Road construction paving quality detection method, system equipment and medium

Abstract

The invention discloses a road construction paving quality detection method, system equipment and a medium, and relates to the technical field of road construction quality detection.The method comprises the steps that laser point cloud data, infrared thermal imaging data and vibration frequency spectrum data of a road paving layer are collected through a multi-mode sensor set carried by a vehicle-mounted mobile platform, and the laser point cloud data, the infrared thermal imaging data and the vibration frequency spectrum data are obtained; performing data preprocessing to obtain a standardized engineering data set; calculating the pavement flatness, the transverse gradient variation coefficient and the tectonic depth variation of the to-be-detected area to obtain geometrical characteristic parameters; identifying a void area of the road pavement layer, and calculating a thermal anomaly parameter to obtain a thermodynamic parameter; establishing a vibration response feature library; determining a weight coefficient by adopting an analytic hierarchy process based on the geometrical characteristic parameters, the thermodynamic parameters and the vibration response characteristic library; in the target quality comprehensive evaluation model, determining a quality index of the to-be-detected area, and generating a pavement quality detection result in combination with a quality grading standard; according to the invention, the detection accuracy of road construction pavement quality detection is improved.

Description

Technical Field

[0001] This application relates to the field of road construction quality testing technology, and in particular to a method, system equipment and medium for testing the quality of road construction paving. Background Technology

[0002] Currently, in order to ensure the quality of the pavement layer during road construction, various methods such as visual inspection, manual measurement, and instrument testing are commonly used. Although these traditional methods guarantee the quality of the project to a certain extent, they cannot meet the high standards required by modern engineering due to problems such as low efficiency, low accuracy, and susceptibility to human factors.

[0003] With the development of sensor technology, detection methods based on automated equipment have been gradually introduced into road construction quality inspection. However, traditional road construction quality inspection methods rely on manual measurement or a single sensor (such as lidar or camera), which still have the drawback of insufficient accuracy in road construction paving quality inspection and need to be improved. Summary of the Invention

[0004] To improve the accuracy of road construction paving quality testing, this application provides a road construction paving quality testing method, system equipment, and medium.

[0005] Firstly, the objective of this invention is achieved through the following technical solution: A method for testing the quality of road construction paving includes: A multimodal sensor array mounted on a vehicle-mounted mobile platform is used to collect laser point cloud data, infrared thermal imaging data, and vibration spectrum data of the road pavement layer in the area to be inspected, and the data is preprocessed to obtain a standardized engineering dataset. Based on the standardized engineering dataset, the road surface smoothness, cross slope variation coefficient and structural depth variation of the area to be inspected are calculated to obtain geometric feature parameters. Based on preprocessed infrared thermal imaging data, void areas in the road pavement layer are identified, thermal anomaly parameters are calculated, and thermodynamic parameters are obtained; based on preprocessed vibration spectrum data, the natural frequency and damping ratio parameters of the pavement structure are extracted, and a vibration response feature library is established. Based on the geometric feature parameters, the thermodynamic parameters, and the vibration response feature library, the weight coefficients are determined by the analytic hierarchy process (AHP) and the preset comprehensive evaluation model for paving quality is optimized to obtain the target comprehensive evaluation model for paving quality. In the target quality comprehensive evaluation model, the quality index of the area to be inspected is determined, and the paving quality test results are generated by combining the preset quality grade classification standards.

[0006] By adopting the above technical solution, a method for detecting the quality of road construction pavement based on multimodal data fusion is provided. The area to be inspected is the pavement layer of the road construction to be inspected. Unlike the detection based on single sensor technology in the prior art, this application realizes full-dimensional detection of the geometric morphology (such as smoothness), internal defects (such as void areas), and structural stability (such as natural frequency) of the road pavement layer. At the same time, in practical applications, the vehicle-mounted mobile platform can also complete data preprocessing, feature extraction, and comprehensive evaluation during driving. Combined with dynamic evaluation capabilities, it is beneficial to improve the efficiency of quality detection. The target quality comprehensive evaluation model constructed by the Analytic Hierarchy Process (AHP) is combined with the multi-parameter weighted evaluation method to integrate the three major indicators of geometry, thermodynamics, and kinetics, output the quality index of the area to be inspected, and compare it with the preset quality level standard to generate a detection report containing a visualization model, heat map, and defect distribution map. This is conducive to realizing full-parameter, high-precision, and dynamic detection of road pavement quality.

[0007] In a preferred embodiment of this application, the calculation of road surface smoothness, cross slope variation coefficient, and texture depth variation specifically includes: Local features are extracted using the moving window analysis method, and the plane equation is fitted by the random sample consensus algorithm to calculate the road surface smoothness within the moving window. A digital elevation model is generated based on the lane centerline of the area to be inspected, and the road cross slope and coefficient of variation are calculated. The road construction depth is identified based on the concave-convex hull method, and the variability of the road construction depth is calculated using quantile statistics.

