avoiding passive responses to quality issues.

[0017] 4. The automated inspection process reduces manual intervention and operational errors, while improving inspection efficiency and meeting the timeliness requirements of project acceptance. The built-in intelligent evaluation logic, combined with preset quality grade standards, automatically outputs clear quality conclusions and targeted optimization suggestions (such as formwork sealing process adjustment, vibration parameter optimization, etc.), directly guiding on-site construction improvements. It can adapt to the inspection requirements of different formwork solutions (such as acrylic panels and formwork paint), accurately identify the quality differences between different processes, and provide an objective basis for project selection. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of a quality inspection method for mirror-finished plain concrete proposed by the present invention; Figure 2This is a logical structural block diagram of a quality detection system for mirror-finished plain concrete proposed by the present invention. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0020] Reference Figure 1 A method for detecting the quality of mirror-surfaced plain concrete comprises the following steps: S100, 3D detection start: Determine the area to be detected, establish a 3D coordinate system with the lower left corner of the detection 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 modules, digital modeling tools and environmental sensors to complete the equipment's time and space reference calibration; Specifically, clarify the inspection scope: define the inspection area according to the project requirements (such as the entire surface of the frame column, specific area of ​​the wall), mark the area boundary with ink lines (error ≤ 5mm), and record the actual size of the area (length × height).

[0021] Pre-treatment: Before testing, remove dust, water stains and other interferences from the surface, mark obvious protrusions (such as concrete tumors) (circle them with a red marker) and analyze them separately later.

[0022] Origin setting: The lower left corner of the detection area (the intersection of the ground and the area boundary) is the coordinate origin (0,0,0).

[0023] Axis definition: X-axis: along the length of the detection area (horizontally to the right), in meters (m); Y-axis: along the height direction of the detection area (vertically upward), in meters (m); Z axis: perpendicular to the concrete surface (positive when pointing outward), unit is millimeter (mm, used to represent the surface normal depth, such as bubble depth and flatness deviation).

[0024] Coordinate marking: Coordinate reference points (using stainless steel markers) are set every 1m at the boundary of the area and calibrated using a total station (accuracy ±0.5mm) to ensure the orthogonality of the axis system (deviation ≤0.1°).

[0025] S200, spatial dimension multi-index collaborative detection, details are as follows: S201. Use a laser flatness meter (accuracy ±0.01mm) and a 60° gloss meter to synchronously collect surface data. Divide the inspection area into a 10cm×10cm grid, take 3 sampling points in each grid, and generate a global distribution heat map of flatness and gloss using the Kriging interpolation algorithm. Specifically, place a standard calibration block (known dimensions 100mm×100mm×5mm) in the center of the inspection area. Use the laser flatness meter and 3D scanner to measure separately. The deviation must be ≤0.02mm, otherwise adjust the equipment position. Use a hyperspectral camera to shoot a standard color card (X-Rite Color Checker), and use Image-proplus6.0 to calibrate the white balance and spectral baseline.

[0026] The actual operation is as follows: Meshing: Use a tape measure and ink line to mark a 10 cm × 10 cm grid line in the test area (error ≤ 1 mm), and mark each grid vertex with a white marker (diameter ≤ 5 mm).

[0027] Sampling point settings: Three sampling points were taken in each grid: one in the center and two at the quarter of the diagonal (coordinates were the lower left corner of the grid (x+2.5cm, y+2.5cm) and (x+7.5cm, y+7.5cm)).

[0028] Synchronous acquisition: Start the electric slide, and the laser flatness meter and gloss meter move along the X-axis (speed 5cm / s). They automatically stop for 0.5s every time they reach a sampling point to collect data. Record the three-dimensional coordinates (X, Y, Z), flatness value (Z direction deviation, mm), and gloss value (GU) of each point and save them in CSV format (fields: X, Y, Z, flatness, gloss, timestamp).

