Method for detecting and evaluating quality of shotcrete construction site

Abstract

This invention belongs to the field of tunnel support testing technology. It discloses a method for testing and evaluating the on-site construction quality of shotcrete. By establishing a gridded testing network to realize the spatial correspondence between thickness and strength, and employing collaborative testing to obtain parameter distribution within a unified spatial framework, it innovatively introduces self-similarity index and multiple self-similarity spectrum analysis to achieve a quantitative description of parameter distribution patterns. A simplified mechanical model is established to transform material parameters into safety indicators, intuitively identifying potential instability risk points. A comprehensive quality evaluation index system is constructed to achieve scientific grading and evaluation. A closed-loop feedback mechanism of testing-evaluation-optimization is formed to continuously improve construction quality. This invention expands quality evaluation from simple parameter testing to safety performance assessment, providing a more accurate and comprehensive quality control solution for tunnel engineering, effectively improving engineering safety and durability.

Description

Technical Field

[0001] This invention relates to the field of tunnel support testing technology, and more specifically, to a method for testing and evaluating the on-site construction quality of shotcrete. Background Technology

[0002] In tunnel construction practice, the quality inspection and evaluation of shotcrete for initial support faces numerous technical challenges and methodological limitations. Traditional inspection systems generally suffer from arbitrary inspection point placement and a lack of spatial correspondence. Thickness and strength testing are often conducted independently by different teams, resulting in a disconnect and making it impossible to establish spatial mapping relationships between parameters, which is particularly evident in sections with complex and varying surrounding rock. Existing evaluation methods rely excessively on macroscopic statistical indicators such as average values ​​and pass rates, lacking in-depth analysis of the spatial distribution characteristics of parameters. This makes it difficult to accurately characterize the degree and pattern of unevenness in the shotcrete layer, causing local weak areas to be masked by overall data. For example, the systematic insufficient thickness of the transition zone between the arch crown and arch foot is difficult to detect in a timely manner. More importantly, current inspection results only remain at the level of material parameters, disconnected from the structural stress state. There is a lack of an effective mechanism to transform inspection data into safety evaluation indicators, making it impossible to quantitatively assess the actual bearing capacity and potential instability risk of the support system. This problem is particularly prominent under conditions of weak surrounding rock and high ground stress. Furthermore, the existing testing and evaluation system lacks a systematic connection with construction process optimization. Testing results cannot effectively guide subsequent adjustments to construction parameters, creating data silos. This leads to the recurrence of similar quality problems in different sections, such as the repeated occurrence of thickness control issues under similar geological conditions within the same project. These technical bottlenecks severely restrict the accurate assessment and continuous improvement of tunnel shotcrete support quality, increasing engineering safety hazards and subsequent maintenance costs.

[0003] In view of this, the present invention proposes a method for testing and evaluating the on-site construction quality of shotcrete to solve the above problems. Summary of the Invention

[0004] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: a method for detecting and evaluating the quality of on-site sprayed concrete construction, comprising:

[0005] A gridded detection network is established within the tunnel section to be inspected. Multiple detection sections are set at fixed intervals along the tunnel direction. Multiple detection positions are set at fixed angles along the circumferential direction within each detection section, forming a detection point array corresponding to spatial positions.

[0006] At each detection location of the detection point array, the thickness and strength of the spray layer are collected simultaneously to generate a set of thickness and strength detection values ​​for each detection section.

[0007] Based on the set of thickness detection values ​​and the set of strength detection values, the average detection value and the degree of cross-sectional offset value of each detection section are calculated respectively.

[0008] The thickness and strength test values ​​are normalized to construct a continuous distribution curve, and the thickness self-similarity index and strength self-similarity index of each test section are calculated based on the grid coverage analysis method.

[0009] Based on the thickness self-similarity index and the strength self-similarity index, multi-scale self-similarity features are extracted for each test section, multiple self-similarity spectrum parameters are calculated, and a correlation model between the multiple self-similarity spectrum parameters and construction process parameters is established.

[0010] The regional design offset value is calculated based on the design value of each detection section, and the difference rate index of each detection section is established by combining the average detection value of the section. Then, the regional average difference rate, regional average deviation rate and regional average self-similarity index are statistically analyzed.

[0011] A simplified mechanical model is established based on the average detection value of the cross section as an equivalent parameter. The stress at key locations and the cross section safety factor of each detection cross section are calculated, the instability risk points and stress concentration points are identified, and the comprehensive safety factor of the detection section is calculated.

[0012] Based on the regional average difference rate, regional average deviation rate, regional average self-similarity index and comprehensive safety factor, a comprehensive evaluation index for regional construction quality is constructed, and the on-site construction quality level of shotcrete is determined according to the preset allowable value.

[0013] The technical effects and advantages of the method for detecting and evaluating the on-site construction quality of shotcrete according to the present invention are as follows:

[0014] This invention establishes a complete testing and evaluation system, breaking down the information barrier between thickness and strength parameters. This makes construction quality assessment more comprehensive and multi-dimensional, especially in tunnel engineering with complex geological conditions, enabling the accurate identification of local quality hazards that are easily overlooked by traditional methods. This invention organically combines material testing results with structural safety, moving quality evaluation beyond simple parameter compliance checks to a substantive assessment of the structure's working state and safety margins, providing more forward-looking risk warnings for engineering decisions. This invention achieves closed-loop management of quality evaluation and construction optimization. Through data-driven continuous optimization of construction process parameters, it effectively avoids the recurrence of quality problems and significantly improves construction efficiency and resource utilization. In the long run, this systematic quality control mechanism significantly reduces the life-cycle cost of the project, decreases the frequency of maintenance interventions during tunnel operation, and improves the reliability and durability of infrastructure. Attached Figure Description

[0015] Figure 1This is a schematic diagram of a method for detecting and evaluating the on-site construction quality of shotcrete according to the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] This application provides a method for detecting and evaluating the on-site construction quality of shotcrete. The method can be implemented by various entities, including but not limited to: tunnel engineering quality management systems, on-site testing equipment, and engineering monitoring platforms, which can be considered as general computing nodes in this application. The detection and evaluation system includes, but is not limited to: at least one cloud-based data analysis engine, a distributed on-site testing network, and an intelligent quality evaluator.