[0008] By adopting the above technical solutions, in order to improve the recognition accuracy of local features of road surface smoothness, this application uses the moving window analysis method combined with random sample consistent fitting of plane equations to extract and calculate the road surface smoothness within each moving window. At the same time, it combines a digital elevation model based on the lane centerline to enhance the accuracy of road cross slope calculation, which helps to discover potential road structure inhomogeneity problems. The application also uses the convex hull method to identify road construction depth and uses quantile statistics to calculate its variability in order to quantitatively assess the roughness of the road surface.

[0009] In a preferred embodiment of this application, the establishment of the vibration response feature library specifically includes: Fast Fourier Transform is performed on the vibration signals of the preprocessed vibration spectrum data to analyze the energy distribution in a fixed frequency band; the first-order natural frequency and damping ratio are calculated using a peak detection algorithm. A probabilistic neural network classifier is constructed, with the input parameters being the ratio of the first-order natural frequency and the damping ratio, and the spectral entropy, to generate a vibration response feature library; the training sample set of the vibration response feature library includes defect types such as base layer voids, asphalt aging, and / or aggregate loosening. The vibration response feature library has a dynamic update mechanism. When a new type of defect is identified, the weights of the vibration response feature library are adjusted through online incremental learning. The vibration response feature library is stored in a hierarchical structure, with graded storage based on the severity of defects, and supports real-time access from multiple terminals.

[0010] By adopting the above technical solutions, combined with Fast Fourier Transform (FFT) and peak detection algorithms, the first-order natural frequency and damping ratio can be efficiently identified, improving the ability to identify defects in road pavement layers. At the same time, based on the constructed probabilistic neural network classifier, which uses the ratio of its first-order natural frequency to damping ratio and spectral entropy as a basis, it can effectively distinguish various typical defect types such as base layer voids and asphalt aging, improving the accuracy of defect classification. Based on the dynamic update mechanism of the vibration response feature library, the weights can be adjusted incrementally online, effectively ensuring the timeliness and adaptability of the vibration response feature library.

[0011] In a preferred embodiment, this application describes the optimization of a pre-defined comprehensive evaluation model for paving quality by determining weighting coefficients using the analytic hierarchy process (AHP) based on the geometric feature parameters, the thermodynamic parameters, and the vibration response feature library. Specifically, this includes: A comprehensive evaluation factor is predefined, which includes, but is not limited to, road surface smoothness, cross slope variation coefficient, structural depth variation, thermal anomaly parameters, first-order natural frequency, and damping ratio. The area to be inspected is divided into several evaluation units, and each evaluation unit is assigned a value according to the comprehensive evaluation factor to obtain the preliminary evaluation score of each evaluation unit. The weight coefficients of each comprehensive evaluation factor are calculated using the analytic hierarchy process (AHP), and the quality index of each evaluation unit is calculated in combination with the preliminary evaluation score. Based on the quality index and the preset quality grade classification standard, the comprehensive evaluation factors and their weight coefficients in the preset comprehensive evaluation model of paving quality are dynamically adjusted to adapt to the testing needs of the current environmental conditions.

[0012] By adopting the above technical solution, a comprehensive evaluation of the quality of road pavement is achieved based on multiple predefined comprehensive evaluation factors and the weight coefficients of each comprehensive evaluation factor calculated using the analytic hierarchy process (AHP). To improve the reliability and scientific rigor of the road quality evaluation results, this application divides the area to be inspected into several evaluation units and assigns values ​​according to the comprehensive evaluation factors to obtain preliminary evaluation scores. Then, the AHP is applied to calculate the quality index. Based on the quality index and the preset quality grade classification standards, the evaluation factors and their weight coefficients in the comprehensive pavement quality evaluation model are dynamically adjusted to adapt to different environmental conditions and testing requirements.

[0013] In a preferred embodiment of this application, the steps for generating the preset comprehensive evaluation model for paving quality include: The factors affecting the quality of the pavement layer and their change thresholds were determined, resulting in several evaluation factors and corresponding reference ranges for pavement quality changes. Based on the preliminary design quality standards and the final design quality standards, the thickness of the base layer, and the predicted construction period and the designed construction period, a quality control table is generated to guide the adjustment of the paving process. The quality control table includes quality assessment indicators for different stages and specific repair suggestions for different quality problems. By combining the reference range for changes in paving quality and the quality control table, a comprehensive evaluation model for paving quality is constructed using machine learning algorithms.

[0014] By adopting the above technical solutions, the factors affecting the quality of the paving layer and their change thresholds can be accurately identified, which helps to discover potential problems in a timely manner during construction. The generated quality control table not only includes quality assessment indicators for different stages, but also provides specific repair suggestions for different quality problems, which enhances the guidance and feasibility in actual operation. At the same time, the constructed comprehensive paving quality evaluation model can be dynamically adjusted to cope with different construction conditions and environmental changes, making the model more adaptable.