[0029] Heatmap generation: Import the data into ArcGIS or MATLAB and use the Kriging interpolation algorithm (select the Gaussian model as the variogram model and the search radius as 5 cm); Generates global heat maps of flatness (color scale: blue ≤ 1mm, red ≥ 3mm) and glossiness (color scale: red ≥ 40GU, ​​blue ≤ 20GU) with a resolution of 1mm×1mm.

[0030] S202: Use a hyperspectral camera (400-700 nm band, spectral resolution 1 nm) to collect surface reflectance spectra, and use Image-proplus 6.0 software to extract the reflectance characteristics of the red light band (630-670 nm) to calculate the color uniformity index; The specific operations are as follows: Hyperspectral image acquisition: Shoot in 50 cm × 50 cm sections, with 10 cm overlap between sections (to avoid stitching errors), and an exposure time of 50 ms (adjusted according to lighting conditions to ensure no overexposure). After capturing every three sections, calibrate the spectral baseline using a standard white plate (reflectivity 99%).

[0031] Reflectance extraction: Import the image into Image-proplus6.0, crop the detection area (eliminate the 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 the three sampling points (n≥300, covering the entire area).

[0032] The color uniformity index is calculated using the following formula: ,in, is the reflectivity of the red light band (630-670nm) of the i-th sampling point, The average reflectance of all sampling points, n is the total number of sampling points (n ≥ 300), and the U value range is 0-1. The closer to 1, the more uniform the color.

[0033] S203. Based on the principle of topological graph construction, bubbles with a diameter greater than 0.5 mm are used as nodes and the distance between bubble centers as edges to establish a bubble distribution topological graph, and calculate the node degree, average path length, and clustering coefficient. Specifically, this is achieved through the following operations: Bubble identification: 3D scanner point cloud data is imported into Geomagic Studio, and surface bubbles (diameter > 0.5mm) are identified using the "curvature mutation" algorithm. The center coordinates (X, Y, Z), diameter (d), and depth (z, Z-axis value) of each bubble are recorded. Manual review: Use a vernier caliper (accuracy 0.01mm) to randomly check 10% of the bubbles identified by the algorithm, and the deviation must be ≤0.1mm.

[0034] Topology construction: Node definition: Each bubble is a node, and its attributes include diameter, depth, and coordinates; Edge definition: When the distance between the centers of two bubbles is less than 5 cm, an edge is established (the weight is the distance value); Tools: Use Python's networkx library to build a topology graph and save it in .graphml format.

[0035] Topology parameter calculation: Node degree: the number of edges connected to a single bubble (reflecting the density of surrounding bubbles); Average path length: the average value of the shortest paths between all node pairs (reflecting the dispersion of bubble distribution); Clustering coefficient: the ratio of the actual number of edges between node neighbors to the possible number of edges (reflecting the degree of clustering).

[0036] Repeat node semantic inference: If there are duplicate bubble nodes identified at the same location (e.g., due to scanning errors), the semantic intersection of the preceding and succeeding nodes is used. For example: the preceding node "diameter 2-3mm, depth 0.5mm" + the succeeding node "diameter 2-3mm, depth 1mm" → the target semantics "diameter 2-3mm, depth 0.5-1mm".

[0037] S204. Use an ultrasonic flaw detector (detection depth 0-50 mm) to detect hidden bubbles within 5 mm below the surface and record the spatial correlation between the hidden bubbles and the surface bubbles. Specifically, ultrasonic flaw detection is used: probe selection: 5MHz straight probe (diameter 10mm), motor oil is used as coupling agent (to ensure there are no bubbles); detection range: 0-5mm below the surface (Z axis -0.1mm to -5mm), stop detection at 3 sampling points in each 10cm×10cm grid, and record the center coordinates (X, Y, Z) and diameter of the hidden bubble.

