Laboratory static load evaluation method for surrounding rock control effect of coal mine spraying support material

Abstract

The invention relates to the technical field of coal mine tunnel support, and provides a laboratory static load evaluation method for the surrounding rock control effect of a coal mine spraying support material. The method comprises the following steps: acquiring a static load test data set of a spraying sample group and an unprocessed sample group, wherein the data set comprises load, strain and acoustic emission signals acquired through a static load test; based on the static load test data set, multi-dimensional evaluation indexes representing the cooperative control effect of the spraying material and the surrounding rock are calculated; inputting the multi-dimensional evaluation indexes into a trained neural network model to obtain a comprehensive judgment result of the surrounding rock control effect of the spraying support material; and the neural network model outputs a comprehensive judgment result by checking physical consistency of changes among different multi-dimensional evaluation indexes. The problem that in the prior art, the actual supporting effect after interaction of the spraying supporting material and the surrounding rock cannot be reflected only by carrying out physical and mechanical property testing on the spraying supporting material can be solved, and the control effect of different spraying supporting materials on the surrounding rock under the static loading condition can be scientifically evaluated.

Description

Technical Field

[0001] This invention relates to the field of coal mine roadway support technology, and in particular to a laboratory static load evaluation method for the surrounding rock control effect of coal mine sprayed support materials. Background Technology

[0002] In underground coal mine engineering, roadways are the critical lifeline for ensuring production, transportation, and ventilation. With the continuous increase in mining depth and ground stress, the stability control of the surrounding rock in roadways faces increasingly severe challenges, especially in soft rock roadways with complex geological conditions, where problems such as large deformation, loosening, and fracturing of the surrounding rock are particularly prominent, seriously threatening the safe production of the mine.

[0003] In recent years, novel coal mine spray support technology has emerged and developed rapidly. This technology involves spraying special polymer materials (such as polyurethane, epoxy resin, and modified cement-based composite materials) onto the surface of the surrounding rock in the roadway. After rapid solidification, a dense, high-strength sprayed layer with good adhesion and toughness is formed. This artificial structure can effectively seal cracks on the surface of the surrounding rock, prevent weathering and deliquescence, and provide continuous radial and tangential support forces through coordinated deformation with the surrounding rock. This improves the stress state of the surrounding rock and controls its deformation and failure, making it a highly promising active and flexible support method.

[0004] Currently, the evaluation of the mechanical properties of sprayed support materials under static loading conditions mainly focuses on the physical and mechanical properties of the sprayed support materials themselves. Although these tests can reflect the basic performance of the sprayed support materials, they cannot directly reveal the actual support effect after their interaction with the surrounding rock. Summary of the Invention

[0005] This invention provides a laboratory static load evaluation method for the rock-control effect of sprayed support materials in coal mines. This method addresses the problem in existing technologies where testing the physical and mechanical properties of the sprayed support material itself cannot reflect its actual support effect after interaction with the surrounding rock. It enables a scientific evaluation of the rock-control effect of different sprayed support materials under static loading conditions. The technical solution proposed by this invention is as follows: In a first aspect, the present invention provides a laboratory static load evaluation method for the surrounding rock control effect of sprayed support materials in coal mines, comprising: Obtain static load test datasets for sprayed and untreated sample groups, the datasets containing load, strain, and acoustic emission signals acquired through static load tests; Based on the static load test dataset, a multidimensional evaluation index characterizing the synergistic control effect of the sprayed material and the surrounding rock is calculated. The multidimensional evaluation indicators are input into a trained neural network model to obtain a comprehensive judgment result on the rock control effect of the sprayed support material; wherein, the neural network model outputs the comprehensive judgment result on the rock control effect of the sprayed support material by verifying the physical consistency of the changes between different multidimensional evaluation indicators.

[0006] Optionally, based on the static load test dataset, a multidimensional evaluation index characterizing the synergistic control effect of the sprayed material and the surrounding rock is calculated, including: From the static load test dataset, the raw data used to calculate various evaluation indicators are extracted; the raw data includes: stress-strain curve feature point data, cumulative parameters of acoustic emission signals, test peak intensity, and relevant geometric parameters; Based on the original data, determine the parameters used for inter-group comparison of the sprayed sample group and the untreated sample group for each of the evaluation indicators; The evaluation value of each evaluation index is determined based on the parameters applied to the inter-group comparison between the sprayed sample group and the untreated sample group. The calculated evaluation values ​​are combined to form the multidimensional evaluation index.

[0007] Optionally, the static load test includes uniaxial compression test, uniaxial tensile test, shear strength test, point load strength test and triaxial compression test; The acoustic emission signal was acquired synchronously during the uniaxial compression test.

[0008] Optionally, the multidimensional evaluation index includes multiple parameters such as uniaxial compressive strength change rate, uniaxial compressive elastic modulus change rate, uniaxial compressive Poisson's ratio change rate, uniaxial secant modulus change rate, uniaxial initial deformation index, uniaxial post-peak residual strength change rate, uniaxial failure strain change rate, brittleness index change rate, fracture suppression index, acoustic emission energy release change rate, tensile strength enhancement index, shear strength change rate, point load strength change rate, triaxial compressive strength change rate, cohesion change index, internal friction angle change rate, triaxial compressive elastic modulus change rate, triaxial residual strength change rate, triaxial failure strain change rate, and performance dispersion improvement coefficient.

[0009] Optionally, the neural network model is trained in the following manner: Obtain a training dataset, which contains multiple sets of sample data. Each set of sample data includes a multidimensional evaluation index calculated from the mechanical response data of sprayed and untreated samples, as well as the corresponding benchmark evaluation level label. The training dataset is augmented using a variational autoencoder model to generate augmented samples for model training, resulting in an augmented dataset. Using the evaluation metrics in the augmented dataset as input features and the corresponding benchmark evaluation level labels as training objectives, the deep neural network model is trained by minimizing the composite loss function, and the trained deep neural network model is used as the neural network model. The composite loss function is determined based on the classification error loss and the physical consistency constraint loss. The classification error loss is determined based on the predicted probability distribution of the model output and the true benchmark rating label. The physical consistency constraint loss is determined based on the predicted score relationship of sample pairs in the training batch. The sample pairs are selected based on the physical performance relationship of the input features of the sample pairs, and the sample with better physical performance should have a predicted score no lower than that of the sample with worse physical performance.

[0010] Optionally, the physical consistency constraint loss is determined as follows: For all selected sample pairs, the hinge loss is calculated based on the prediction score relationship of each sample pair, and the physical consistency constraint loss is determined based on the hinge loss of all sample pairs. Specifically, for a sample pair consisting of a sample with better physical performance and a sample with worse physical performance, based on the input features, the prediction score of the sample with better physical performance and the prediction score of the sample with worse physical performance are calculated respectively. If the prediction score of the sample with better physical performance is less than or equal to the prediction score of the sample with worse physical performance, the hinge loss is calculated based on the prediction scores of the sample with better physical performance and the sample with worse physical performance. Otherwise, the hinge loss of the sample pair is zero.

[0011] Secondly, the present invention also provides a laboratory static load evaluation device for the surrounding rock control effect of coal mine sprayed support materials, comprising the following modules: The data acquisition module is used to acquire static load test datasets of the sprayed sample group and the untreated sample group. The datasets include load, strain and acoustic emission signals collected through static load tests. The parameter calculation module is used to calculate multi-dimensional evaluation indicators characterizing the synergistic control effect of the sprayed material and the surrounding rock based on the static load test dataset. The effect evaluation module is used to input the multi-dimensional evaluation indicators into a trained neural network model to obtain a comprehensive judgment result on the rock control effect of the sprayed support material; wherein, the neural network model outputs the comprehensive judgment result on the rock control effect of the sprayed support material by verifying the physical consistency of the changes between different multi-dimensional evaluation indicators.

[0012] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the laboratory static load evaluation method for the surrounding rock control effect of coal mine sprayed support materials as described in the first aspect above.

[0013] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a laboratory static load evaluation method for the surrounding rock control effect of coal mine sprayed support materials as described in the first aspect above.

[0014] Fifthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements a laboratory static load evaluation method for the surrounding rock control effect of coal mine sprayed support materials as described in the first aspect above.