[0015] In a preferred embodiment of this application, the method further includes: Based on the standardized engineering dataset, the pavement quality index QI is calculated using formula (1): QI=α×P j +β×T k +γ×V m Among them, P j T is the normalized value of the j-th geometric characteristic parameter (road surface smoothness, cross slope coefficient of variation, and texture depth variability); k V is the normalized value of the k-th thermodynamic parameter (thermal anomaly index, thermal conductivity); m is the normalized value of the m-th vibration response parameter (natural frequency, damping ratio); α, β, and γ are the combined weighting coefficients of the geometric characteristic parameter, thermodynamic parameter, and vibration response parameter, respectively, determined by the combined analytic hierarchy process and entropy weighting method.

[0016] By adopting the above technical solution, geometric characteristic parameters, thermodynamic parameters, and vibration response parameters are comprehensively considered, enabling quantitative evaluation of pavement quality from multiple dimensions. The multi-dimensional evaluation method is more comprehensive and objective, reducing the bias that may be caused by a single indicator. At the same time, by normalizing each parameter, it is ensured that different types of parameters are compared on the same scale, avoiding unfairness caused by different dimensions.

[0017] In a preferred example of this application: the combined weighting coefficients are optimized using formula (2): The initial weights calculated by the analytic hierarchy process, D(w) i ) represents the information entropy value. α ij To determine the importance scale of the i-th parameter relative to the j-th parameter in the matrix.

[0018] By adopting the above technical solution and combining the initial weights and information entropy values ​​calculated by the analytic hierarchy process to optimize the combined weight coefficients, the effective integration of subjective judgments (such as expert opinions) and objective data (such as information entropy values) is achieved, making the weight allocation more reasonable and scientific. This application further introduces the evaluation factor of information entropy values, which can measure the amount of information of each parameter, thereby adjusting its importance in the comprehensive evaluation.

[0019] Secondly, the objective of this invention is achieved through the following technical solution: A road construction paving quality inspection system, the system comprising: The vehicle-mounted mobile platform carries a multimodal sensor array to collect laser point cloud data, infrared thermal imaging data, and vibration spectrum data of the road pavement layer in the area to be inspected. The data processing module is used to preprocess the collected multimodal sensor data to generate a standardized engineering dataset; based on the standardized engineering dataset, it calculates the road surface smoothness, cross slope variation coefficient, and structural depth variation of the area to be inspected to obtain geometric feature parameters; based on the preprocessed infrared thermal imaging data, it identifies the void areas of the road pavement layer and calculates thermal anomaly parameters to obtain thermodynamic parameters. The vibration response feature library establishment module extracts the natural frequency and damping ratio parameters of the pavement structure based on preprocessed vibration spectrum data to establish a vibration response feature library; The model optimization module is used to optimize the preset pavement quality comprehensive evaluation model to obtain the target quality comprehensive evaluation model based on the geometric feature parameters, the thermodynamic parameters and the vibration response feature library, using the analytic hierarchy process to determine the weight coefficients; in the target quality comprehensive evaluation model, the quality index of the area to be inspected is determined, and the pavement quality inspection results are generated by combining the preset quality grade classification standards.

[0020] By adopting the above technical solutions, the vehicle-mounted mobile platform equipped with a multimodal sensor group can quickly and accurately collect laser point cloud data, infrared thermal imaging data and vibration spectrum data in the area to be inspected, thereby improving the efficiency of data acquisition. The model optimization module uses the analytic hierarchy process to determine the weight coefficients and optimizes the preset comprehensive evaluation model of paving quality to obtain the target quality comprehensive evaluation model. The optimization process considers a variety of influencing factors and their relative importance, making the final paving quality assessment results more scientific and reasonable.

[0021] Thirdly, the objective of this invention is achieved through the following technical solution: A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the aforementioned method for detecting the quality of road construction paving.

[0022] Fourthly, the objective of this invention is achieved through the following technical solution: A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the aforementioned method for detecting the quality of road construction paving.

[0023] In summary, this application includes at least one of the following beneficial technical effects: 1. It has achieved an integrated solution from data acquisition, processing to analysis and evaluation, which has significantly improved the accuracy and efficiency of road construction paving quality inspection; 2. By comprehensively considering geometric characteristic parameters, thermodynamic parameters, and vibration response parameters, a quantitative assessment of the pavement quality from multiple dimensions is achieved. This multi-dimensional assessment method is more comprehensive and objective, reducing the bias that may be caused by a single indicator. Attached Figure Description

[0024] Figure 1 This is a flowchart of a road construction paving quality inspection method according to one embodiment of this application; Figure 2 This is a schematic diagram of a device according to one embodiment of this application. Detailed Implementation

[0025] The present application will be further described in detail below with reference to the accompanying drawings.

[0026] In one embodiment, such as Figure 1 As shown, this application discloses a method for testing the quality of road construction paving, which specifically includes the following steps: S1: Using a multimodal sensor array mounted on a vehicle-mounted mobile platform, laser point cloud data, infrared thermal imaging data, and vibration spectrum data of the road pavement layer are collected in the area to be inspected, and the data is preprocessed to obtain a standardized engineering dataset.