[0038] Spatial correlation analysis: Calculate the straight-line distance between the latent bubble and the nearest surface bubble. If it is ≤2cm, it is determined to be a "surface-latest bubble cluster" (there is a causal relationship). The formula for calculating the proportion of latent bubbles is: ; S300, dynamic attenuation analysis in time dimension: Repeat the spatial dimension test 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, calculate the attenuation coefficient and attenuation acceleration of each indicator, and establish a quadratic attenuation prediction model including time variables; The calculation formula of the attenuation coefficient k is: , the calculation formula of the attenuation acceleration a is: in, is the index value at time t, is the initial index value of 1 day old, 、 are the attenuation coefficients at time t1 and t2 respectively. A negative k indicates that the indicator is attenuating, and a negative a indicates that the attenuation speed is accelerating.

[0039] The quadratic decay prediction model form is: , 、 Fitting coefficient) Fitting method: Use the least squares method to fit 1-day, 7-day, and 28-day data, and verify it with 90-day data (deviation must be ≤5%); application: predict 180-day indicator values ​​to determine whether they meet long-term use requirements.

[0040] S400, Environmental Adaptability Calibration: Environmental sensors collect real-time data during testing, including temperature (0-50°C, accuracy ±0.5°C), humidity (20%-95%RH, accuracy ±5%RH), and light intensity (0-10,000 lux, accuracy ±100 lux). It should be noted that the temperature (T), humidity (H), and light intensity (L) during testing are simultaneously recorded, with the average value stored every 5 minutes to avoid instantaneous fluctuations. Gloss and color indicators are calibrated based on a preset environmental impact coefficient matrix. The environmental impact coefficient matrix is ​​a 3×3 matrix ,in is the temperature influence coefficient (-0.005 / ℃), is the humidity influence coefficient (-0.002 / %RH), is the illumination influence coefficient (0.001 / lux), gloss calibration value ,in is the measured glossiness, 、 、 are the deviation values ​​between the measured environment and the standard environment respectively.

[0041] The standard environment is: 25°C (temperature T), 50%RH (humidity H), 5000lux (light intensity L); The deviation value is: .

[0042] S500, Digital Dimension Topological Modeling: Maps spatial detection data into a 3D point cloud (0.1mm pitch). Based on the normal distribution transformation map construction method, a digital twin model with a 0.5mm resolution base layer and a 2mm resolution application layer is generated. The bubble node topology is optimized through repeated node semantic inference, enabling index mapping and rapid switching between models of different resolutions. Specifically, the 3D point cloud generation process is as follows: 3D scanner data stitching: Use Cyclone software to align multi-station scanning data (stitching error ≤ 0.3mm) and remove noise points (retain points with confidence ≥ 95%); Point cloud attribute mapping: associates smoothness, glossiness, reflectivity, bubble parameters, etc. to corresponding 3D coordinate points to generate a point cloud with attributes (.las format).

[0043] 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 indicator values ​​of each 0.5mm×0.5mm×0.1mm unit; the base layer is used for high-precision defect analysis (such as micro-bubble positioning).

[0044] The application layer aggregates the base layer data into 2mm×2mm×0.5mm voxel units through voxel fusion algorithm. The units store the following data: flatness average (average Z deviation of all points in the unit); gloss average (average GU after calibration); color uniformity (U value in the unit); bubble density (number of bubbles in the unit / unit volume); the application layer is used for lightweight display and engineering collaboration. Implement the associated call of two-layer models. / / is divisible, index value The unit is mm. 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 perpendicular to the surface of the detection area, respectively.

[0045] S600, comprehensive quality assessment: Based on the spatial distribution heat map, time decay curve, environmental calibration value and digital twin model, the analytic hierarchy process (AHP) is used to calculate the weight of each indicator (flatness 0.25, gloss 0.3, color 0.2, bubble parameter 0.25), and output quality grades (excellent, good, qualified, unqualified) and targeted optimization suggestions.