[0015] Based on the above technical solution, the beneficial effects of the present invention compared with the prior art are as follows: This invention provides a laboratory static load evaluation method for the rock-surrounding control effect of sprayed support materials in coal mines. It obtains static load test datasets for sprayed and untreated sample groups, including load, strain, and acoustic emission signals collected through static load tests. By setting up a comparison between the sprayed and untreated sample groups and incorporating various key data (i.e., load, strain, and acoustic emission signals) generated during interaction with the surrounding rock, it can directly obtain the actual performance data of the sprayed support material under interaction with the surrounding rock. This overcomes the shortcomings of existing technologies that only test the material itself, providing data closer to actual working conditions for subsequent evaluation. Based on the obtained static load test dataset, this invention calculates multi-dimensional evaluation indicators characterizing the synergistic control effect between the sprayed material and the surrounding rock. These multi-dimensional evaluation indicators comprehensively consider various factors of the sprayed support material and the surrounding rock under static loading conditions, reflecting the control effect under their synergistic action from multiple perspectives. Compared to existing technologies that only focus on the material's own performance, this method more scientifically and accurately reflects the actual support effect. This invention inputs the calculated multidimensional evaluation indicators into a trained neural network model to obtain a comprehensive judgment result on the rock control effect of sprayed support materials. The neural network model outputs results by verifying the physical consistency of changes among different multidimensional evaluation indicators. This evaluation method based on a neural network model can fully utilize the complex relationships between multidimensional evaluation indicators, simulate the influence of multiple factors interacting on the rock control effect under actual working conditions, avoid the one-sidedness of single-indicator evaluation, and realize the scientific evaluation of the rock control effect of different sprayed support materials under static loading conditions. It effectively solves the problem that existing technologies cannot scientifically evaluate the actual support effect.

[0016] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is a schematic flowchart of the laboratory static load evaluation method for the surrounding rock control effect of coal mine sprayed support materials provided by the present invention.

[0020] Figure 2 This is a technical roadmap for the laboratory static load evaluation method of the surrounding rock control effect of coal mine sprayed support materials provided by the present invention.

[0021] Figure 3 This is a schematic diagram of the immersion method for preparing sprayed rock samples provided by the present invention.

[0022] Figure 4 This is a uniaxial compression axial stress-strain curve provided by the present invention.

[0023] Figure 5 This is the tensile stress-displacement curve provided by the present invention.

[0024] Figure 6 This is a triaxial compression axial stress-strain curve provided by the present invention.

[0025] Figure 7 This is a flowchart of the comprehensive evaluation process provided by the present invention.

[0026] Figure 8 This is a schematic diagram of the laboratory static load evaluation device for the control effect of coal mine sprayed support material on surrounding rock provided by the present invention.

[0027] Figure 9 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0029] Currently, the evaluation of the mechanical properties of sprayed support materials under static loading conditions mainly focuses on testing the physical and mechanical properties of the materials themselves. For example, basic parameters such as tensile strength, bond strength, and shear strength are measured through laboratory tests. While these tests can reflect the basic performance of the sprayed support materials, they cannot directly reveal their actual support effect after interaction with the surrounding rock. The essence of sprayed support lies in stabilizing the roadway through the synergistic effect of the material and the surrounding rock. Its rock control effect depends not only on the properties of the material itself but also on the properties of the surrounding rock, the interfacial bonding condition, and the performance of the combined structure formed by the support layer and the surrounding rock. Therefore, considering the coupled mechanical properties of the sprayed material and the rock in the evaluation of the mechanical properties of sprayed support materials better reflects the actual role the material plays in the field, providing a scientific reference for the research and development of sprayed support materials and the evaluation of their support effectiveness.

[0030] Currently, the evaluation of the performance of such sprayed support materials mostly draws on traditional testing methods in the fields of concrete, shotcrete, or rock mechanics. However, natural rocks are formed due to geological origins, and their interiors inevitably contain many defects such as cracks and cleavage. This results in a significant difference in the density of the rock itself compared to artificial materials such as concrete. Therefore, directly drawing on traditional testing methods still has obvious limitations.

[0031] In summary, the current evaluation system for sprayed support materials in coal mines is incomplete. Therefore, there is an urgent need in this field for a relatively simple and repeatable testing method to scientifically evaluate the rock control effect of different sprayed support materials under static loading conditions, in order to overcome the shortcomings of existing technologies. Based on this, this invention provides a laboratory static load evaluation method for the rock control effect of sprayed support materials in coal mines.

[0032] Reference Figure 1 As shown, the laboratory static load evaluation method for the surrounding rock control effect of the coal mine sprayed support material includes the following: S110. Obtain static load test datasets for the sprayed sample group and the untreated sample group, wherein the datasets include load, strain and acoustic emission signals collected through static load tests.

[0033] Reference Figure 2As shown, the first step is to process rock samples. Representative rock blocks are retrieved from the coal mine site and processed into standard cylindrical samples of various sizes according to different mechanical testing requirements. Representative rock blocks refer to rock samples that accurately reflect the actual geological and mechanical characteristics of the surrounding rock in the underground roadway. Specifically, these rock blocks should be collected from the main rock strata of the roof, floor, and sidewalls actually exposed in the roadway, covering major rock types such as sandstone, mudstone, and sandy mudstone, and including rock samples with different degrees of weathering and structural characteristics. In terms of geological characteristics, the samples should include both intact rock blocks and fractured rock blocks, preserving the original structural surfaces of the natural rock mass such as bedding and joints, to accurately reflect the strength range of the rocks in the mining area. The sampling locations should be spatially representative, systematically distributed along the roadway's strike and dip, covering areas of different burial depths and geological structures, including stress concentration zones and ordinary zones.

[0034] The specimen types include standard specimens for uniaxial compressive strength testing, triaxial compressive strength testing, tensile strength testing, and shear strength testing. The dimensions of the standard specimens can be set according to actual needs. As an example, the standard specimens for uniaxial and triaxial compressive strength testing are machined into standard cylindrical specimens with a height × diameter of 100 mm × 50 mm. This size specification can well simulate the stress state of actual surrounding rock under compression, providing a basis for accurately evaluating the compressive performance of surrounding rock under uniaxial and triaxial pressure. The standard specimens for tensile strength testing are machined into standard cylindrical specimens with a height × diameter of 25 mm × 50 mm. This size design meets the requirements of tensile strength testing, effectively measuring the mechanical properties of rock blocks under tension and reflecting the ability of surrounding rock to resist tensile failure. The standard specimens for shear strength testing are machined into standard cylindrical specimens with a height × diameter of 50 mm × 50 mm. This specification of specimen is suitable for shear strength testing, simulating the mechanical response of surrounding rock under different shear forces, providing data support for analyzing the shear failure mechanism of surrounding rock.

[0035] Due to their geological formation, natural rock blocks inevitably contain inherent structural defects such as fissures and cleavage. When these rock blocks are processed into standard test specimens, the number, size, and distribution of these internal defects exhibit randomness, directly leading to significant dispersion in the mechanical properties of the specimens. This inherent heterogeneity causes large fluctuations in test results, severely interfering with the objective evaluation of the true support effect of sprayed materials, and may even mask the performance differences between different sprayed materials.

[0036] To eliminate the interference of individual differences in rock samples on experimental results and ensure the reliability and comparability of static load tests, a sample homogenization screening step was introduced before the formal tests. Non-destructive testing techniques such as ultrasonic testing or CT three-dimensional crack scanning were used to assess the internal integrity of candidate samples. Ultrasonic testing indirectly reflects the internal density and defect situation of the sample by measuring the propagation speed of ultrasonic waves within the sample; CT three-dimensional crack scanning can visually present the three-dimensional distribution of cracks within the sample. Based on the assessment results, rock samples with excessively developed defects were eliminated to ensure good consistency in the physical and mechanical properties of the samples used. The screening method can be set according to actual needs. For example, a wave velocity threshold can be set, and samples whose wave velocity value deviates from the group average wave velocity by more than this threshold are eliminated. A wave velocity dispersion threshold can also be set to eliminate samples with dispersion exceeding this threshold. Based on the CT three-dimensional crack scanning results, a crack density threshold is set, and samples with crack densities exceeding this threshold are excluded. A maximum crack size threshold is set, and samples with a single crack exceeding the maximum crack size threshold are eliminated. A threshold for crack distribution uniformity is set, and samples with a crack distribution non-uniformity coefficient exceeding the threshold are discarded. Samples are selected according to the screening method, and the number of samples is determined based on the experimental design, but should not be less than 300 to meet the requirements of sufficient repeated testing and ensure the statistical reliability of the static load test data set.

[0037] Using the screened samples, sprayed rock samples were prepared by immersion method. Sprayed support material was uniformly applied to all surfaces of half of the samples to prepare sprayed sample groups. (Refer to...) Figure 3 As shown, in specific operations, the sample is completely immersed in the sprayed support material using a lifting grid, ensuring the material sufficiently covers the sample. The flowability and adhesion of the sprayed support material are monitored in real time to ensure consistent immersion time and uniform material application on each sample surface. This immersion method ensures the sprayed support material adheres evenly to all surfaces of the test sample, preventing uneven application from affecting the static load test results. The remaining half of the samples are left untreated as the untreated sample group, used for comparative analysis with the sprayed sample group to highlight the impact of the sprayed support material on the mechanical properties of the surrounding rock.