[0027] In this embodiment, the multimodal sensor group refers to a sensing system that integrates a lidar (accuracy ±2mm), an infrared thermal imager (resolution 640×480 pixels, wavelength 8-14μm), and a three-dimensional accelerometer (range ±16g). During data acquisition, it is necessary to ensure that the vehicle's driving speed remains stable. Data preprocessing includes noise filtering, coordinate registration, and data normalization. The standardized engineering dataset establishes a unified geographic coordinate system (WGS84), and the data format conforms to the LAS1.2 standard.

[0028] Specifically, the multi-source heterogeneity processing is performed on the multi-field data collected by the multimodal sensor group: the laser point cloud data is downsampled using voxel rasterization (resolution ≤ 5cm); Infrared thermal imaging data undergoes radiometric calibration (atmospheric transmittance correction) and non-uniformity correction; Vibration spectrum data are filtered by bandpass (filter range: 1-100Hz) to eliminate environmental noise interference; (2) Spatial reference unification: Based on the ICP algorithm, coordinate registration between laser point cloud and infrared image is achieved (error < 3mm); (3) Data standardization: The registered multimodal data is normalized by Z-score to generate a standardized engineering dataset containing three-dimensional coordinates, temperature gradient and vibration amplitude.

[0029] S2: Based on a standardized engineering dataset, calculate the road surface smoothness, cross slope variation coefficient, and structural depth variation of the area to be inspected to obtain geometric feature parameters.

[0030] In this embodiment, step S2 includes: S21: The moving window analysis method is used to extract local features, and the plane equation is fitted by the random sample consensus algorithm to calculate the road surface smoothness within the moving window.

[0031] Specifically, in this embodiment, the moving window size is 2m×2m, and the step size is 0.5m; the random sample consensus algorithm is also known as the RANSAC algorithm.

[0032] S22: Generate a digital elevation model based on the lane centerline of the area to be inspected, and calculate the road cross slope and coefficient of variation.

[0033] Specifically, the cross slope of the road is calculated segment by segment along the center line of the lane.

[0034] S23: Identify road construction depth based on the concave-convex hull method, and use quantile statistics to calculate the variability of road construction depth.

[0035] Specifically, the size of the movable window can be adjusted according to actual needs to balance calculation accuracy and efficiency, and the construction depth refers to the depth of the road surface texture.

[0036] S3: Based on preprocessed infrared thermal imaging data, identify void areas in the road pavement layer, calculate thermal anomaly parameters, and obtain thermodynamic parameters; based on preprocessed vibration spectrum data, extract the natural frequency and damping ratio parameters of the road structure, and establish a vibration response feature library.

[0037] In this embodiment, the thermodynamic parameters include the thermal anomaly index and thermal conductivity parameters. The identification of voids in road pavement layers is typically based on temperature differences. Because voids (such as gaps under concrete layers) exhibit temperature anomalies due to the poor thermal conductivity of air, infrared thermal imagers can capture these temperature differences. In other words, the identification of voids is based on the principle of thermal conductivity anomalies: the thermal conductivity of voids (air gaps) (approximately 0.024 W / m·K) is much lower than that of concrete (approximately 1.7 W / m·K). Under sunlight or changes in ambient temperature, the voided area will form obvious abnormalities in temperature accumulation or dissipation; the FLIRT1020 thermal imager can be selected; in actual detection, different thresholds can be selected according to the actual application scenario. For example, the local threshold method is suitable for non-uniform heating scenarios (such as urban roads), while the global threshold method is suitable for highway detection under uniform environmental conditions.

[0038] Establishing a vibration response feature library includes: S31: Perform a fast Fourier transform on the vibration signal of the preprocessed vibration spectrum data to analyze the energy distribution of the fixed frequency band; calculate the first-order natural frequency and damping ratio through the peak detection algorithm.

[0039] Specifically, Fast Fourier Transform (FFT) analysis refers to converting time-domain vibration signals into the frequency domain to extract energy distribution characteristics of a specific frequency band (e.g., 1-50Hz); peak detection algorithm refers to identifying spectral peaks using the second derivative method and calculating the corresponding first-order natural frequency and damping ratio; energy distribution analysis refers to using a rectangular window (Hamming window) to integrate the spectrum in segments and obtain the energy proportion of each sub-band; for example, when configuring parameters, the FFT window length is 256 points (corresponding to 256ms at a sampling rate of 1kHz); the peak detection threshold is 0.05 × the maximum value; and the frequency band resolution is 0.05Hz.

[0040] S32: Construct a probabilistic neural network classifier with input parameters being the ratio of the first-order natural frequency and the damping ratio, and the spectral entropy, to generate a vibration response feature library; the training sample set of the vibration response feature library includes defect types such as base layer voids, asphalt aging, and / or aggregate loosening.