[0018] Please see Figure 1 In this embodiment of the invention, a method for detecting and evaluating the on-site construction quality of shotcrete includes:

[0019] A gridded detection network is established within the tunnel section to be inspected. Multiple inspection sections are set at fixed intervals along the tunnel direction. Within each inspection section, multiple inspection positions are set at fixed angles along the circumferential direction, forming a spatially corresponding inspection point matrix. This step establishes a one-to-one spatially corresponding gridded detection network, ensuring a unified spatial reference system for thickness and strength testing. The gridded detection network consists of multiple inspection sections along the tunnel direction and inspection positions evenly distributed at circumferential angles on each section, forming a systematic three-dimensional inspection point matrix. The spacing between inspection sections is determined based on the tunnel length and the frequency of changes in construction quality, typically 10-20 meters; the angular interval of the inspection positions is determined based on the section shape and spraying process characteristics, typically 30° or 45°. This gridded point layout strategy avoids the limitations of traditional random point selection methods, achieving full coverage and systematic parameter acquisition of shotcrete, providing a spatially corresponding data foundation for subsequent multidimensional analysis.

[0020] At each detection location of the detection array, the thickness and strength of the sprayed layer are simultaneously acquired, generating sets of thickness and strength values ​​for each detection section. This step achieves coordinated detection of two key parameters, thickness and strength, ensuring spatial correspondence and temporal consistency of the data. Thickness detection primarily employs ground-penetrating radar (GPR), determining the sprayed layer thickness by analyzing the reflection characteristics of electromagnetic waves at different media interfaces. Strength detection utilizes an ultrasonic echo synthesis method, combining ultrasonic wave propagation characteristics and surface rebound hardness to estimate the actual strength of the concrete. Both detection methods are non-destructive or micro-destructive testing techniques, capable of efficiently acquiring a large number of data points and forming a complete spatial distribution image of the parameters. The synchronous detection strategy establishes a one-to-one correspondence between thickness and strength data at the same location, revealing the potential correlation between the two types of parameters and laying the foundation for comprehensive quality evaluation.

[0021] Based on the sets of thickness and strength test values, the average test value and the degree of deviation of each test section are calculated. This step performs preliminary statistical processing on the raw test data, extracting basic indicators reflecting the overall level and internal dispersion of the section. The average test value of the section is calculated by arithmetic mean, reflecting the overall level of the spray layer thickness or strength of the section; the degree of deviation of the section is quantified by calculating the standard deviation of the test values, thus quantifying the uniformity of parameter distribution. These two indicators together constitute the basic evaluation framework for section quality, examining both the overall compliance and the uniformity of distribution, and can effectively identify hidden defects of "overall compliance, local instability".

[0022] The thickness and intensity measurement sets are normalized to construct continuous distribution curves, and the thickness and intensity self-similarity indices for each measurement section are calculated based on the grid coverage analysis method. This step is an innovative method for quantitatively describing the morphological characteristics of parameter distribution, revealing the complex geometric properties of the spray layer parameter distribution through self-similarity analysis. Normalization maps parameters of different dimensions to a unified interval, facilitating comparative analysis; the continuous distribution curve transforms discrete measurement points into continuous functions, more comprehensively reflecting the parameter variation trend. The grid coverage analysis method, drawing on fractal geometry principles, calculates the self-similarity index through the coverage characteristics of the distribution curves by a multi-scale grid. This index intuitively reflects the "irregularity" of the parameter distribution, providing deep geometric characteristics that traditional statistical indicators cannot provide.

[0023] Based on thickness and strength self-similarity indices, multi-scale self-similarity features were extracted from each inspection section, multi-self-similarity spectrum parameters were calculated, and a correlation model between these parameters and construction process parameters was established. This step further deepened the multi-scale analysis of distribution morphology, expanding from a single self-similarity index to a complete multi-self-similarity spectrum, comprehensively characterizing the scale variation law of parameter distribution. The multi-self-similarity spectrum, by analyzing the scale behavior of different moments, revealed the local singularities and global regularities of the distribution, particularly by quantitatively distinguishing different types of non-uniform distributions through spectral width and spectral asymmetry indices. The correlation model then established a mapping relationship between these geometric features and spraying process parameters (such as wind pressure, distance, and accelerator dosage), providing a data-driven theoretical basis for process optimization.

[0024] Based on the design values ​​of each inspection section, the regional design offset value is calculated, and combined with the average inspection value of the section, a difference rate index for each inspection section is established. Then, the regional average difference rate, regional average deviation rate, and regional average self-similarity index are statistically analyzed. This step extends single-section analysis to the entire inspection section, establishing a global quality evaluation index system. The regional design offset value reflects the inherent variability of design requirements; the section difference rate index quantifies the degree of deviation between actual inspection values ​​and design values; and the regional average index integrates information from multiple sections to form a comprehensive evaluation of the entire inspection section. This multi-level index system considers both local characteristics and overall performance, providing a comprehensive and objective basis for the final quality grade assessment.

[0025] A simplified mechanical model is established based on the average cross-sectional test values ​​as equivalent parameters. This model calculates the stress and safety factor at key locations on each test section, identifies instability risk points and stress concentration points, and calculates the comprehensive safety factor for the tested section. This step overcomes the limitations of traditional testing methods that only focus on material parameters, introducing a mechanical analysis perspective and organically combining material quality with structural safety. The simplified mechanical model treats the shotcrete layer as a thin-shell structure under uniformly distributed confining pressure, calculating the stress state at key locations (such as the arch crown, arch waist, and arch foot) based on the principles of elasticity. By comparing the calculated stress with the material strength, the local safety factor is determined, thereby identifying potential instability risk points; by comparing the uniformity of stress distribution, stress concentration points are identified. These analytical results directly reflect the actual working state and safety margin of the support structure.

[0026] Based on the regional average difference rate, regional average deviation rate, regional average self-similarity index, and comprehensive safety factor, a comprehensive evaluation index for regional construction quality is constructed, and the on-site construction quality grade of shotcrete is determined according to preset allowable values. This step integrates statistical indicators, geometric characteristics, and mechanical analysis results to construct a systematic quality evaluation system. The comprehensive evaluation index, through weighted fusion of evaluation elements from multiple dimensions, comprehensively reflects the construction quality status of shotcrete. The preset allowable values ​​are determined based on engineering experience and theoretical analysis, serving as threshold standards for quality grading. The final quality grade assessment considers both material performance indicators and structural safety analysis, providing a scientific basis for engineering decisions.