[0038] All treated samples (including sprayed sample groups and untreated sample groups) were placed in a curing chamber for constant temperature and humidity curing. The curing conditions were set at 90% humidity and 25℃ for 3 days. Curing the samples allowed the sprayed support material to fully solidify under suitable environmental conditions, fully utilizing its mechanical properties and making the static load test results more reflective of the application effect of the sprayed support material in actual engineering. Test samples were selected according to the sprayed sample group and the untreated sample group; specifically, three sprayed samples and three untreated samples were selected as one test group for static load testing.

[0039] The static loading test equipment was started, and loading tests were conducted on the sprayed sample group and the untreated sample group according to the set loading method and rate. During the loading process, load, strain, and acoustic emission signal data were collected in real time by sensors, and these data were transmitted to the data acquisition system for recording and storage to form a static load test dataset.

[0040] S120. Based on the static load test dataset, calculate a multi-dimensional evaluation index characterizing the synergistic control effect of the sprayed material and the surrounding rock.

[0041] Based on the acquired static load test dataset, a multidimensional feature vector is constructed, containing multiple evaluation indicators characterizing the synergistic control effect between the sprayed material and the surrounding rock. The specific process is as follows: a series of fundamental mechanical parameters are calculated from the raw data in the dataset, such as stress-strain curves and acoustic emission signals. These parameters include, but are not limited to: strength, various moduli, and post-peak parameters obtained from the stress-strain curves; cumulative ringing counts and energy obtained from the acoustic emission signals; and tensile, shear, and triaxial strength obtained from other tests. Representative values ​​within each fundamental parameter are calculated for both the sprayed and untreated sample groups. By comparing the representative values ​​of the two groups, a series of evaluation parameters directly characterizing performance changes are generated, such as strength ratio, modulus change rate, and energy release change rate. These evaluation parameters are combined to form a multidimensional feature vector for subsequent comprehensive evaluation.

[0042] S130. Input the multidimensional evaluation index into the trained neural network model to obtain a comprehensive judgment result on the control effect of the sprayed support material on the surrounding rock; wherein, the neural network model outputs the comprehensive judgment result on the control effect of the sprayed support material on the surrounding rock by verifying the physical consistency of the changes between different multidimensional evaluation indices.

[0043] Extensive experimental data was collected, including multidimensional evaluation index data of different types of sprayed support materials under various surrounding rock conditions, and corresponding evaluation results of actual surrounding rock control effects, such as excellent, good, average, and poor. This data was divided into training, validation, and test sets. A neural network model structure was selected, such as a multilayer perceptron, convolutional neural network, or recurrent neural network. The number of nodes in the input, hidden, and output layers of the model, as well as the connection methods and activation functions between layers, were determined. The neural network model was trained using the training set data, and the model parameters were continuously adjusted using the backpropagation algorithm to minimize the error between the model's output and the actual evaluation results. During training, the model was validated using the validation set data, and the model's hyperparameters, such as the learning rate and the number of training iterations, were adjusted based on the validation results to prevent overfitting. Finally, the trained model was tested using the test set data to evaluate its generalization ability. A physical consistency verification mechanism was introduced during model training. For example, based on relevant theories of materials mechanics and rock mechanics, the relationship between different multidimensional evaluation indices was analyzed to determine whether it conformed to physical laws. If the model outputs results that violate these physical laws, the model should be adjusted and corrected to ensure that the overall judgment result output by the model has physical rationality.

[0044] The calculated multidimensional evaluation indicators are input into a trained and validated neural network model. Based on the input multidimensional indicator data and the complex nonlinear relationships learned internally, the model outputs a comprehensive judgment result on the rock control effect of the sprayed support material, such as giving a specific score or qualitative evaluation (e.g., "the sprayed support material has a good rock control effect"), providing a scientific basis for the selection and application of sprayed support materials in coal mines.

[0045] The specific evaluation process includes: The Z-Score method is used to standardize the multidimensional evaluation indicators to eliminate the influence of dimensions. The standardized feature vectors are input into a neural network model, and the model outputs a probability distribution vector. Based on the probability distribution vector p, the level of rock control effect of the sprayed support material is determined. If max(p) = p I And p I >θ, classified as Level I (excellent control effect); If max(p) = p II It was determined to be Level II (good control effect); If max(p) = p III It is classified as Level III (moderate control effect). If max(p) = p IV It is classified as Level IV (poor control effect).

[0046] Where, pI p II p III p IV These correspond to the predicted probabilities of Level I, Level II, Level III, and Level IV, respectively, with θ being the confidence threshold, such as 0.8.

[0047] Existing technologies primarily focus on testing the physical and mechanical properties of the sprayed support materials themselves, failing to reflect the actual support effect after interaction with the surrounding rock. This invention, however, obtains static load test datasets for sprayed and untreated sample groups. These datasets include load, strain, and acoustic emission signals collected through static load tests. By setting up a comparison between the sprayed and untreated sample groups and incorporating various key data (i.e., load, strain, and acoustic emission signals) generated during interaction with the surrounding rock, the actual performance data of the sprayed support material under interaction with the surrounding rock can be directly obtained. This overcomes the shortcomings of existing technologies that only test the material itself, providing data closer to actual working conditions for subsequent evaluation. Based on the obtained static load test dataset, this invention calculates multi-dimensional evaluation indicators characterizing the synergistic control effect between the sprayed material and the surrounding rock. These multi-dimensional evaluation indicators comprehensively consider various factors of the sprayed support material and the surrounding rock under static loading conditions, comprehensively reflecting the control effect under their synergistic action from multiple perspectives. Compared to existing technologies that only focus on the material's inherent performance, this more scientifically and accurately reflects the actual support effect. This invention inputs the calculated multidimensional evaluation indicators into a trained neural network model to obtain a comprehensive judgment result on the rock control effect of sprayed support materials. The neural network model outputs results by verifying the physical consistency of changes among different multidimensional evaluation indicators. This evaluation method based on a neural network model can fully utilize the complex relationships between multidimensional evaluation indicators, simulate the influence of multiple factors interacting on the rock control effect under actual working conditions, avoid the one-sidedness of single-indicator evaluation, and realize the scientific evaluation of the rock control effect of different sprayed support materials under static loading conditions. It effectively solves the problem that existing technologies cannot scientifically evaluate the actual support effect.

[0048] In an optional embodiment, the calculation of multidimensional evaluation indicators characterizing the synergistic control effect of the sprayed material and the surrounding rock based on the static load test dataset described in S120 above includes: S1201. Extract the raw data for calculating various evaluation indicators from the static load test dataset; the raw data includes: stress-strain curve feature point data, acoustic emission signal cumulative parameters, test peak intensity, and related geometric parameters.

[0049] Key characteristic parameters obtained from stress-strain curves include peak stress, elastic stage start and end points, and residual strength points. Cumulative parameters obtained from acoustic emission signals include time-domain characteristic parameters such as cumulative ring count and cumulative energy release. Test peak strengths obtained from specialized tests such as shear strength, point load strength, and triaxial compression include ultimate bearing parameters such as shear strength, point load strength, and triaxial compressive strength; relevant geometric parameters include geometric characteristic parameters such as specimen size and loading area.

[0050] S1202. Based on the original data, determine the parameters used for inter-group comparison of the sprayed sample group and the untreated sample group for each of the evaluation indicators.

[0051] Based on the obtained raw data, parameters for inter-group comparison were determined for each evaluation index for the sprayed sample group and the untreated sample group. For continuous variables such as uniaxial compressive strength and elastic modulus, the arithmetic mean and standard deviation of each group were calculated; for acoustic emission characteristic parameters, the cumulative statistics of each group were calculated; and a statistical model based on probability distribution was established for the strength index. To ensure statistical significance, the number of samples in each group was no less than 30.

[0052] S1203. Based on the parameters applied to the inter-group comparison of the sprayed sample group and the untreated sample group, determine the evaluation value of each evaluation index.

[0053] Based on the parameters used for inter-group comparison between the sprayed sample group and the untreated sample group, the evaluation value of each evaluation index is calculated. Appropriate calculation methods are used for different types of indices. For indicators reflecting the material's load-bearing capacity, such as uniaxial compressive strength, triaxial compressive strength, tensile strength, shear strength, point load strength, and cohesion, relative change rates are used for quantitative calculation. The specific calculation formula is: Evaluation value = (Average value of sprayed group - Average value of untreated group) / Average value of untreated group × 100%. This method directly reflects the extent to which the sprayed material improves the strength characteristics of the rock mass.