[0041] Specifically, the probabilistic neural network (PNN) classifier is a feedforward network based on radial basis function (RBF); the training sample set of the vibration response feature library should contain a sufficient number of learning samples, such as 1000 sets of labeled data: 300 cases of base layer voids, 400 cases of asphalt aging, and 300 cases of aggregate loosening; the sensitivity of feature parameters is ranked as follows: spectral entropy > relative frequency ratio.

[0042] S33: The vibration response feature library has a dynamic update mechanism. When a new type of defect is identified, the weights of the vibration response feature library are adjusted through online incremental learning.

[0043] In this embodiment, the vibration response feature library is dynamically updated to trigger a complete weight reset when the number of new defects appears > 5 times, and the model is automatically validated weekly (using the K-fold cross-validation method); online incremental learning uses a Bayesian weight update formula to dynamically adjust the classifier parameters.

[0044] S34: The vibration response feature library is stored in a hierarchical structure, with graded storage based on the severity of defects, and supports real-time access from multiple terminals.

[0045] In this embodiment, the hierarchical storage structure adopts a tree-shaped directory storage, which is layered according to defect severity → detection time → geographical location. To facilitate the analysis of different defect types, histograms of spectral entropy distribution for different defect types and performance comparison curves of the model before and after incremental learning can also be generated. Data synchronization between the Web terminal, mobile terminal, and vehicle terminal is realized based on RESTful API.

[0046] Specifically, the API interface is designed as follows: data upload interface is POST / API; query interface is GET / api; and real-time synchronization interface is WebSocket.

[0047] S4: Based on the geometric characteristic parameters, thermodynamic parameters, and vibration response characteristic library, the weight coefficients are determined by the analytic hierarchy process (AHP) and the preset comprehensive evaluation model for paving quality is optimized to obtain the comprehensive evaluation model for target quality.

[0048] In this embodiment, the weight parameter optimization includes: Establish the parameter weight matrix based on the Analytic Hierarchy Process (AHP): A judgment matrix is ​​constructed, and the weights of geometric / thermodynamic / vibration parameters are determined through a consistency test (CR < 0.1). Then, the entropy weight method is introduced to correct the subjective weights, and the combined weight coefficients are calculated. Specifically, based on the geometric feature parameters, thermodynamic parameters, and vibration response feature library, the weight coefficients are determined using the analytic hierarchy process (AHP) to optimize the preset comprehensive evaluation model for pavement quality. This optimization includes: S41: Predefine comprehensive evaluation factors, including but not limited to road surface smoothness, cross slope variation coefficient, structural depth variation, thermal anomaly parameters, first-order natural frequency, and damping ratio.

[0049] In this embodiment, an evaluation factor library is first preset, that is, an evaluation factor database containing historical engineering data is established, which supports dynamic addition and deletion of comprehensive evaluation factors.

[0050] S42: Divide the area to be inspected into several evaluation units. Assign a value to each evaluation unit according to the comprehensive evaluation factor to obtain the preliminary evaluation score of each evaluation unit.

[0051] In this embodiment, several evaluation units are divided based on spatial location (e.g., a 2m×2m grid) or functional area (e.g., lane lines, bridge joints); that is, a hybrid partitioning strategy is adopted, combining spatial geometric features and functional attributes for dynamic partitioning, and the Voronoi diagram algorithm is used to generate the initial grid, while the grid size supports dynamic adjustment. Urban roads: 2m×2m (high-density detection); Highway: 5m x 5m (balancing efficiency and accuracy); Bridges / tunnels: 1m x 1m (structurally sensitive area).

[0052] When adjusting functional areas, segmentation is based on lane lines, that is, lane boundaries are identified based on image semantic segmentation, and the grid boundaries are forced to align with the lane lines; at the same time, special area isolation is supported, with each side of the bridge expansion joint extended by 1m to form a separate area, and the drainage grate area is merged into an independent unit. For the edge parts, morphological closing operations are used to fill the grid holes.

[0053] Specifically, based on fuzzy membership functions and weighted scoring models, multi-source heterogeneous data are first mapped to a unified score, i.e., parameter standardization. Geometric feature parameters are standardized using Min-Max; thermodynamic parameters are standardized using Gaussian normalization; and vibration parameters are standardized using quantiles to eliminate the influence of outliers. Example weight configuration (verified by AHP): The parameters and their corresponding weights are as follows: flatness -0.3, cross slope coefficient of variation -0.15, thermal anomaly parameter -0.25, first natural frequency -0.2, damping ratio -0.1.

[0054] S43: Calculate the weight coefficients of each comprehensive evaluation factor using the analytic hierarchy process (AHP), and combine them with the preliminary evaluation scores to calculate the quality index of each evaluation unit.

[0055] Specifically, the analytic hierarchy process (AHP) refers to constructing a judgment matrix through expert scoring and calculating the weights of each comprehensive evaluation factor.

[0056] S44: Based on the quality index and the preset quality grade classification standards, dynamically adjust the comprehensive evaluation factors and their weight coefficients in the preset comprehensive evaluation model of paving quality to adapt to the testing needs of the current environmental conditions.