[0046] Defective areas are located using the digital twin model, and the causes are analyzed by combining topology maps and decay curves. For example, if bubbles cluster in the central area (clustering coefficient > 0.7) and a high proportion of hidden bubbles are present, it is recommended to optimize the sealing of the formwork joints (widening the sealing tape) and extend the vibration time by 5-10 seconds. If the gloss decays rapidly (k < -0.02 / day), it is recommended to increase the number of surface curing agent applications (from 2 to 3).

[0047] In addition, the quality grade judgment standards are as follows: high quality must meet the requirements of flatness ≤ 2mm, gloss ≥ 35GU, color uniformity index ≥ 0.9, maximum diameter of surface bubbles ≤ 3mm and the proportion of hidden bubbles < 5%; good must meet the requirements of flatness ≤ 3mm, gloss ≥ 25GU, color uniformity index ≥ 0.8, maximum diameter of surface bubbles ≤ 5mm and the proportion of hidden bubbles < 10%; qualified must meet the requirements of flatness ≤ 4mm, gloss ≥ 15GU, color uniformity index ≥ 0.7, maximum diameter of surface bubbles ≤ 8mm and the proportion of hidden bubbles < 15%; if the qualified standards are not met, it is unqualified.

[0048] Reference Figure 2 The present invention also proposes a mirror surface plain concrete quality detection system, comprising: 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 to the same mobile platform via a rigid bracket and synchronized with a unified data acquisition controller for data collection and multi-indicator data collection in spatial dimensions. 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 a temperature and humidity sensor, a light sensor, a timestamp unit, and a calibration algorithm unit. It is used to record time series and environmental parameters and perform real-time calibration of gloss and color indicators. 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 a built-in normal distribution transformation algorithm and topology optimization module to convert point cloud data into a multi-resolution digital twin model, supporting model compression and lightweight transmission. Comprehensive evaluation platform: This platform adopts a B / S architecture, is deployed in the central processing unit, and is connected to the digital modeling engine through a network interface, including but not limited to an RJ45 network port and a 4G / 5G network. The comprehensive evaluation platform includes an AHP weight calculation module, a quality grade determination module, and a visualization display module, supporting 3D visualization of test data and retrospective analysis of historical data. Data interface module: integrated into the comprehensive assessment platform, supports data interaction with the BIM platform and construction management system, and realizes real-time sharing of test results.

[0049] The 3D holographic inspection module has a sampling frequency of at least 10Hz and a scanning speed of at least 0.5m² / min, ensuring full data collection for a 10m² inspection area within two hours. The comprehensive evaluation platform's visualization module supports dynamic rendering of bubble topology maps, attenuation curve prediction and simulation, and digital twin model cross-section analysis. Touch controls enable precise positioning and detailed viewing of the inspection area.

[0050] Furthermore, the present invention selected six C40 mirrored exposed concrete frame columns (KZ1-KZ6, 1.6m x 1.6m, 5.505m high) from the concourse level of an intercity railway station. KZ1-KZ3 used a "steel formwork + 1.5mm acrylic sheet" formwork scheme, while KZ4-KZ6 used a "steel formwork + HD-6 formwork paint" formwork scheme. The following examples and comparative examples were designed: Example 1: Steel formwork + acrylic board (method of the present invention) Test object: KZ1 frame column (1.5mm acrylic sheet pasted inside the steel formwork, laser cutting and splicing).

[0051] Detection method: Spatial dimension: 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; Time dimension: Four time point tests: 1 day, 28 days, 90 days, and 180 days; Environmental calibration: When testing, the temperature is 25℃ and the humidity is 50%, according to the formula Calibrated gloss, where is the temperature influence coefficient, is the humidity influence coefficient, is the illumination influence coefficient, where is the measured glossiness, 、 、 are the deviation values ​​between the measured environment and the standard environment respectively; Digital modeling: Generate a 0.5mm resolution digital twin model and optimize the bubble topology. The results are shown in the following table: Table 1: Test results of various parameters of C40 mirrored plain concrete frame columns in Example 1

[0052] Example 2: Steel template + HD-6 template paint (method of the present invention) Inspection object: KZ4 frame column (the inside of the steel formwork is painted with HD-6 formwork paint, roller coating).