[0054] For indicators reflecting a material's resistance to deformation, such as elastic modulus, secant modulus, initial deformation modulus, and triaxial compressive modulus, a standardized scoring method is used for quantification. The specific calculation formula is: Evaluation value = 100 × (average value of the sprayed group - minimum value) / (maximum value - minimum value), where the maximum and minimum values ​​are taken from the extreme range of all test data. This method effectively eliminates the influence of dimensions and facilitates comparative analysis between different stiffness indicators.

[0055] For indicators reflecting the energy accumulation and release characteristics of materials, such as cumulative acoustic emission ring count, cumulative acoustic emission energy release, and brittleness index, a significance test method based on logarithmic transformation is adopted. First, the original data are logarithmically transformed to meet the normal distribution requirement. Then, a t-test is used to analyze the significant differences between the two sets of data. Finally, the test statistic is transformed into a standardized evaluation value. This method fully considers the statistical characteristics of acoustic emission data, ensuring the reliability of the evaluation results.

[0056] For indicators reflecting data fluctuation characteristics, such as the performance dispersion improvement coefficient, the coefficient of variation ratio is used for quantification. The specific calculation method is as follows: first, calculate the coefficient of variation (standard deviation / mean) for the sprayed group and the untreated group respectively; then, calculate the ratio of the two coefficients of variation. This indicator can objectively evaluate the improvement effect of the sprayed material on the stability of rock mass mechanical properties.

[0057] S1204. The calculated evaluation values ​​are combined to form the multidimensional evaluation index.

[0058] This invention extracts raw data from stress-strain curves, acoustic emission signals, and mechanical tests to establish a multi-dimensional evaluation index system encompassing strength, deformation, and damage characteristics, enabling a comprehensive quantitative assessment of the rock-control effect. The uniaxial compressive strength index characterizes the degree of improvement in the rock-bearing capacity, while the cumulative acoustic emission energy release index reflects the suppression effect on internal damage. The comprehensive analysis of multi-dimensional indicators overcomes the limitations of single evaluation indicators, ensuring the accuracy and completeness of the assessment results. Using an inter-group comparative analysis method, the quantitative differences in various evaluation indicators between the sprayed sample group and the untreated sample group are calculated to establish an evaluation benchmark for the rock-control effect of the sprayed support material. A significant improvement in shear strength reflects the material's enhancement of the rock's shear capacity, while changes in point load strength reflect the material's improvement in local bearing capacity. This comparative quantitative method forms an objective basis for evaluating the rock-control effect.

[0059] Analysis of characteristic point data from stress-strain curves provides a basis for evaluating the effect of sprayed support materials on surrounding rock deformation control. Changes in the elastic modulus reflect the material's adjustment of surrounding rock stiffness, while changes in peak strain reflect the material's control over the deformation development process. Analysis of cumulative acoustic emission signal parameters yields evaluation parameters for the sprayed support material's damage control effect on surrounding rock. Changes in the cumulative acoustic emission ringing count reflect the material's control over microcrack initiation and propagation, while changes in acoustic emission energy release characteristics reflect the material's dissipation of damage evolution energy. Based on comprehensive analysis of specialized mechanical test data, an evaluation method for the collaborative bearing mechanism between sprayed materials and surrounding rock is established. Changes in shear strength reflect the material's improved resistance to shear failure, while the development law of triaxial compressive strength reflects the material's collaborative bearing effect under complex stress conditions. These analytical results collectively constitute an evaluation system for the surrounding rock control effect of sprayed support materials.

[0060] In an optional embodiment, the static load test includes a uniaxial compression test, a uniaxial tensile test, a shear strength test, a point load strength test, and a triaxial compression test; The acoustic emission signal was acquired synchronously during the uniaxial compression test.

[0061] Three coated samples and three untreated samples were selected as one test group, and the experiment was divided into 10 test groups, each of which underwent uniaxial compression testing. Acoustic emission monitoring was also performed during the testing process to fully record the acoustic emission signals during the loading process.

[0062] Three sprayed samples and three untreated samples were selected as one test group. The experiment was divided into 10 test groups, and uniaxial tensile tests were carried out in each group.

[0063] Three sprayed samples and three untreated samples were selected as one test group. The experiment was divided into 10 test groups, and shear strength tests were carried out in each group.

[0064] Three sprayed samples and three untreated samples were selected as one test group. The experiment was divided into 10 test groups, and point load strength tests were performed on each group.

[0065] Three sprayed samples and three untreated samples were selected as one test group, and the experiment was divided into 10 test groups, each of which underwent triaxial compression testing. The confining pressure in the triaxial compression test can be determined according to the engineering background or actual experimental needs, such as 5MPa, 10MPa, or 15MPa.

[0066] All test processes record the load and strain magnitudes throughout the entire loading process to ensure data integrity and accuracy.

[0067] This invention utilizes various static load tests, including uniaxial compression tests, uniaxial tensile tests, point load strength tests, triaxial compression tests, and shear strength tests, to comprehensively evaluate the improvement effect of sprayed support materials on the mechanical properties of surrounding rock from different angles and under different stress states. For example, uniaxial compression tests can reflect the effect of sprayed materials on improving the compressive strength of surrounding rock; uniaxial tensile tests can assess the effect of sprayed materials on enhancing the tensile properties of surrounding rock; triaxial compression tests can study the support effect of sprayed materials on surrounding rock under different confining pressures; and shear strength tests can analyze the influence of sprayed materials on the shear properties of surrounding rock. By combining the results of these tests, the applicability and effectiveness of sprayed support materials in practical engineering can be evaluated more accurately.

[0068] Synchronous acquisition of acoustic emission signals during uniaxial compression tests provides an effective means of analyzing the internal damage evolution process of specimens. Acoustic emission is an elastic wave generated by the release of strain energy due to the propagation of internal defects during the stress process of a material. By acquiring and analyzing acoustic emission signals, the development process of internal damage in rocks can be monitored in real time. Changes in the number of acoustic emission events reflect the degree of development of internal microcracks, while changes in acoustic emission energy characterize the severity of crack propagation. Compared with traditional methods that analyze the mechanical properties of specimens solely through macroscopic parameters such as load-displacement curves, acoustic emission technology can provide more detailed information on internal damage, helping to accurately reveal the failure mechanism of specimens and the mechanism of action of sprayed support materials.

[0069] In an optional embodiment, the multidimensional evaluation index includes multiple parameters such as uniaxial compressive strength, uniaxial compressive modulus, uniaxial compressive Poisson's ratio, uniaxial secant modulus, uniaxial initial deformation modulus, uniaxial post-peak residual strength, uniaxial failure strain, brittleness index, cumulative acoustic emission ringing count, cumulative acoustic emission energy release, uniaxial tensile strength, shear strength, point load strength, triaxial compressive strength, cohesion, internal friction angle, triaxial compressive modulus, triaxial post-peak residual strength, triaxial failure strain, and performance dispersion improvement coefficient.

[0070] This invention systematically reflects the degree to which the sprayed material improves the strength characteristics of rock mass under different stress states through seven strength variation indicators: uniaxial compressive strength variation rate, tensile enhancement index, shear strength variation rate, point load strength variation rate, triaxial compressive strength variation rate, cohesion variation index, and internal friction angle variation rate, providing a complete strength gain parameter system for support design. Based on five stiffness variation indicators—uniaxial compressive elastic modulus variation rate, uniaxial compressive Poisson's ratio variation rate, uniaxial secant modulus variation rate, uniaxial initial deformation index, and triaxial compressive elastic modulus variation rate—the invention accurately characterizes the quantitative degree of the sprayed material's effect on controlling the deformation behavior of rock mass, revealing the stiffness improvement law of the material at different deformation stages. Through four post-peak characteristic variation indicators—uniaxial post-peak residual strength variation rate, uniaxial failure strain variation rate, triaxial residual strength variation rate, and triaxial failure strain variation rate—the invention comprehensively quantifies the improvement effect of the rock mass's mechanical behavior from peak strength to residual strength, accurately evaluating the control effectiveness of the sprayed material on the rock mass failure process. By utilizing two acoustic emission characteristic change parameters—the fracture suppression index and the acoustic emission energy release change rate—a quantitative assessment of the suppression effect on the damage evolution process within the rock mass was achieved, providing direct quantitative evidence for revealing the reinforcement mechanism of sprayed materials. Through comprehensive analysis of the brittleness index change rate and the performance dispersion improvement coefficient, the degree of improvement of the brittleness characteristics of the sprayed material on the rock mass and its effect on improving mechanical property stability were objectively evaluated. This multidimensional evaluation index system, with the change rate as its core, covers key aspects such as strength gain, stiffness improvement, post-peak characteristic control, damage suppression, and performance stability improvement. It establishes a systematic quantitative evaluation method from macroscopic mechanical behavior improvement to microscopic damage mechanism regulation, providing complete quantitative technical support for the performance optimization and engineering application of coal mine sprayed support materials, and establishing a systematic evaluation standard from changes in basic material mechanical properties to the synergistic control effect on surrounding rock.