[0027] In this embodiment of the invention, the detailed implementation steps for establishing a gridded detection network within the tunnel section to be detected include:

[0028] Along the tunnel's direction, the tunnel section to be inspected is divided into equal intervals according to a pre-set longitudinal spacing to determine the number and location of inspection sections. The selection of the longitudinal spacing considers factors such as tunnel design characteristics, changes in surrounding rock conditions, and construction process transition points. A denser layout strategy is typically used in high-risk sections (such as fault fracture zones and water-bearing sections). In practical applications, the longitudinal spacing is generally set at 10-20m, but can be appropriately reduced to 5-10m for special geological conditions or important engineering parts. For long tunnel projects, a segmented inspection strategy can be adopted, with each segment typically controlled within the range of 100-300m in length to ensure the continuity and representativeness of the inspection. Priority is given to key locations such as construction process transition points, areas of geological condition change, and special design sections to ensure the detection of potential quality weaknesses.

[0029] Within each inspection section, using the tunnel's central axis as a reference, the section is divided circumferentially at predetermined angles to determine the number and angular coordinates of inspection positions. The circumferential point layout fully considers the geometric characteristics and stress features of the tunnel section, ensuring that key areas (such as the arch crown, arch waist, and arch foot) are covered by inspection points. Under standard conditions, the circumferential angle interval is typically set at 30°, meaning 12 inspection positions are set for each section. This density comprehensively reflects the parameter distribution while controlling the workload. For sections with special cross-sectional shapes or challenging spraying processes (such as the arch crown area of ​​long-span tunnels), the local inspection density can be appropriately increased, reducing the angle interval to 15°. Each inspection position is precisely located using angular coordinates and radial position, ensuring the repeatability of the inspection and the accuracy of comparative analysis.

[0030] Each detection location is assigned a dual index of cross-section number and circumferential position number, establishing a spatial mapping relationship between the detection cross-section and the detection location, and generating a detection point matrix. The dual indexing system is a key technology for achieving spatial data association, ensuring the identification and spatial location of each detection point through unique coding. The coding rule adopts the format of "cross-section number - circumferential position number," for example, "D03-P06" represents the 6th circumferential position of the 3rd cross-section. This spatial mapping relationship not only facilitates data management and querying but also provides a unified framework for the spatial distribution analysis and visualization of parameters. The complete detection point matrix constitutes a three-dimensional spatial grid, covering the entire volume space of the tunnel section to be detected, realizing the collaborative detection and analysis of thickness and strength parameters under the same spatial reference system.

[0031] In this embodiment of the invention, the detailed implementation steps for simultaneously acquiring spray layer thickness and spray layer intensity detection values ​​at each detection position of the detection array include:

[0032] The surface of the sprayed layer at each detection location undergoes pretreatment to remove surface laitance and protrusions. Surface pretreatment is a crucial step in ensuring detection accuracy, improving signal quality and data reliability by eliminating influencing factors. Pretreatment includes: lightly brushing the surface with a wire brush to remove loose laitance and dust; blowing away surface residue with a high-pressure air gun; and lightly grinding any localized protrusions that may affect detection using a handheld grinder to ensure surface smoothness. The pretreatment area covers a circular region approximately 30-50 cm in diameter centered on the detection point, providing a stable contact surface for the subsequent instrument probe. This seemingly simple step is critical to detection accuracy; good surface conditions significantly improve the quality of radar reflection signals and the stability of ultrasonic wave propagation.

[0033] Ground-penetrating radar (GPR) is used to detect the thickness of the shotcrete layer. A radar antenna moves along the surface at the detection location to collect reflected signals. Based on a preset wave velocity, the two-way travel time of the reflecting interface is converted into a thickness measurement value. GPR is an efficient and non-destructive thickness detection technique that determines the interlayer distance by observing the reflection characteristics of electromagnetic waves at the interface of different media. The detection process first selects an antenna frequency suitable for the shotcrete layer thickness range, typically 1.0-2.5 GHz. Higher frequency antennas provide higher resolution but have a smaller penetration depth. Then, on-site wave velocity calibration is performed. Using the reflection time at a known thickness point, the propagation speed of electromagnetic waves in the shotcrete is calculated. Finally, a system scan is performed to record the transmitted and received electromagnetic signals, and the concrete-surrounding rock interface is identified based on the reflected waveform characteristics. The thickness calculation formula is:

[0034] ;

[0035] in, For the thickness of the spray layer, The speed at which electromagnetic waves propagate in concrete. This represents the two-way travel time of the reflected wave. Wave velocity calibration is a crucial step in ensuring measurement accuracy and is typically determined using local core drilling verification. In concrete under different moisture content and density conditions, the electromagnetic wave velocity usually varies within the range of 9-12 cm / ns.

[0036] The ultrasonic echo composite method is used to test the strength of the sprayed layer. Ultrasonic propagation parameters and rebound test data are simultaneously acquired at the same test location, and the strength value is calculated based on a pre-defined strength relationship model. The ultrasonic echo composite method is a combined strength assessment technique that integrates acoustic characteristics and surface hardness, overcoming the limitations of single methods and improving the accuracy of strength estimation. The testing process first involves ultrasonic testing, where longitudinal wave velocities are acquired through transmitting and receiving probes, reflecting the internal structure and elastic properties of the concrete. Then, a rebound test is performed at the same location to obtain surface hardness values, reflecting the surface strength of the concrete. Finally, the two types of parameters are converted into concrete strength values ​​using a comprehensive relationship model. The general form of strength calculation is as follows:

[0037] ;

[0038] in, For concrete strength, For ultrasonic speed, This is the rebound value. , , The regression coefficients are determined through calibration using standard specimens. This comprehensive evaluation method considers both the internal structural characteristics of concrete and surface hardness indicators, making it more accurate and reliable than strength estimation using single methods, and is particularly suitable for the needs of on-site non-destructive testing.

[0039] Thickness measurements from all locations within the same inspection section are aggregated into a thickness measurement set, and intensity measurements from all locations are aggregated into an intensity measurement set. Data aggregation is a necessary step in forming a complete inspection dataset. Unified organization and formatting facilitate subsequent statistical analysis and spatial modeling. The aggregation process employs structured data management, associating the location information of each inspection point (section number, circumferential position) with its corresponding thickness and intensity values, forming a standardized data table. Each row in the data table represents an inspection location, and columns include location code, spatial coordinates, thickness measurement value, and intensity measurement value. In this way, each inspection section forms two basic data sets: a thickness measurement value set and an intensity measurement value set, which share the same spatial reference system, creating conditions for correlation analysis between parameters. The complete dataset serves as the foundation for statistical analysis and a data source for visualizing the spatial distribution of parameters, directly impacting the scientific rigor and comprehensiveness of subsequent evaluations.