[0057] Example dynamic weighting adjustment: When the temperature is high (>35℃), the weight of thermodynamic parameters is increased to 0.35, and the normal temperature is set between 26-30℃; when the speed is high (>60km / h), the weight of vibration parameters is reduced to 0.05.

[0058] In one embodiment, the steps for generating the preset comprehensive evaluation model for paving quality include: S401: Determine the influencing factors and change thresholds that affect the pavement quality, and obtain several evaluation factors and corresponding reference ranges for pavement quality changes.

[0059] Specifically, influencing factors include material characteristics, construction parameters, and environmental factors. Material characteristics include concrete slump, aggregate gradation, and asphalt viscosity; construction parameters include paving temperature, number of compaction passes, and vibration frequency; and environmental factors include temperature gradient, humidity variation, and rainfall. The threshold for variation is determined by combining the requirements of JTGF40-2004 "Technical Specification for Construction of Asphalt Pavement of Highway" and statistical analysis of historical engineering data (such as the mean ± 3σ principle).

[0060] S402: Based on the preliminary design quality standards and the final design quality standards, the thickness of the base layer, and the predicted construction period and the designed construction period, a quality control table is generated to guide the adjustment of the paving process. The quality control table includes quality assessment indicators for different stages and specific repair suggestions for different quality problems.

[0061] Specifically, the preliminary design standards are related to the thickness of the pavement structure layer and the material mix ratio; the final design quality standards are related to acceptance indicators such as skid resistance coefficient and permeability coefficient; different construction stages include the paving stage and the curing stage, and the quality assessment indicators are: compaction degree ≥98% and smoothness ≤2mm / m in the paving stage; and temperature gradient <5℃ / m and moisture content ≤3% in the curing stage; the repair suggestions are: compaction and partial milling and repaving in the paving stage; and covering with insulation film and water curing in the curing stage.

[0062] S403: Combining the reference range for paving quality changes and the quality control table, a comprehensive evaluation model for paving quality is constructed using machine learning algorithms.

[0063] S5: In the target quality comprehensive evaluation model, determine the quality index of the area to be inspected, and generate the paving quality inspection results by combining the preset quality grade classification standards.

[0064] In this embodiment, the paving quality inspection results include visual charts and reports; based on preset quality grade classification standards (e.g., Excellent > 0.85, Good 0.7-0.85, Qualified 0.5-0.7, Unqualified < 0.5); combined with spatiotemporal correlation analysis, an inspection report containing quality index, defect location, and maintenance suggestions is generated.

[0065] In one embodiment, a method for inspecting the quality of road construction paving further includes: Based on the standardized engineering dataset, the pavement quality index QI is calculated using formula (1): QI=α×P j +β×T k +γ×V m Among them, P j T is the normalized value of the j-th geometric characteristic parameter (road surface smoothness, cross slope coefficient of variation, and texture depth variability); k V is the normalized value of the k-th thermodynamic parameter (thermal anomaly index, thermal conductivity); m is the normalized value of the m-th vibration response parameter (natural frequency, damping ratio); α, β, and γ are the combined weighting coefficients of the geometric characteristic parameter, thermodynamic parameter, and vibration response parameter, respectively, determined by the combined analytic hierarchy process and entropy weighting method.

[0066] The combined weight coefficients are optimized using formula (2): The initial weights calculated by the analytic hierarchy process, D(w) i ) represents the information entropy value. α ij To determine the importance scale of the i-th parameter relative to the j-th parameter in the matrix.

[0067] Furthermore, the weight range for geometric parameters is 0.4-0.6; the weight range for thermodynamic parameters is 0.2-0.3; and the weight range for vibration parameters is 0.2-0.3. The weight range can be adaptively adjusted under different environmental conditions.

[0068] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0069] In one embodiment, a road construction paving quality inspection system is provided, which corresponds to a road construction paving quality inspection method described in the above embodiment.

[0070] A road construction paving quality inspection system includes a vehicle-mounted mobile platform, a data processing module, a vibration response feature library establishment module, and a model optimization module. Detailed descriptions of each functional module are as follows: The vehicle-mounted mobile platform carries a multimodal sensor array to collect laser point cloud data, infrared thermal imaging data, and vibration spectrum data of the road pavement layer in the area to be inspected. The data processing module is used to preprocess the collected multimodal sensor data to generate a standardized engineering dataset; based on the standardized engineering dataset, it calculates the road surface smoothness, cross slope variation coefficient, and structural depth variation of the area to be inspected to obtain geometric feature parameters; based on the preprocessed infrared thermal imaging data, it identifies the void areas of the road pavement layer and calculates thermal anomaly parameters to obtain thermodynamic parameters. The vibration response feature library establishment module extracts the natural frequency and damping ratio parameters of the pavement structure based on preprocessed vibration spectrum data to establish a vibration response feature library; The model optimization module is used to optimize the preset comprehensive evaluation model of paving quality based on geometric feature parameters, thermodynamic parameters and vibration response feature library, and to determine the weight coefficients using the analytic hierarchy process (AHP) to obtain the target comprehensive evaluation model of paving quality. In the target comprehensive evaluation model of paving quality, the quality index of the area to be inspected is determined, and the paving quality inspection results are generated in combination with the preset quality grade classification standards.