[0053] Detection method: Same as Example 1.

[0054] The test results are shown in the following table: Table 2: Test results of various parameters of C40 mirrored plain concrete frame columns in Example 2

[0055] The comparative example is the KZ2 frame column in the same area as Example 1 (the template scheme is the same). The detection method adopts traditional manual detection (straightedge + feeler gauge to measure flatness, single-point gloss meter), without environmental calibration, time tracking and hidden bubble detection.

[0056] Table 3: Test results of various parameters of comparative example C40 mirrored plain concrete frame column

[0057] Combining the above embodiments and comparative examples, it is not difficult to see that in terms of comprehensiveness of detection: Examples 1 and 2 detect 6 core indicators through the method of the present invention, covering explicit / hidden defects, time decay and environmental impact; while the comparative example only detects 2 indicators, missing key parameters such as color uniformity and hidden bubbles, and the evaluation is significantly one-sided.

[0058] In terms of data accuracy: In Example 1, environmental calibration controls the gloss error within ±1GU, and the flatness measurement deviation is ≤0.03mm; the comparison example has an error of 6.6%-25% due to no calibration and low sampling density, which cannot reflect the actual quality status.

[0059] In terms of time dimension effectiveness: Examples 1 and 2 established an attenuation model through 180-day tracking, predicting that the 180-day gloss deviation is ≤1%; the control example has no time tracking, and the long-term performance cannot be evaluated, which poses a later quality risk (such as exposure of hidden bubbles).

[0060] In terms of adaptability of different template solutions: Example 1 (acrylic board) is superior to Example 2 (HD-6 template paint) in flatness, glossiness, and bubble control, verifying that the present invention can accurately distinguish the quality differences between different construction solutions; the test data of Example 2 provides a basis for optimizing the template solution (such as increasing the thickness of the HD-6 template paint to reduce bubbles).

[0061] In terms of efficiency and cost: The method of the present invention (Example 1) takes 1.5 hours per column to detect, which saves 40% time compared with the traditional method (Comparative Example); the digital twin model can be called repeatedly to avoid repeated detection, and the long-term cost is reduced by more than 30%.

[0062] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A quality inspection method for mirror-finished plain concrete, characterized in that: The steps include: S100, 3D detection start: 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 modules, digital modeling tools, and environmental sensors to complete the equipment's time and space reference calibration; S200, spatial dimension multi-index collaborative detection, details are as follows: S201. Use a laser flatness meter and a 60° gloss meter to synchronously collect surface data. Divide the test area into 10 cm × 10 cm grids, take three sampling points in each grid, and generate a heat map of the global distribution of flatness and gloss using the Kriging interpolation algorithm. S202. Use a hyperspectral camera to collect the surface reflectance spectrum, use Image-proplus6.0 software to extract the red light band reflectance characteristics, and calculate the color uniformity index; S203. Based on the principle of topological graph construction, bubbles with a diameter greater than 0.5 mm are used as nodes and the distance between bubble centers as edges to establish a bubble distribution topological graph, and calculate the node degree, average path length, and clustering coefficient. S204. Detect hidden bubbles within 5 mm below the surface using an ultrasonic flaw detector, and record the spatial position correlation between the hidden bubbles and the surface bubbles; S300, dynamic attenuation analysis in time dimension: Repeat the spatial dimension test 1 day, 7 days, 28 days, 90 days, and 180 days after pouring, calculate the attenuation coefficient and attenuation acceleration of each indicator, and establish a quadratic attenuation prediction model that includes the time variable; S400, Environmental Adaptability Calibration: The environmental sensor collects the temperature, humidity and light intensity during the test in real time, and calibrates and corrects the gloss and color indexes based on the preset environmental impact coefficient matrix; S500, Digital Dimension Topological Modeling: Maps spatial detection data into a 3D point cloud. Based on the normal distribution transformation map construction method, a digital twin model with a 0.5mm resolution base layer and a 2mm resolution application layer is generated. The bubble node topology is optimized through repeated node semantic inference, enabling index mapping and rapid switching between models of different resolutions. S600, comprehensive quality assessment: Based on spatial distribution heat maps, time decay curves, environmental calibration values ​​and digital twin models, the hierarchical analysis method is used to calculate the weights of each indicator, and output quality grades and targeted optimization suggestions.