[0040] In this embodiment of the invention, the detailed implementation steps for normalizing the set of thickness detection values ​​and the set of strength detection values, constructing a continuous distribution curve, and calculating the thickness self-similarity index and strength self-similarity index of each detection section based on the grid coverage analysis method include:

[0041] The circumferential coordinates of the detection location are normalized to the interval [0,1] to generate normalized coordinates. Circumferential coordinate normalization is a key step in achieving comparability of data from different cross-sections. By mapping physical angles to a unified interval, the influence of geometric differences in the cross-sections is eliminated. The normalization process starts from the tunnel design reference point (usually the arch) and proceeds clockwise, converting the circumferential angles into relative position values ​​within the interval [0,1] according to the cross-section perimeter. For standard circular or horseshoe-shaped cross-sections, the normalized coordinates can be directly calculated according to the angle ratio; for unconventional cross-sections, the actual arc length ratio must be considered to ensure a linear correspondence between the coordinate values ​​and the physical position. This normalization process makes the data distribution between different cross-sections and different projects comparable, creating conditions for the accumulation of experience and the summarization of patterns across projects.

[0042] The thickness and strength test values ​​are normalized to the interval [0,1], generating normalized coordinates for thickness and strength. Parameter normalization is a crucial step in eliminating dimensional differences and achieving comparability between different parameters. The normalization process uses a minimum-maximum scaling method to linearly map the original test values ​​to the [0,1] interval. For the thickness test value, the normalization formula is:

[0043] ;

[0044] in, The normalized thickness value. This is the original thickness measurement value. and These represent the minimum and maximum thickness values ​​within the cross-section, respectively. The strength values ​​are normalized using the same principle. This normalization method preserves the relative relationships and morphological characteristics of the parameter distributions while eliminating the influence of dimensional differences and varying numerical ranges. This makes the distribution characteristics of thickness and strength parameters directly comparable, facilitating the identification of their correlation patterns and differences.

[0045] The data points composed of normalized position coordinates and normalized thickness coordinates are connected sequentially by straight line segments to form the thickness distribution curve; similarly, the data points composed of normalized position coordinates and normalized intensity coordinates are connected sequentially by straight line segments to form the intensity distribution curve. The construction of the distribution curves is a transitional step from discrete data points to a continuous function expression, establishing a parameter variation model across the entire circumferential range through linear interpolation. The construction process arranges the detection positions in circumferential order, connecting adjacent points with straight line segments to form piecewise linear curves. This method preserves the authenticity of the original data while filling in the parameter variations between detection points, transforming discrete detection results into a continuous function expression. The continuous distribution curves not only visually demonstrate the parameter variation trend along the circumferential direction but also provide a basic graphical basis for subsequent mesh coverage analysis, serving as a prerequisite for calculating the self-similarity index. The thickness and intensity distribution curves reflect the spatial distribution characteristics of the two types of parameters, and their morphological differences and similarities directly reflect the uniformity of the spraying process and the consistency of material properties.

[0046] Within the [0,1]×[0,1] plane, the plane is divided using multiple proportionally reduced grids. For each grid, the number of grids intersecting the thickness distribution curve and the number intersecting the intensity distribution curve are counted. Grid coverage analysis is a curve complexity quantification method based on fractal geometry principles. It reveals the self-similarity and complexity of the distribution curve through the intersection relationship between multi-scale grids and curves. The analysis process first sets up a series of square grids of different sizes on the normalized coordinate plane, with grid side lengths typically decreasing in powers of 2 to form a multi-level coverage grid. Then, the number of grids intersecting the distribution curve at each scale is calculated; this number directly reflects the "coverage degree" of the curve at that scale. Finally, the scale relationship between the grid side length and the number of intersecting grids is analyzed to extract a quantitative complexity index. Intersection determination uses computational geometry algorithms to determine whether a curve segment crosses the grid boundary or is located inside the grid. This multi-scale analysis method can capture subtle changes and overall trends in parameter distribution, making it particularly suitable for evaluating engineering materials such as shotcrete, which exhibit multi-scale inhomogeneity.

[0047] Logarithmic transformations are applied to the mesh edge length and the number of intersecting meshes. The absolute values ​​of the slopes of the fitted lines are obtained through linear fitting, serving as the thickness and intensity self-similarity indices, respectively. Calculating the self-similarity indices is the final step in mesh coverage analysis, extracting the fractal characteristics of the distribution curves through a logarithmic-linear relationship. The calculation process first involves logarithmic transformation of the mesh edge length and the number of intersecting meshes, converting the power-law relationship into a linear one; then, linear regression fitting is performed using the least squares method to obtain the slope and intercept; finally, the absolute value of the slope is taken as the self-similarity index. Ideal fractal curves follow a strict power-law relationship:

[0048] ;

[0049] in, For a grid with side length of The number of intersecting grids at time, The self-similarity index is approximately equal to the fractal dimension of the curve. A larger self-similarity index indicates a more complex and jagged distribution curve, reflecting a higher degree of non-uniformity in parameter distribution; a value closer to 1 indicates a smoother distribution and more uniform parameter distribution. This index overcomes the limitations of traditional statistical methods, quantifying the geometric complexity of parameter distribution and providing a new mathematical tool for evaluating the uniformity of spraying processes.

[0050] In this embodiment of the invention, the detailed implementation steps for extracting multi-scale self-similarity features from each detection section, calculating multiple self-similarity spectrum parameters, and establishing a correlation model between the multiple self-similarity spectrum parameters and construction process parameters include:

[0051] The probability measure for each detection location is calculated based on the normalized thickness and intensity coordinates. The probability measure calculation is a transformation step from parameter values ​​to a statistical distribution. By interpreting the normalized coordinates as probability densities, it forms the basis for multifractal analysis. The calculation process first treats the normalized parameter values ​​as relative weights, then transforms them into a probability distribution through standardization, ensuring the sum is 1. For the thickness parameter, the probability measure calculation formula is:

[0052] ;

[0053] in, For the first Thickness probability measure at each detection location For the corresponding normalized thickness value, This represents the total number of detected locations. The probability measure of the intensity parameter is calculated using a similar method. The probability measure transforms static parameter values ​​into dynamic distribution characteristics, providing a statistical basis for subsequent multiple self-similarity analysis, and is particularly suitable for analyzing complex distributions exhibiting local singularities.