[0071] For specific limitations regarding a road construction paving quality inspection system, please refer to the limitations of a road construction paving quality inspection method mentioned above, which will not be repeated here. Each module in the aforementioned road construction paving quality inspection system can be implemented entirely or partially through software, hardware, or a combination thereof. Each module can be embedded in or independent of the processor in a computer device in hardware form, or it can be stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0072] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 2As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores laser point cloud data, infrared thermal imaging data, and vibration spectrum data, etc. The network interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for detecting the quality of road construction paving.

[0073] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps: S1: Using a multimodal sensor array mounted on a vehicle-mounted mobile platform, laser point cloud data, infrared thermal imaging data, and vibration spectrum data of the road pavement layer are collected in the area to be inspected, and the data is preprocessed to obtain a standardized engineering dataset.

[0074] S2: Based on a standardized engineering dataset, calculate the road surface smoothness, cross slope variation coefficient, and structural depth variation of the area to be inspected to obtain geometric feature parameters.

[0075] S3: Based on preprocessed infrared thermal imaging data, identify void areas in the road pavement layer, calculate thermal anomaly parameters, and obtain thermodynamic parameters; based on preprocessed vibration spectrum data, extract the natural frequency and damping ratio parameters of the road structure, and establish a vibration response feature library.

[0076] S4: Based on the geometric characteristic parameters, thermodynamic parameters, and vibration response characteristic library, the weight coefficients are determined by the analytic hierarchy process (AHP) and the preset comprehensive evaluation model for paving quality is optimized to obtain the comprehensive evaluation model for target quality.

[0077] S5: In the target quality comprehensive evaluation model, determine the quality index of the area to be inspected, and generate the paving quality inspection results by combining the preset quality grade classification standards.

[0078] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor: S1: Using a multimodal sensor array mounted on a vehicle-mounted mobile platform, laser point cloud data, infrared thermal imaging data, and vibration spectrum data of the road pavement layer are collected in the area to be inspected, and the data is preprocessed to obtain a standardized engineering dataset.

[0079] S2: Based on a standardized engineering dataset, calculate the road surface smoothness, cross slope variation coefficient, and structural depth variation of the area to be inspected to obtain geometric feature parameters.

[0080] S3: Based on preprocessed infrared thermal imaging data, identify void areas in the road pavement layer, calculate thermal anomaly parameters, and obtain thermodynamic parameters; based on preprocessed vibration spectrum data, extract the natural frequency and damping ratio parameters of the road structure, and establish a vibration response feature library.

[0081] S4: Based on the geometric characteristic parameters, thermodynamic parameters, and vibration response characteristic library, the weight coefficients are determined by the analytic hierarchy process (AHP) and the preset comprehensive evaluation model for paving quality is optimized to obtain the comprehensive evaluation model for target quality.

[0082] S5: In the target quality comprehensive evaluation model, determine the quality index of the area to be inspected, and generate the paving quality test results by combining the preset quality grade classification standards.

[0083] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0084] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0085] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for inspecting the quality of road construction paving, characterized in that, include: A multimodal sensor array mounted on a vehicle-mounted mobile platform is used to collect laser point cloud data, infrared thermal imaging data, and vibration spectrum data of the road pavement layer in the area to be inspected, and the data is preprocessed to obtain a standardized engineering dataset. Based on the standardized engineering dataset, the road surface smoothness, cross slope variation coefficient and structural depth variation of the area to be inspected are calculated to obtain geometric feature parameters. Based on preprocessed infrared thermal imaging data, void areas in the road pavement layer are identified, thermal anomaly parameters are calculated, and thermodynamic parameters are obtained. Based on the preprocessed vibration spectrum data, the natural frequency and damping ratio parameters of the pavement structure are extracted, and a vibration response feature library is established. Based on the geometric feature parameters, the thermodynamic parameters, and the vibration response feature library, the weight coefficients are determined by the analytic hierarchy process (AHP) and the preset comprehensive evaluation model for paving quality is optimized to obtain the target comprehensive evaluation model for paving quality. In the target quality comprehensive evaluation model, the quality index of the area to be inspected is determined, and the paving quality test results are generated by combining the preset quality grade classification standards.

2. The method for testing the quality of road construction paving according to claim 1, characterized in that, The calculation of road surface smoothness, cross slope variation coefficient, and texture depth variation specifically includes: Local features are extracted using the moving window analysis method, and the plane equation is fitted by the random sample consensus algorithm to calculate the road surface smoothness within the moving window. A digital elevation model is generated based on the lane centerline of the area to be inspected, and the road cross slope and coefficient of variation are calculated. The road construction depth is identified based on the concave-convex hull method, and the variability of the road construction depth is calculated using quantile statistics.