[0071] Each evaluation indicator was calculated in the following way: The ratio of the uniaxial compressive strength of the sprayed sample group to that of the untreated sample group is calculated, and an exponential function is used to map this ratio to a percentage score to obtain the evaluation value of the uniaxial compressive strength change rate. The evaluation value of the uniaxial compressive elastic modulus change rate is obtained by calculating the rate of change of elastic modulus of the sprayed sample group relative to the untreated sample group. The evaluation value of the uniaxial compression Poisson's ratio change rate is obtained by calculating the change rate of the Poisson's ratio of the sprayed sample group relative to the untreated sample group. The evaluation value of the uniaxial secant modulus change rate is obtained by mapping the ratio of the secant modulus of the sprayed sample group to the untreated sample group to a percentage score. The evaluation value of the uniaxial initial deformation index is obtained by mapping the ratio of the initial deformation modulus of the sprayed sample group to that of the untreated sample group to a percentage score. The evaluation value of the uniaxial post-peak residual strength change rate is obtained by calculating the post-peak residual strength change rate of the sprayed sample group relative to the untreated sample group. The evaluation value of the uniaxial failure strain change rate is obtained by calculating the failure strain change rate of the sprayed sample group relative to the untreated sample group. The evaluation value of the brittleness index change rate is obtained by calculating the rate of change of the brittleness index of the sprayed sample group relative to the untreated sample group. The evaluation value of the fracture suppression index is obtained by mapping the ratio of acoustic emission ring counts of the sprayed sample group to the untreated sample group to a percentage score. An evaluation value for the acoustic emission energy release change rate is obtained by calculating the cumulative change rate of acoustic emission energy release of the sprayed sample group relative to the untreated sample group. The evaluation value of the tensile strengthening index is obtained by calculating the ratio of the uniaxial tensile strength of the sprayed sample group to that of the untreated sample group, and mapping the ratio to a percentage score using an exponential function. The evaluation value of the shear strength change rate is obtained by calculating the rate of change of the shear strength of the sprayed sample group relative to the untreated sample group. The evaluation value of the point load strength change rate is obtained by calculating the point load strength change rate of the sprayed sample group relative to the untreated sample group; The evaluation value of the triaxial compressive strength change rate is obtained by calculating the triaxial compressive strength change rate of the sprayed sample group relative to the untreated sample group. The cohesion ratio of the sprayed sample group to the untreated sample group is calculated, and an exponential function is used to map the ratio to a percentage score to obtain the evaluation value of the cohesion change index. An evaluation value for the rate of change of the internal friction angle is obtained by calculating the rate of change of the internal friction angle of the sprayed sample group relative to the untreated sample group. The evaluation value of the triaxial compressive modulus change rate is obtained by calculating the triaxial elastic modulus change rate of the sprayed sample group relative to the untreated sample group. The evaluation value of the triaxial residual strength change rate is obtained by calculating the rate of change of residual strength after the triaxial peak of the sprayed sample group relative to the untreated sample group. The evaluation value of the triaxial failure strain change rate is obtained by calculating the triaxial failure strain change rate of the sprayed sample group relative to the untreated sample group. The evaluation value of the performance dispersion improvement coefficient is obtained by comparing the coefficient of variation of uniaxial compressive strength between the sprayed sample group and the untreated sample group, and by combining the mean change.

[0072] The following explains the calculation process of the evaluation values ​​of each evaluation index under five static load tests: uniaxial compression test, uniaxial tensile test, shear strength test, point load strength test, and triaxial compression test.

[0073] 1. The data post-processing and evaluation index selection methods for uniaxial compression tests are as follows: After the test is completed, plot the axial stress-axial strain curve, transverse strain-axial strain curve, acoustic emission count, and energy diagram for each specimen.

[0074] (1) Uniaxial compressive strength UCS is the maximum compressive stress (σ) that a rock can withstand under uniaxial compressive load. ucs Uniaxial compressive strength is the most intuitive indicator for evaluating the coating effect. Its value is the maximum axial stress value on the axial stress-axial strain curve, and its value is as follows: Figure 4 σ ucs As shown. For each test group, UCS uses the average value of the samples under the same conditions (sprayed or untreated) within that test group. Based on the strength change of the sprayed sample compared to the untreated sample, the evaluation level is divided into five levels according to the rate of change of uniaxial compressive strength: (1) (2) In the formula, R ucs S is the uniaxial compressive strength ratio index. ucs This represents the uniaxial compressive strength change rate. It should be noted that parameters with subscripts for sprayed samples in the formula of this invention represent the average value of that parameter for all sprayed samples within the test group; parameters with subscripts for untreated samples represent the average value of that parameter for all sprayed samples within the test group. For example, UCS... 喷涂试样 The average uniaxial compressive strength of all sprayed specimens in the test group, UCS 未处理试样 This represents the average uniaxial compressive strength of all untreated specimens within the test group.

[0075] Table 1

[0076] (2) The elastic modulus E is the ability of rock to resist deformation during the elastic deformation stage. The elastic modulus characterizes the stiffness of the rock sample and is an indicator for evaluating the anti-deformation effect of spraying. The value of the elastic modulus corresponds to the slope of the elastic stage of the axial stress-strain curve, and its value is 0.5σ corresponding to the axial stress-strain curve of the rock. cd The slope calculated from linear fitting in the ±0.5 MPa segment; for linear fitting, R0 is required. 2 >0.9, if R 2 If the requirements are not met, the linear fitting range can be increased by 0.5 MPa both above and below it until R0 is reached. 2 >0.9. (For example) Figure 4 As shown in Figure E. The elastic modulus of this method is taken as the average value of samples under the same conditions in each test group. Based on the change in elastic modulus between sprayed and untreated samples, and according to the uniaxial compressive elastic modulus change rate η... E Formula (3) divides the evaluation level into three levels: (3) Table 2

[0077] (3) Poisson's ratio Poisson's ratio is the ratio of transverse normal strain to axial normal strain when rock is subjected to uniaxial tension or compression. It describes the degree of transverse strain in the sprayed sample and can be included in the evaluation index. In this method, 0.5σ is used. ucs The ratio of transverse strain to axial strain corresponding to the ±0.5 MPa range. Poisson's ratio is taken as the average value of specimens under the same conditions in each test group, based on the uniaxial compression Poisson's ratio change rate η. v Formula (4) divides the evaluation level into three levels: (4) Table 3

[0078] (4) The secant modulus Es is the slope of the line connecting a point on the uniaxial compressive axial stress-strain curve and the origin of the coordinate system. It reflects the overall deformation characteristics of the sprayed rock sample and can be included in the evaluation index. In this method, 0.5σ on the curve is taken. ucs The slope of the line connecting the corresponding point to the origin is the value of the secant modulus, such as... Figure 4 As shown in Es. The secant modulus is taken as the average value of the samples under the same conditions in each test group, based on the uniaxial secant modulus change rate S. Es Formula (5) has four evaluation levels: (5) Table 4

[0079] (5) Initial deformation modulus E i The slope of the tangent line at the origin of the coordinate system for uniaxial compressive axial stress and strain in rock reflects the number of microcracks in the sprayed rock sample, thus characterizing the initial density of the sample after spraying. Therefore, it can be included in the evaluation index. The initial deformation modulus is taken as follows: Figure 4 China E i As shown. The initial deformation modulus is taken as the average value of specimens under the same conditions in each test group, based on the uniaxial initial deformation index S. Ei Formula (6) has four evaluation levels: (6) Table 5

[0080] (6) After the rock reaches its uniaxial compressive strength, its internal structure is destroyed, and the load-bearing capacity of the specimen decreases with increasing strain, eventually stabilizing at a certain strength value, which is called the post-peak residual strength σ. r It can characterize the residual bearing capacity of rocks and can be included in the evaluation index. The post-peak residual strength value is as follows: Figure 4 σ r As shown. The residual strength is the average value of specimens under the same conditions in each test group. The residual strength is calculated based on the rate of change of residual strength after the uniaxial peak. Formula (7) divides the evaluation level into three levels: (7) Table 6

[0081] (7) Damage strain ε r This is the strain corresponding to the residual strength of rock after uniaxial compression reaches its peak. The larger the failure strain, the stronger the rock's toughness, and it can be included in the evaluation index. The failure strain value is as follows: Figure 4 ε r As shown, the failure strain is taken as the average value of specimens under the same conditions in each test group. This is based on the uniaxial failure strain change rate. Formula (8) divides the evaluation level into three levels: (8) Table 7

[0082] (8) The brittleness index (BI) is used to evaluate the tendency of rocks to undergo brittle fracture (severe fracture) under stress. The higher the brittleness index, the more easily the rock will undergo brittle fracture. It can evaluate the control effect of sprayed materials on severe rock fracture and can be included in the evaluation index. In this method, the brittleness index (BI) is calculated by formula (9). The brittleness index is taken as the average value of the samples under the same conditions in each test group, and the brittleness index is calculated based on the rate of change of the brittleness index. Formula (10) divides the evaluation level into three levels: (9) In the formula: BI is the brittleness index; E is the rock elastic modulus; G is the rock shear modulus. The rock shear modulus can be calculated from the rock elastic modulus E and Poisson's ratio. The calculated relationship is as follows: .