[0054] For sub-intervals divided by different grid side lengths, the sum of probability measures within each sub-interval is calculated to generate a grid probability measure. The grid probability measure is a clustered calculation of distribution characteristics at different scales, revealing the scale-dependent characteristics of the distribution by changing the observation scale. The calculation process first divides the [0,1] interval into 1 / ε equal-length sub-intervals according to the grid side length ε; then, the sum of the probability measures of the detection points falling within each sub-interval is calculated as the grid probability measure for that sub-interval; finally, a complete grid probability measure sequence is formed, reflecting the clustering characteristics of the probability distribution at a specific scale. This multi-scale clustering analysis can effectively capture the local concentration and overall uniformity of the distribution, and is an important method for understanding the intrinsic structure of complex distributions.

[0055] A sequence of moment orders, including negative, zero, and positive values, is defined. For each moment order, a partition function is calculated, and the quality index is obtained through linear regression. Partition function calculation is a core step in multifractal analysis, comprehensively characterizing the multi-scale properties of the distribution through the scale behavior of different moment orders. The calculation process first defines a set of moment orders q, typically selecting several points within the range [-5, 5]. Then, for each moment order and grid size, the partition function is calculated, which is the sum of the q powers of the grid probability measure. Finally, logarithmic linear regression is performed on the partition functions under different grid sizes to obtain the slope as the quality index τ(q). This process extends from a single self-similarity index to a complete quality index spectrum, enabling the capture of scale behavior differences across different probability regions.

[0056] The singularity index is calculated using the difference method based on the mass index, and the multiple self-similar spectrum is calculated by combining the mass index and the singularity index. The calculation of the singularity index and the multiple self-similar spectrum are crucial steps in revealing the local singularity and overall regularity of a distribution. The singularity index α(q) is obtained by differentiating the mass index τ(q), reflecting the local singularity intensity corresponding to a specific q value; the multiple self-similar spectrum f(α) is calculated using the Legendre transform, representing the fractal dimension of the subset with the singularity index α. The multiple self-similar spectrum provides a panoramic view of the distribution's complexity; the spectral width indicates the degree of non-uniformity, and the spectral shape reflects the dominant non-uniformity pattern. A narrow spectrum indicates a relatively uniform distribution, while a wide spectrum indicates significant multi-scale non-uniformity.

[0057] Spectral width and spectral asymmetry index are extracted as parameters from multi-self-similar spectra. Spectral parameter extraction simplifies complex spectral information into key indicators, capturing the main features of the spectrum through a small number of parameters. Spectral width is defined as the difference between the maximum and minimum singularity indices, intuitively reflecting the degree of heterogeneity of the distribution. The spectral asymmetry index is calculated based on the shape difference between the left and right sides of the spectral curve, reflecting the contrast in singularity between high-probability and low-probability regions. A positive asymmetry index indicates stronger singularity in high-probability regions, while a negative value indicates more significant singularity in low-probability regions. These spectral parameters are distributional morphological features that traditional statistical methods cannot provide, and they have unique value in identifying potential anomalies in shotcrete quality.

[0058] A nonlinear regression-based correlation model was established using construction process parameter vectors as independent variables and multiple self-similar spectral parameters as dependent variables. This correlation model serves as a bridge connecting geometric features and process parameters, revealing their intrinsic relationship through a data-driven approach. The modeling process first collects multiple sets of construction data, including process parameters such as wind pressure, spraying distance, and accelerator dosage, as well as corresponding spectral parameters such as spectral width and asymmetry index. Then, nonlinear mapping relationships are established using machine learning methods (such as random forests or support vector regression). Finally, cross-validation is used to evaluate the model's performance and generalization ability. The establishment of this correlation model not only enables the diagnostic function of inferring process conditions from spectral parameters but also provides theoretical guidance for optimizing construction parameters, promoting the transformation of spraying processes from experience-driven to data-driven.

[0059] In this embodiment of the invention, a simplified mechanical model is established based on the average cross-sectional detection value as an equivalent parameter. The detailed implementation steps for calculating the stress at key locations and the cross-sectional safety factor of each detection cross-section, identifying instability risk points and stress concentration points, and calculating the comprehensive safety factor of the detection section include:

[0060] The average thickness and average strength of each test section are used as the equivalent thickness and equivalent strength, respectively. Defining the equivalent parameters is a transitional step from distributed data to a simplified model, reflecting overall characteristics through representative parameters. The definition process is based on the average cross-sectional value, simplifying the complex spatial distribution into a single parameter, facilitating mechanical calculations and safety assessments. The equivalent thickness directly uses the average cross-sectional thickness, reflecting the overall geometric characteristics of the sprayed layer; the equivalent strength uses the average cross-sectional strength, representing the overall mechanical properties of the material. This simplification significantly reduces the complexity of mechanical analysis while preserving overall characteristics, making rapid safety assessments possible. In practical applications, the selection principles for the equivalent parameters can be appropriately adjusted according to risk preferences and safety requirements. For example, in high-risk projects, the lower quartile values ​​can be used instead of the arithmetic mean to increase the conservatism of the assessment.

[0061] Based on the tunnel design profile and surrounding rock conditions, the shotcrete layer is treated as a thin-shell structure under uniformly distributed confining pressure, and a simplified mechanical model is established. Establishing this simplified mechanical model is a crucial step in abstracting complex engineering structures into a computable theoretical model. The model construction first determines the cross-sectional geometry based on the tunnel design drawings, typically simplified to a circular or elliptical cross-section. Then, the surrounding rock pressure distribution is determined based on geological reports and monitoring data, generally assumed to be a uniformly distributed load. Finally, the shotcrete layer is treated as an elastic thin-shell structure, and shell theory is applied for stress analysis. While this simplified model neglects some complex factors (such as the nonlinear characteristics of the rock-support interaction), it effectively captures the main stress mechanisms, providing a reasonable engineering approximation solution for safety assessment, and is particularly suitable for the needs of rapid on-site assessment.