3. The method for testing the quality of road construction paving according to claim 1, characterized in that, The establishment of the vibration response feature library specifically includes: Fast Fourier Transform is performed on the vibration signals of the preprocessed vibration spectrum data to analyze the energy distribution in a fixed frequency band; the first-order natural frequency and damping ratio are calculated using a peak detection algorithm. A probabilistic neural network classifier is constructed, with the input parameters being the ratio of the first-order natural frequency and the damping ratio, and the spectral entropy, to generate a vibration response feature library; the training sample set of the vibration response feature library includes defect types such as base layer voids, asphalt aging, and / or aggregate loosening. The vibration response feature library has a dynamic update mechanism. When a new type of defect is identified, the weights of the vibration response feature library are adjusted through online incremental learning. The vibration response feature library is stored in a hierarchical structure, with graded storage based on the severity of defects, and supports real-time access from multiple terminals.

4. The method for testing the quality of road construction paving according to claim 1, characterized in that, The optimization of the pre-set comprehensive evaluation model for paving quality, based on the geometric feature parameters, the thermodynamic parameters, and the vibration response feature library, uses the analytic hierarchy process (AHP) to determine weighting coefficients. Specifically, this includes: A comprehensive evaluation factor is predefined, which includes, but is not limited to, road surface smoothness, cross slope variation coefficient, structural depth variation, thermal anomaly parameters, first-order natural frequency, and damping ratio. The area to be inspected is divided into several evaluation units, and each evaluation unit is assigned a value according to the comprehensive evaluation factor to obtain the preliminary evaluation score of each evaluation unit. The weight coefficients of each comprehensive evaluation factor are calculated using the analytic hierarchy process (AHP), and the quality index of each evaluation unit is calculated in combination with the preliminary evaluation score. Based on the quality index and the preset quality grade classification standard, the comprehensive evaluation factors and their weight coefficients in the preset comprehensive evaluation model of paving quality are dynamically adjusted to adapt to the testing needs of the current environmental conditions.

5. The method for testing the quality of road construction paving according to claim 1, characterized in that, The steps for generating the preset comprehensive evaluation model for paving quality include: The factors affecting the quality of the pavement layer and their change thresholds were determined, resulting in several evaluation factors and corresponding reference ranges for pavement quality changes. Based on the preliminary design quality standards and the final design quality standards, the thickness of the base layer, and the predicted construction period and the designed construction period, a quality control table is generated to guide the adjustment of the paving process. The quality control table includes quality assessment indicators for different stages and specific repair suggestions for different quality problems. By combining the reference range for changes in paving quality and the quality control table, a comprehensive evaluation model for paving quality is constructed using machine learning algorithms.

6. The method for testing the quality of road construction paving according to claim 1, characterized in that, The method further includes: calculating the pavement quality index QI using formula (1) based on a standardized engineering dataset: QI=α×P j +β×T k +γ×V m Among them, P j T is the normalized value of the j-th geometric characteristic parameter (road surface smoothness, cross slope coefficient of variation, and texture depth variability); k V is the normalized value of the k-th thermodynamic parameter (thermal anomaly index, thermal conductivity); m is the normalized value of the m-th vibration response parameter (natural frequency, damping ratio); α, β, and γ are the combined weighting coefficients of the geometric characteristic parameter, thermodynamic parameter, and vibration response parameter, respectively, determined by the combined analytic hierarchy process and entropy weighting method.

7. The method for testing the quality of road construction paving according to claim 6, characterized in that, The combined weight coefficients are optimized using formula (2): The initial weights calculated by the analytic hierarchy process, D(w) i ) represents the information entropy value. α ij To determine the importance scale of the i-th parameter relative to the j-th parameter in the matrix.

8. A road construction paving quality inspection system, characterized in that, The system includes: The vehicle-mounted mobile platform carries a multimodal sensor array to collect laser point cloud data, infrared thermal imaging data, and vibration spectrum data of the road pavement layer in the area to be inspected. The data processing module is used to preprocess the collected multimodal sensor data to generate a standardized engineering dataset; based on the standardized engineering dataset, it calculates the road surface smoothness, cross slope variation coefficient, and structural depth variation of the area to be inspected to obtain geometric feature parameters; based on the preprocessed infrared thermal imaging data, it identifies the void areas of the road pavement layer and calculates thermal anomaly parameters to obtain thermodynamic parameters. The vibration response feature library establishment module extracts the natural frequency and damping ratio parameters of the pavement structure based on preprocessed vibration spectrum data to establish a vibration response feature library; The model optimization module is used to optimize the preset pavement quality comprehensive evaluation model to obtain the target quality comprehensive evaluation model based on the geometric feature parameters, the thermodynamic parameters and the vibration response feature library, using the analytic hierarchy process to determine the weight coefficients; in the target quality comprehensive evaluation model, the quality index of the area to be inspected is determined, and the pavement quality inspection results are generated by combining the preset quality grade classification standards.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the road construction paving quality inspection method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the road construction paving quality inspection method as described in any one of claims 1 to 7.

Images