[0083] (10) Table 8

[0084] (9) Acoustic emission ringing count and energy characteristic map are plotted using acoustic emission data monitored during the test. The cumulative acoustic emission ringing count can reflect the degree of internal fracture damage during rock loading and can be included in the evaluation index. (Accumulated acoustic emission ringing count) The average value of samples under the same conditions in each test group is taken, and the results are calculated based on the cumulative count of acoustic emission ringing and the fracture suppression index. Formula (11) divides the evaluation level into four levels: (11) Table 9

[0085] (10) The rock loading and fracturing process is accompanied by energy release, and the cumulative release characteristics of acoustic emission energy can reflect the scale of rock fracturing. Therefore, the cumulative release characteristics of acoustic emission energy are included in the evaluation index. (Accumulated release of acoustic emission energy) Take the average value of the samples under the same conditions in each test group, and calculate based on the rate of change of acoustic emission energy release. Formula (12) divides the evaluation level into three levels: (12) Table 10

[0086] 2. The data post-processing and evaluation index selection methods for uniaxial tensile tests are as follows: The uniaxial tensile strength test was conducted using the Brazilian splitting method. After the test, the failure load was obtained, and the uniaxial tensile strength was calculated using formula (13). (13) In the formula: R t t is the uniaxial compressive strength, MPa; P is the maximum load at rock failure, N; t is the rock specimen thickness, mm; D is the rock specimen diameter, mm.

[0087] (11) Uniaxial tensile strength R t Tensile strength is the maximum tensile stress that a rock can withstand under uniaxial tensile stress. It can be used to evaluate the enhancing effect of the sprayed material on the tensile properties of the rock and can be included in the evaluation index. The tensile strength is taken as the average value of the samples under the same conditions in each test group, and its value is [value missing]. Figure 5 As shown. Based on the uniaxial tensile strength enhancement index S... Rt The evaluation levels are divided into four levels: (14) (15) In the formula, R Rt S is the tensile strength index, representing the uniaxial tensile strength ratio.Rt This is a score for uniaxial tensile strength.

[0088] Table 11

[0089] 3. The data post-processing and evaluation index selection methods for shear strength tests are as follows: (12) Shear strength It can reflect the rock's ability to resist shear failure and can be used to evaluate the enhancing effect of sprayed materials on the rock's shear resistance under shear loads, thus it can be included in the evaluation index. The shear strength is taken as the average value of samples from each test group under the same conditions, based on the shear strength variation rate η. τ The evaluation levels are divided into three levels: (16) Table 12

[0090] 4. The data post-processing and evaluation index selection methods for point load strength tests are as follows: (13) Point load strength It can reflect the ability of rock to resist failure under localized concentrated loads and can be used to evaluate the enhancing effect of sprayed materials on the local strength of rock under concentrated loads, thus it can be included in the evaluation index. The point load strength is taken as the average value of samples under the same conditions in each test group, and the point load strength variation rate η... Is The evaluation level is divided into three levels: (17) Table 13

[0091] 5. The data post-processing and evaluation index selection methods for triaxial compression tests are as follows: The evaluation metrics selected for triaxial compression are similar to those for uniaxial compression. The evaluation metrics and grading are as follows: (14) Triaxial compressive strength (TCS) is the maximum compressive stress (σ) that a rock can withstand under triaxial compressive load. tcs Therefore, triaxial compressive strength is the most intuitive indicator for evaluating the coating effect under confining pressure. Its value is the maximum axial stress value on the axial stress-strain curve under confining pressure, such as... Figure 6 As shown. TCS uses the average value of samples under the same conditions in each test group, based on the strength change of the sprayed sample compared to the untreated sample, and according to the triaxial compressive strength change rate η. tcs Formula (18) divides the evaluation level into three levels: (18) Table 14

[0092] Table 15

[0093] (15) Cohesion Cohesion is one of the important indicators of rock shear strength. From a microscopic perspective, cohesion is defined as the attractive force between adjacent parts of a rock. Incorporating cohesion into the evaluation index aims to evaluate the enhancing effect of sprayed materials on the "bond strength" of rock. In this method, the cohesion is taken as the average value of samples under the same conditions in each test group, and the cohesion variation index is used. Formula (20) divides the evaluation level into four levels: (19) (20) In the formula, S c The cohesion score is given.

[0094] Table 16

[0095] (16) Angle of internal friction Internal friction angle is one of the important indicators of rock friction characteristics, referring to the relationship between frictional force and directional pressure during the sliding of particles in the rock shear failure process. In this method, the internal friction angle is taken as the average value of samples under the same conditions in each test group, based on the rate of change of internal friction angle η. θ Formula (21) divides the evaluation level into three levels: (twenty one) Table 17

[0096] (17) Triaxial compressive modulus The elastic modulus represents the rock's ability to resist deformation during the elastic deformation stage. It characterizes the stiffness of a rock sample and is an indicator for evaluating the anti-deformation effect of spraying. The value of the elastic modulus corresponds to the slope of the axial stress-strain curve in the elastic stage, and its value is 0.5σ of the rock's axial stress-strain curve. ucs The slope of the curve in the ±0.5 MPa segment, such as Figure 6 As shown. The elastic modulus of this method is taken as the average value of samples under the same conditions in each test group. Based on the change in elastic modulus between sprayed and untreated samples, and according to the triaxial compressive elastic modulus change rate η... E’ Formula (22) divides the evaluation level into three levels: (twenty two) Table 18

[0097] (18) After the rock reaches its triaxial compressive strength, its internal structure is destroyed, and the load-bearing capacity of the specimen decreases with increasing strain, eventually stabilizing at a certain strength value, which is called the residual strength after the triaxial peak. It can characterize the residual bearing capacity of rocks and can be included in the evaluation index. The residual strength after the triaxial peak is taken as follows: Figure 6 As shown. The residual strength after the triaxial peak is taken as the average value of the specimens under the same conditions in each test group. Based on the triaxial residual strength variation rate... Formula (23) divides the evaluation level into three levels: (twenty three) Table 19

[0098] (19) Triaxial failure strain ε tr This is the strain corresponding to the residual strength after the rock reaches its peak under triaxial compression. The larger the peak strain, the stronger the rock's toughness, and it can be included in the evaluation index. The triaxial failure strain value is as follows: Figure 6 As shown, the failure strain is taken as the average value of specimens under the same conditions in each test group. This is based on the triaxial failure strain rate η. tr Formula (24) divides the evaluation level into three levels: (twenty four) Table 20

[0099] (20) To evaluate the effect of the spraying material on improving the consistency of mechanical properties of rock samples, a performance dispersion improvement coefficient S is introduced. Cv This index assesses the material's ability to control performance dispersion by comparing the coefficients of variation between different samples in the sprayed group and between different samples in the untreated group. It evaluates whether the sprayed material can reduce the fluctuations in the mechanical properties of rock samples and improve the reliability and consistency of the support effect. This index uses the uniaxial compressive strength σ, which most directly reflects the mechanical properties of the sprayed material, as its most direct indicator. ucs As the basis for discreteness analysis, the calculation method is as follows: First, calculate the standard deviation between different samples in the sprayed group and between different samples in the untreated group. The mean μ, standard deviation and mean can be calculated by formula (25): (25) Where: σ ucsi Let be the uniaxial compressive strength of the i-th sample. Then, the coefficient of variation Cv between different samples in the sprayed group and between different samples in the untreated group is calculated using formula (26): (26) Finally, the performance dispersion improvement coefficient S is calculated using formula (27). Cv : (27) The performance dispersion improvement coefficient S Cv The evaluation level is divided into five levels: Table 21

[0100] In an optional embodiment, this method proposes a deep learning evaluation system based on physical constraints and data augmentation (Physics-Constrained Deep Neural Network with VAE Augmentation, PC-DNA). This method aims to address the problem of limited sample size in rock mechanics experiments, which makes it difficult to meet the training requirements of traditional deep learning, while ensuring that the evaluation results conform to the fundamental principles of rock mechanics through a physical constraint mechanism. (Refer to...) Figure 7 As shown, the trained neural network model is obtained through training in the following manner: S1301. Obtain the training dataset, which contains multiple sets of sample data. Each set of sample data includes a multi-dimensional evaluation index calculated from the mechanical response data of the sprayed sample and the untreated sample, as well as the corresponding benchmark evaluation level label.