[0062] For each inspection section, the tangential stress at three key locations—the arch crown, arch waist, and arch foot—is calculated, and the local safety factor at each key location is calculated based on the ratio of equivalent strength to tangential stress. Calculating stress at key locations is a core step in assessing the stress state of the support structure. By analyzing the stress levels of typical locations, potential weak points are identified. The calculation process is based on the classical solution of thin-shell theory, considering the tunnel geometry, shotcrete thickness, and external loads to solve for the stress state at each point. Since shotcrete primarily operates through membrane stress, the calculation focuses on the tangential stress component, the magnitude of which directly affects the stability and durability of the structure. The local safety factor is determined by the ratio of equivalent strength to calculated stress, reflecting the safety margin between material strength and actual stress. This analysis method based on key locations avoids the complexity of detailed calculations across the entire section while effectively identifying potential safety hazards, providing an intuitive and effective technical basis for engineering decisions.

[0063] The minimum local safety factor at each critical location is taken as the cross-sectional safety factor for that inspection section. The determination of the cross-sectional safety factor follows the "barrel principle," meaning that overall safety is controlled by the weakest link. The strategy of taking the minimum value ensures the conservatism of the safety assessment, avoiding the risk of high-safety-factor areas masking low-safety-factor areas. As an overall indicator, the cross-sectional safety factor reflects the overall safety status of the shotcrete support at that section and is an important basis for risk assessment and quality level determination. The threshold for the safety factor is usually determined based on the importance and risk level of the project, generally within the range of 1.2-1.5, with higher safety requirements adopted for high-risk projects.

[0064] Locations with local safety factors less than a preset safety threshold are marked as instability risk points, and locations with tangential stress greater than the product of the average cross-sectional stress and a preset amplification factor are marked as stress concentration points. Risk point identification is the concrete implementation of safety assessment, screening potential problem areas through quantitative standards. Instability risk points represent areas with insufficient safety margins, which may crack or deform during service life; stress concentration points indicate areas of abnormally concentrated stress, which are potential sources of fatigue damage or progressive failure. The identification of these two types of risk points allows safety assessment to delve from overall indicators to specific locations, providing target areas for precise maintenance and reinforcement, greatly improving the practicality and operability of the assessment results. The spatial distribution patterns of risk points can also reflect systemic problems in construction techniques or design schemes, providing directional guidance for quality improvement.

[0065] The average safety factor of all tested sections within the tested area is used to generate the comprehensive safety factor. The comprehensive safety factor is a quantitative expression of the overall safety status of the section, reflecting the safety level of the entire tested area through statistical averaging. The calculation process uses an arithmetic mean method to ensure the equal contribution of each section, ultimately yielding a single index representing the overall safety margin of the section. The comprehensive safety factor is not only an important component of comprehensive quality evaluation but also a key reference for construction management decisions. A higher index value indicates a safer and more reliable support structure; a value closer to the critical threshold indicates greater potential risk, requiring more cautious management strategies and more frequent monitoring plans.

[0066] In this embodiment of the invention, the detailed implementation steps for constructing a comprehensive evaluation index for regional construction quality and determining the on-site construction quality level of shotcrete according to preset allowable values ​​include:

[0067] The ratio of the regional average deviation rate to the regional design offset is calculated to generate a dimensionless regional average deviation rate index. Dimensionless processing is a crucial step in eliminating the influence of dimensions and achieving objective comparison. The process converts dimensional statistics into a dimensionless index by constructing the ratio of the regional average deviation rate to the design offset. This conversion eliminates the influence of measurement units and numerical values, allowing direct comparison of deviation rates between different items and parameters. The dimensionless deviation rate index intuitively reflects the degree of actual variation relative to the design allowable variation. A value closer to 1 indicates that the actual deviation is closer to the design expectation, while a value significantly greater than 1 indicates that the variation exceeds the design allowable range, potentially indicating quality problems.

[0068] The regional average difference rate, the dimensionless regional average deviation rate, and the regional average self-similarity index are weighted and summed to construct regional thickness construction quality evaluation indicators and regional strength construction quality evaluation indicators, respectively. The construction of individual quality indicators is a key step in integrating multi-dimensional evaluation elements, forming a comprehensive evaluation standard through weighted fusion. The construction process first determines the weight coefficients of each indicator, reflecting its relative importance in quality evaluation; then, the normalized indicators are linearly combined according to their weights to form a single comprehensive score. Weight settings are usually based on expert judgment and historical data analysis. Thickness evaluation generally focuses more on the difference rate (reflecting design compliance), while strength evaluation focuses more on the deviation rate (reflecting uniformity). This weighting strategy ensures that the evaluation indicators simultaneously consider multiple dimensions of quality characteristics, avoiding the one-sidedness of single-indicator evaluation and providing a comprehensive basis for scientific and objective quality judgment.

[0069] The regional construction quality comprehensive evaluation index is generated by weighted summing of the regional thickness construction quality evaluation index, the regional strength construction quality evaluation index, and the reciprocal of the comprehensive safety factor. Constructing the comprehensive evaluation index is the final step in integrating thickness, strength, and safety into a unified quality standard. The construction process uses a similar weighted summation method, comprehensively considering the two individual indicators and the reciprocal of the safety factor (converted into a "risk index," with higher values ​​indicating higher risk) to form the final quality score. The weighting of the three indicators reflects the emphasis of the project's focus; typically, thickness and strength indicators have similar weights, while the safety indicator has a slightly higher weight, reflecting the "safety first" engineering philosophy. The safety factor is used in reciprocal form to align its direction with the other two indicators (higher values ​​indicate worse quality), facilitating linear combination. The comprehensive evaluation index, as the final judgment basis, comprehensively reflects the multi-dimensional characteristics of shotcrete construction quality and is a scientifically sound quality evaluation standard.

[0070] Preset allowable values ​​are calculated based on tunnel type, safety level, and surrounding rock stability. Determining these preset allowable values ​​is fundamental to establishing quality level standards. By considering project characteristics and risk levels, reasonable quality thresholds are set. The calculation process first determines benchmark values ​​based on tunnel type (e.g., highway, railway, hydraulic engineering) to reflect the basic requirements of different project types. Then, coefficients are adjusted according to safety level (e.g., special grade, first grade, second grade), with higher safety level projects using stricter standards. Finally, surrounding rock stability is considered, and allowable values ​​are appropriately relaxed for adverse geological conditions, reflecting the principle of "rationality" in engineering. This multi-factor adjustment mechanism ensures that the preset allowable values ​​not only conform to standards but also have engineering relevance, more accurately reflecting the actual quality requirements of specific projects.