2. The quality inspection method of mirror-surface clear-faced concrete according to claim 1, characterized in that: The color uniformity index is calculated by the following formula: ,in, is the red light band reflectivity of the i-th sampling point, The average reflectance of all sampling points, n is the total number of sampling points, and the U value range is 0-1. The closer to 1, the more uniform the color.

3. The quality inspection method of mirror-surface clear-faced concrete according to claim 1, characterized in that: The calculation formula of the attenuation coefficient k is: , the calculation formula of the attenuation acceleration a is: in, is the index value at time t, is the initial index value of 1 day old, 、 are the attenuation coefficients at time t1 and t2 respectively. A negative k indicates that the indicator is attenuating, and a negative a indicates that the attenuation speed is accelerating.

4. The quality inspection method of mirror-surface clear-faced concrete according to claim 1, characterized in that: The environmental impact coefficient matrix is ​​a 3×3 matrix ,in is the temperature influence coefficient, is the humidity influence coefficient, is the lighting influence coefficient, gloss calibration value ,in is the measured glossiness, 、 、 are the deviation values ​​between the measured environment and the standard environment respectively.

5. The quality inspection method of mirror-surface plain 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 through the voxel fusion algorithm. Each voxel unit stores the flatness mean, gloss mean, color uniformity and bubble density parameters. Realize the associated call of the two-layer model, where x, y, and z are the coordinate values ​​of the voxel unit in the three-dimensional coordinate system, respectively, in millimeters, and 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 of mirror-surface plain concrete according to claim 1, characterized in that: In the bubble distribution topology diagram, the semantic inference of repeated nodes adopts the adjacent node semantic intersection method.

7. A mirror surface clear concrete quality detection system for implementing the method according to any one of claims 1 to 6, characterized in that: include: 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 to the same mobile platform via a rigid bracket and synchronized with a unified data acquisition controller for data collection and multi-indicator data collection in spatial dimensions. 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 a temperature and humidity sensor, a light sensor, a timestamp unit, and a calibration algorithm unit. It is used to record time series and environmental parameters and perform real-time calibration of gloss and color indicators. 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 a built-in normal distribution transformation algorithm and topology optimization module to convert point cloud data into a multi-resolution digital twin model, supporting model compression and lightweight transmission. Comprehensive evaluation platform: This platform adopts a B / S architecture, is deployed in the central processing unit, and is connected to the digital modeling engine through a network interface, including but not limited to an RJ45 network port and a 4G / 5G network. The comprehensive evaluation platform includes an AHP weight calculation module, a quality grade determination module, and a visualization display module, supporting 3D visualization of test data and retrospective analysis of historical data. Data interface module: integrated into the comprehensive assessment platform, supports data interaction with the BIM platform and construction management system, and realizes real-time sharing of test results.

8. The quality inspection system for mirror-finished plain concrete according to claim 7, characterized in that: 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.

9. The quality inspection system for mirror-finished plain concrete according to claim 7, characterized in that: The visualization display module of the comprehensive evaluation platform supports dynamic rendering of bubble topology maps, attenuation curve prediction simulation and digital twin model section analysis, and can achieve precise positioning and detailed viewing of the detection area through touch operation.

Images