[0101] Obtain a training dataset containing multiple sets of sample data. Each set of sample data includes input features and corresponding benchmark evaluation level labels. The input features are 20 evaluation indicators calculated based on the mechanical response data of sprayed and untreated samples. A structured feature vector x∈R is constructed based on each evaluation indicator. 20 The feature vector contains evaluation index data in 20 dimensions: uniaxial compressive strength change rate S UCS 1. Uniaxial elastic modulus change rate η E 1 / 2 uniaxial compression Poisson's ratio change rate η v , uniaxial secant modulus change rate S Es Uniaxial initial deformation index S Ei η, the rate of change of residual intensity after uniaxial peak σr Uniaxial failure strain rate η εr , Brittleness index change rate η BI rupture inhibition index η AEcount η, the rate of change of acoustic emission energy release AEenergy Tensile strength index S Rt η, shear strength change rate τ Point load strength change rate η Is Triaxial compressive strength variation rate η tcs Cohesion variation index S Cη, the rate of change of internal friction angle θ Triaxial compressive modulus change rate η E’ Triaxial residual strength variation rate η σtr Triaxial failure strain rate η εtr Performance dispersion improvement coefficient S Cv It also includes the corresponding benchmark evaluation level labels: Level I, Level II, Level III, and Level IV.

[0102] The Z-Score standardization method is used to process the feature vectors, eliminating the impact of data differences on the neural network weight updates, resulting in standardized feature vectors. For the observed value x of the j-th evaluation index... j The standardized value x j ∗ (Right now Figure 7 The standardized score calculation formula is as follows: (28) Where, μ j Let σ be the mean of this evaluation index across all samples. j This represents the standard deviation of the evaluation index. After standardization, all input features are mapped to a distribution space with a mean of 0 and a variance of 1.

[0103] Based on each individual evaluation indicator, the initial comprehensive score S is calculated using the weighted average formula. total The baseline evaluation level label is obtained by directly assigning a label based on the score's range. The calculation formula is as follows: (29) Among them, U1,U 2, U 3, U 4, U5 represents the combined score for the uniaxial compression test, uniaxial tensile test, shear strength test, point load strength test, and triaxial compression test, respectively. 2, w 3, w 4, w5 represents the weights corresponding to the split Hopkinson bar test and the dynamic-static combined triaxial compression test, respectively. In this embodiment, the weights are set as follows: w1=0.5, w2=0.5, w1=0.3, w2=0.14, w3=0.14, w4=0.12, w5=0.3.

[0104] The scores for each dimension are calculated as a weighted sum of its internal individual evaluation indicators. The overall score U1 for the uniaxial compression test is calculated using the following formula: (30) The overall score U2 of the uniaxial tensile test is calculated using the following formula: (31) The overall score U3 of the shear strength test is calculated using the following formula: (32) The overall score U4 of the point load strength test is calculated using the following formula: (33) The overall score U5 of the triaxial compression test is calculated using the following formula: (34) S1302. The training dataset is augmented using a variational autoencoder model to generate augmented samples for model training, resulting in an augmented dataset.

[0105] To address the problem of insufficient physical sample quantity in rock mechanics experiments, this invention constructs a Variational Autoencoder (VAE) model for probabilistic modeling and expansion. The VAE model comprises an encoder network and a decoder network. The encoder network receives the standardized feature vectors and maps them to Gaussian distribution parameters in a low-dimensional latent space, namely the mean vector μ and the log-variance vector logσ. 2 σ represents the standard deviation. A reparameterization technique is used to sample in the latent space, introducing standard normal distribution noise ε ~ N(0, I) to generate a latent vector z = μ + σ⊙ε, simulating the uncertainty and randomness of the internal microstructure of natural rock materials. I represents the identity matrix. The decoder network maps the sampled latent vector z back to the original feature space, generating virtual experimental data. By interpolating and adding small perturbations in the latent space, a large number of virtual samples conforming to the physical distribution of the original data are generated, which, together with the real samples, constitute the augmented training dataset D. train .

[0106] S1303. Using the evaluation metrics in the augmented dataset as input features and the corresponding benchmark evaluation level labels as training targets, the deep neural network model is trained by minimizing the composite loss function, and the trained deep neural network model is used as the trained neural network model.

[0107] The composite loss function is determined based on the classification error loss and the physical consistency constraint loss. The classification error loss is determined based on the predicted probability distribution of the model output and the true benchmark rating label. The physical consistency constraint loss is determined based on the predicted score relationship of sample pairs in the training batch. The sample pairs are selected based on the physical performance relationship of the input features of the sample pairs, and the sample with better physical performance should have a predicted score no lower than that of the sample with worse physical performance.

[0108] Specifically, the trained neural network model in this invention is obtained by training a deep neural network model. The deep neural network model specifically adopts a Deep Fully Connected Network (DFCN) architecture. First, a deep fully connected neural network is constructed to establish a nonlinear mapping model from multidimensional physical indicators to support effect evaluation levels. The deep fully connected neural network includes an input layer, hidden layers, and an output layer. The input layer receives a 20-dimensional standardized feature vector. The hidden layer consists of four fully connected layers, with neurons arranged in an inverted pyramid structure (64→32→16→8) to progressively compress feature dimensions and extract higher-order abstract features. The output layer contains four neurons, corresponding to four evaluation levels: Level I (Excellent), Level II (Good), Level III (Medium), and Level IV (Poor). The hidden layers uniformly use the LeakyReLU activation function (α=0.01) to enhance the model's nonlinear expression ability in the negative value range. Dropout layers (dropout rate 0.3) and batch normalization layers are introduced between the fully connected layers to prevent overfitting and improve generalization ability. The output layer uses the Softmax function, and the output sample is the probability distribution vector p=[p I ,p II ,p III ,p IV ].

[0109] A composite loss function incorporating physical consistency constraints is defined to guide the parameter updates of a deep fully connected neural network, ensuring that the evaluation results conform to the fundamental laws of rock mechanics. The composite loss function L... total Classification error loss L class And physical consistency constraint loss L phy Weighted composition: (35) Where λ is the weighting coefficient.

[0110] L class This is used to measure the difference between the predicted probability distribution of the output of a deep fully connected neural network and the true benchmark rating level labels. Specifically, it is obtained by calculating the cross-entropy loss between the predicted probability distribution and the true labels. Assume the current training batch size is N, and there are C = 4 rating levels (Level I, Level II, Level III, Level IV). class Calculated using formula (25): (36) In the formula, y i,c Let y be the baseline rating label for the i-th sample. For example, if sample i is at level I, then y i =[1,0,0,0]. p i,cThis is the predicted probability of the i-th sample calculated by the Softmax function from the output layer of the deep fully connected neural network, which is the probability that the i-th sample is predicted by the model to be of the c-th level.

[0111] Physical consistency constraint loss L phy The model is constructed based on the principle of monotonicity constraints. During training, sample pairs (i,j) are randomly selected. If sample i has better performance in all evaluation metrics (such as a higher energy absorption efficiency index and a lower fragmentation score) than sample j, then the model is forced to predict a higher score (or probability I) for sample i than for sample j. If the model output violates this rule, a penalty loss is incurred.

[0112] Using augmented dataset D train The AdamW optimizer is used for iterative training. When the validation set loss no longer decreases and the physical consistency constraint loss approaches zero, training is stopped and the optimal model parameters are saved. The trained deep neural network model is then used as the final trained neural network model.

[0113] This invention utilizes variational autoencoder probabilistic modeling and reparameterized sampling techniques to generate a large number of virtual samples conforming to physical distribution laws based on limited experimental data, effectively improving the applicability and generalization ability of deep learning models in the field of rock mechanics. By introducing physical consistency constraint loss, the basic principles of rock mechanics and expert experience are integrated into the model training process, ensuring that the evaluation results possess both high data-driven accuracy and strict physical rationality. Through the inverted pyramid structure of a deep fully connected neural network, high-order abstract features of 20 evaluation indicators are effectively extracted, establishing a nonlinear mapping relationship from multi-dimensional indicators to comprehensive performance levels. Combined with regularization techniques such as Dropout and batch normalization, the model's anti-overfitting ability and stability are enhanced, ensuring the consistency and comparability of evaluation results for support materials with different formulations and batches.