[0071] The comprehensive evaluation index for regional construction quality is compared with the preset allowable value, and the on-site construction quality grade of shotcrete is determined based on the range of the ratio. Quality grade determination is the final output of the evaluation process. A clear quality grade is defined by comparing the actual score with the standard threshold. The comparison process calculates the ratio of the comprehensive evaluation index to the preset allowable value, and then determines the grade based on the range of the ratio. The conventional grade classification is into four levels: when the ratio is ≤1, it is "Grade A (Excellent)," indicating that all indicators are better than the design requirements; when the ratio is between 1 and 1.2, it is "Grade B (Qualified)," indicating that the requirements are basically met but there are minor deficiencies; when the ratio is between 1.2 and 1.5, it is "Grade C (Warning)," indicating that there are obvious quality defects requiring rectification; when the ratio is >1.5, it is "Grade D (Dangerous)," indicating that there are serious quality problems requiring immediate attention. This ratio-based grading method is both simple and intuitive, and has a sound theoretical basis, providing clear judgment standards and handling guidelines for engineering quality management.

[0072] In this embodiment of the invention, the method further includes:

[0073] A construction case library is established by collecting construction process parameters and corresponding regional construction quality comprehensive evaluation indicators from completed sections. Establishing this library is a crucial step in accumulating experience and transferring knowledge, transforming scattered engineering experience into structured knowledge through systematic recording and organization. The library construction process begins with designing standardized data collection forms to ensure consistency and completeness across projects. Next, a database structure is established, including multiple dimensions such as basic project information, working conditions, construction parameters, and quality indicators. Finally, systematic data entry and review result in a structured case library. Case data includes at least the following process parameters: surrounding rock grade, excavation method, spraying equipment model, air pressure, spraying distance, accelerator dosage, and nozzle movement speed, as well as corresponding comprehensive evaluation indicators and quality grades. The case library not only forms the data foundation for construction optimization but also serves as an important vehicle for engineering knowledge management and technology transfer, possessing long-term value for improving the overall technical level of the industry.

[0074] For the working conditions of the section to be constructed, historical cases with similar working conditions are matched from the construction case library. The construction process parameters corresponding to the case with the best comprehensive evaluation index of regional construction quality are selected as recommended parameters. Construction parameter optimization and recommendation is the core function of the case library application. Through similarity matching and optimal selection, it provides experience-based guidance for the current project. The recommendation process first filters historical cases with similar working conditions from the case library based on the current project's surrounding rock type, groundwater conditions, cross-sectional shape, and other working condition characteristics. Then, these cases are sorted according to the comprehensive evaluation index, and the case with the best index (smallest value) is selected as a reference template. Finally, the combination of construction process parameters from this case is extracted as the recommended parameters for the current project. This case-based recommendation method combines historical experience and data analysis, possessing both practical foundation and scientific rigor. It can effectively guide construction practice, avoid repeated trial and error, and improve construction efficiency and quality stability.

[0075] Construction is carried out according to recommended parameters. After construction is completed, testing and evaluation processes are implemented, and the comprehensive evaluation indicators of the regional construction quality of the new section are added to the construction case library. Closed-loop quality management is a key mechanism for achieving continuous improvement, forming a complete quality improvement cycle through feedback and updates. The management process first applies the recommended parameters to actual construction, strictly controlling parameter stability and operational standardization; after construction is completed, testing and evaluation are carried out according to standard procedures to obtain the quality indicators of the new section; finally, the new construction parameters and quality results are recorded in the case library to enrich the data sample and optimize the model accuracy. This closed-loop mechanism of "recommendation-construction-testing-feedback" realizes the transformation from experience-driven to data-driven quality management, forming a self-improving and continuously enhancing quality management system, which is an important symbol of the modernization of engineering quality management.

[0076] This invention achieves comprehensive detection and scientific evaluation of shotcrete construction quality through gridded detection network construction, collaborative detection of thickness and strength, self-similar feature extraction, safety assessment model establishment, and comprehensive quality evaluation. The multi-dimensional evaluation method of this invention can accurately describe parameter distribution characteristics and effectively identify local defects, providing a systematic solution for quality control in tunnel engineering construction.

[0077] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0078] It should be noted that all formulas in this manual are calculated by removing dimensions and taking their numerical values. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0079] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for testing and evaluating the on-site construction quality of shotcrete, characterized in that, include: A gridded detection network is established within the tunnel section to be inspected. Multiple detection sections are set at fixed intervals along the tunnel direction. Multiple detection positions are set at fixed angles along the circumferential direction within each detection section, forming a detection point array corresponding to spatial positions. At each detection location of the detection point array, the thickness detection value and the intensity detection value of the spray layer are collected simultaneously to generate a set of thickness detection values ​​and a set of intensity detection values ​​for each detection section. Based on the set of thickness detection values ​​and the set of strength detection values, the average detection value and the degree of cross-sectional offset of each detection section are calculated respectively. The set of thickness detection values ​​and the set of strength detection values ​​are normalized to construct a continuous distribution curve, and the thickness self-similarity index and strength self-similarity index of each detection section are calculated based on the grid coverage analysis method. Based on the thickness self-similarity index and the strength self-similarity index, multi-scale self-similarity features are extracted for each detection section, multiple self-similarity spectrum parameters are calculated, and a correlation model between the multiple self-similarity spectrum parameters and construction process parameters is established. The regional design offset value is calculated based on the design value of each detection section, and the difference rate index of each detection section is established in combination with the average detection value of the section. Then, the regional average difference rate, regional average deviation rate and regional average self-similarity index are statistically analyzed. A simplified mechanical model is established based on the average detection value of the cross section as an equivalent parameter. The stress at key locations and the safety factor of the cross section are calculated, the instability risk points and stress concentration points are identified, and the comprehensive safety factor of the detection section is calculated. Based on the regional average difference rate, the regional average deviation rate, the regional average self-similarity index, and the comprehensive safety factor, a comprehensive evaluation index for regional construction quality is constructed, and the on-site construction quality level of shotcrete is determined according to the preset allowable value.

2. The method for testing and evaluating the on-site construction quality of shotcrete according to claim 1, characterized in that, The establishment of a gridded detection network within the tunnel section to be inspected includes: Along the tunnel's direction, the tunnel section to be inspected is divided into equal intervals according to a preset longitudinal spacing to determine the number and location of inspection sections. Within each of the detection sections, the circumference of the section is divided into equal angles according to the preset circumferential angle, with the tunnel centerline as the reference, to determine the number of detection positions and their angular coordinates. Each detection location is assigned a dual index of cross-sectional number and circumferential position number to establish a spatial mapping relationship between the detection cross-section and the detection location, thereby generating the detection point matrix.

3. The method for testing and evaluating the on-site construction quality of shotcrete according to claim 1, characterized in that, The simultaneous acquisition of spray layer thickness and spray layer intensity detection values ​​at each detection position of the detection array includes: The sprayed surface at each detection location is pretreated to remove surface laitance and protrusions; The thickness of the sprayed layer is detected by ground-penetrating radar. The radar antenna moves along the surface at the detection location to collect reflected signals. The two-way travel time of the reflecting interface is converted into a thickness detection value based on a preset wave velocity. The strength of the sprayed layer was tested using the ultrasonic echo synthesis method. Ultrasonic propagation parameters and rebound test data were collected simultaneously at the same test location, and the strength test value was calculated based on a preset strength relationship model. The thickness values ​​at all detection locations within the same detection section are aggregated into a thickness value set, and the strength values ​​at all detection locations are aggregated into a strength value set.

4. The method for testing and evaluating the on-site construction quality of shotcrete according to claim 1, characterized in that, The normalization process for the set of thickness and the set of strength detection values, the construction of a continuous distribution curve, and the calculation of the thickness self-similarity index and strength self-similarity index for each detection section based on the grid coverage analysis method include: The circumferential coordinates of the detected position are normalized to the interval [0,1] to generate normalized position coordinates; The thickness and strength detection values ​​are normalized to the interval [0,1] to generate normalized thickness coordinates and normalized strength coordinates; The data point set composed of the normalized position coordinates and the normalized thickness coordinates is connected sequentially with straight line segments to form a thickness distribution curve; the data point set composed of the normalized position coordinates and the normalized intensity coordinates is connected sequentially with straight line segments to form an intensity distribution curve. Within the [0,1]×[0,1] plane, the plane is divided using multiple proportionally reduced grid side lengths. For each grid side length, the number of grids intersecting the thickness distribution curve and the number of grids intersecting the intensity distribution curve are counted. Logarithmically transform the grid side length and the number of intersecting grids, and obtain the absolute value of the slope of the fitted line through linear fitting, which are used as the thickness self-similarity index and the intensity self-similarity index, respectively.

5. The method for testing and evaluating the on-site construction quality of shotcrete according to claim 1, characterized in that, The process of extracting multi-scale self-similarity features from each detection section, calculating multiple self-similarity spectrum parameters, and establishing a correlation model between the multiple self-similarity spectrum parameters and construction process parameters includes: The probability measure of each detection location is calculated based on the normalized thickness coordinates and intensity coordinates; For sub-intervals divided by different grid side lengths, calculate the sum of probability measures within each sub-interval to generate grid probability measures; Define a sequence of moment orders containing negative, zero, and positive values, calculate the partition function for each moment order, and obtain the quality index through linear regression; The singularity index is calculated based on the difference of the mass index, and the multiple self-similar spectrum is calculated by combining the mass index and the singularity index. The spectral width and spectral asymmetry index are extracted from the multiple self-similar spectrum as parameters of the multiple self-similar spectrum; Using the construction process parameter vector as the independent variable and the multiple self-similar spectrum parameters as the dependent variable, a nonlinear regression-form correlation model is established.

6. The method for testing and evaluating the on-site construction quality of shotcrete according to claim 1, characterized in that, The simplified mechanical model is established based on the average detection value of the cross-section as an equivalent parameter. The stress at key locations and the cross-sectional safety factor are calculated for each detection cross-section. Instability risk points and stress concentration points are identified, and the comprehensive safety factor of the detection section is calculated, including: The average thickness and average strength of each test section are used as the equivalent thickness and equivalent strength, respectively. Based on the tunnel design profile and surrounding rock conditions, the sprayed layer is regarded as a thin shell structure subjected to uniformly distributed confining pressure, and the simplified mechanical model is established. For each test section, the tangential stress at three key locations—the arch crown, arch waist, and arch foot—is calculated, and the local safety factor at each key location is calculated based on the ratio of the equivalent strength to the tangential stress. The minimum local safety factor at each critical location is taken as the cross-sectional safety factor of the detection section. Locations with a local safety factor less than a preset safety threshold are marked as instability risk points, and locations with tangential stress greater than the product of the average cross-sectional stress and a preset amplification factor are marked as stress concentration points. The average value of the safety factors of all test sections within the test area is used to generate the comprehensive safety factor.

7. The method for testing and evaluating the on-site construction quality of shotcrete according to claim 1, characterized in that, The comprehensive evaluation index for construction quality in the constructed area, and the determination of the on-site construction quality level of shotcrete based on preset allowable values, include: The ratio of the regional average deviation rate to the regional design offset value is calculated to generate a dimensionless regional average deviation rate index. The regional average difference rate, the dimensionless regional average deviation rate index, and the regional average self-similarity index are weighted and summed to construct regional thickness construction quality evaluation index and regional strength construction quality evaluation index, respectively. The regional construction quality comprehensive evaluation index is generated by weighting and summing the regional thickness construction quality evaluation index, the regional strength construction quality evaluation index, and the reciprocal of the comprehensive safety factor. The preset allowable value is calculated based on the tunnel type, safety level, and surrounding rock stability; The comprehensive evaluation index of construction quality in the area is compared with the preset allowable value, and the on-site construction quality level of shotcrete is determined according to the range of the ratio.

8. The method for testing and evaluating the on-site construction quality of shotcrete according to claim 1, characterized in that, Also includes: Collect construction process parameters and corresponding regional construction quality comprehensive evaluation indicators for completed sections, and establish a construction case library; For the working conditions of the section to be constructed, historical cases with similar working conditions are matched from the construction case library, and the construction process parameters corresponding to the case with the best regional construction quality comprehensive evaluation index are selected as recommended parameters. Construction shall be carried out according to the recommended parameters. After the construction is completed, the testing and evaluation process shall be carried out, and the comprehensive evaluation index of the regional construction quality of the new section shall be added to the construction case library.

Images