[0114] In an optional embodiment, the physical consistency constraint loss is determined as follows: For all selected sample pairs, the hinge loss is calculated based on the prediction score relationship of each sample pair, and the physical consistency constraint loss is determined based on the hinge loss of all sample pairs. Specifically, for a sample pair consisting of a sample with better physical performance and a sample with worse physical performance as determined by the input features, the prediction score of the sample with better physical performance and the prediction score of the sample with worse physical performance are calculated respectively. If the prediction score of the sample with better physical performance is less than or equal to the prediction score of the sample with worse physical performance, the hinge loss is calculated based on the prediction scores of the sample with better physical performance and the sample with worse physical performance; otherwise, the hinge loss of the sample pair is zero.

[0115] Specifically, since the output of a deep fully connected neural network is a probability vector, in order to perform numerical comparisons, the probability vector is first converted into a continuous scalar score. The weight vectors for each level are defined as w = [1.0, 0.7, 0.4, 0.1] (corresponding to levels I to IV). For the i-th sample, its predicted score The calculation is as follows: (37) In the formula: represents the weight of the c-th evaluation level. A higher predicted score indicates that the model considers the material to provide better support. c is the evaluation level index, representing the four evaluation levels (Level I, Level II, Level III, Level IV). i is the sample index, representing the i-th sample in the current training batch.

[0116] In the current training batch, M pairs of samples (i,j) are randomly selected. For each pair of samples, based on its feature vector x, the relative merits of their physical performance are determined. The determination method is as follows: Calculate the weighted Euclidean distance between the feature vectors of two samples or directly compare their initial composite scores S. total If the physical properties of sample i are significantly better than those of sample j (denoted as x), then... i x j In theory, the model's prediction score must satisfy the following conditions: > .

[0117] Hinge loss and its variants are used to penalize predictions that violate the above physical laws.

[0118] (38) In the formula, Lphy is the physical consistency constraint loss, and max represents taking the maximum value. Ω is the set of sample pairs that satisfy the physical dominance relation xi>xj. M is the number of sample pairs in the set Ω. Let be the predicted score for the i-th sample. Let ζ be the predicted score for the j-th sample. ζ is the safety margin, usually set to 0.1. This means that the score for a good sample must not only be higher than that of a bad sample, but also at least 0.1 points higher; otherwise, a small loss will occur. If sample i has better physical performance, but the model gives sample j a higher predicted score (i.e., ...), then... > If Ssafe(j) - Ssafe(i) is positive, it indicates an error has occurred. max(0,…) is the ReLU function. If the prediction order is correct ( > If the expression in parentheses is negative, the loss is zero; a loss only occurs when the prediction order is incorrect or the discrimination is insufficient.

[0119] This invention, through the aforementioned physical consistency constraint loss, transforms the fundamental mechanical principle that samples with superior physical performance should receive higher evaluation scores into a calculable loss function term, ensuring that the AI ​​model strictly adheres to domain knowledge. Physical consistency constraints effectively prevent the model from making absurd predictions that violate common sense and physical laws, ensuring reasonable evaluation results even in data-sparse regions. As a powerful regularization method, physical constraints guide the model to learn essential characteristics that conform to physical laws, rather than simply fitting surface correlations in the training data, thus improving the prediction stability for unknown materials. In situations with limited data, physical constraints provide guiding signals for model optimization, effectively preventing overfitting, accelerating model convergence, and enhancing practical value in small-sample scenarios.

[0120] The following describes the laboratory static load evaluation device for the rock-surrounding control effect of coal mine sprayed support material provided by the present invention. The laboratory static load evaluation device for the rock-surrounding control effect of coal mine sprayed support material described below can be referred to in correspondence with the laboratory static load evaluation method for the rock-surrounding control effect of coal mine sprayed support material described above.

[0121] The laboratory static load evaluation device for the surrounding rock control effect of coal mine sprayed support materials provided by this invention is based on... Figure 8 As shown, it includes: Data acquisition module 210 is used to acquire static load test datasets of sprayed sample group and untreated sample group, the datasets including load, strain and acoustic emission signals collected through static load test; The parameter calculation module 220 is used to calculate a multi-dimensional evaluation index characterizing the synergistic control effect of the sprayed material and the surrounding rock based on the static load test dataset. The effect evaluation module 230 is used to input the multi-dimensional evaluation indicators into a trained neural network model to obtain a comprehensive judgment result on the rock control effect of the sprayed support material; wherein, the neural network model outputs the comprehensive judgment result on the rock control effect of the sprayed support material by verifying the physical consistency of the changes between different multi-dimensional evaluation indicators.

[0122] Figure 9 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 9As shown, the electronic device may include a processor 810, a communications interface 820, a memory 830, and a communication bus 840. The processor 810, communications interface 820, and memory 830 communicate with each other via the communication bus 840. The processor 810 can call logical instructions from the memory 830 to execute a laboratory static load evaluation method for the rock mass control effect of coal mine sprayed support materials.

[0123] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0124] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to perform the laboratory static load evaluation method for the surrounding rock control effect of coal mine sprayed support materials provided by the above methods.

[0125] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a laboratory static load evaluation method for the surrounding rock control effect of coal mine sprayed support materials provided by the methods described above.

[0126] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0127] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0128] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention 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 technical features; and these 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 the present invention.

Claims

1. A laboratory static load evaluation method for the surrounding rock control effect of sprayed support materials in coal mines, characterized in that, include: Obtain static load test datasets for sprayed and untreated sample groups, the datasets containing load, strain, and acoustic emission signals acquired through static load tests; Based on the static load test dataset, a multidimensional evaluation index characterizing the synergistic control effect of the sprayed material and the surrounding rock is calculated. The multidimensional evaluation indicators are input into a trained neural network model to obtain a comprehensive judgment result on the rock control effect of the sprayed support material; wherein, the neural network model outputs the comprehensive judgment result on the rock control effect of the sprayed support material by verifying the physical consistency of the changes between different multidimensional evaluation indicators.

2. The laboratory static load evaluation method for the surrounding rock control effect of coal mine sprayed support material according to claim 1, characterized in that, Based on the static load test dataset, multidimensional evaluation indices characterizing the synergistic control effect of the sprayed material and the surrounding rock are calculated, including: From the static load test dataset, the raw data used to calculate various evaluation indicators are extracted; the raw data includes: stress-strain curve feature point data, cumulative parameters of acoustic emission signals, test peak intensity, and relevant geometric parameters; Based on the original data, determine the parameters used for inter-group comparison of the sprayed sample group and the untreated sample group for each of the evaluation indicators; The evaluation value of each evaluation index is determined based on the parameters applied to the inter-group comparison between the sprayed sample group and the untreated sample group. The calculated evaluation values ​​are combined to form the multidimensional evaluation index.

3. The laboratory static load evaluation method for the surrounding rock control effect of coal mine sprayed support material according to claim 1, characterized in that, The static load tests include uniaxial compression tests, uniaxial tensile tests, shear strength tests, point load strength tests, and triaxial compression tests. The acoustic emission signal was acquired synchronously during the uniaxial compression test.

4. The laboratory static load evaluation method for the surrounding rock control effect of coal mine sprayed support material according to claim 1, characterized in that, The multidimensional evaluation indices include multiple parameters such as uniaxial compressive strength change rate, uniaxial compressive modulus change rate, uniaxial compressive Poisson's ratio change rate, uniaxial secant modulus change rate, uniaxial initial deformation index, uniaxial post-peak residual strength change rate, uniaxial failure strain change rate, brittleness index change rate, fracture suppression index, acoustic emission energy release change rate, tensile strength enhancement index, shear strength change rate, point load strength change rate, triaxial compressive strength change rate, cohesion change index, internal friction angle change rate, triaxial compressive modulus change rate, triaxial residual strength change rate, triaxial failure strain change rate, and performance dispersion improvement coefficient.

5. The laboratory static load evaluation method for the surrounding rock control effect of coal mine sprayed support material according to claim 1, characterized in that, The neural network model is trained in the following manner: Obtain a training dataset, which contains multiple sets of sample data. Each set of sample data includes a multidimensional evaluation index calculated from the mechanical response data of sprayed and untreated samples, as well as the corresponding benchmark evaluation level label. The training dataset is augmented using a variational autoencoder model to generate augmented samples for model training, resulting in an augmented dataset. Using the evaluation metrics in the augmented dataset as input features and the corresponding benchmark evaluation level labels as training objectives, the deep neural network model is trained by minimizing the composite loss function, and the trained deep neural network model is used as the neural network model. The composite loss function is determined based on the classification error loss and the physical consistency constraint loss. The classification error loss is determined based on the predicted probability distribution of the model output and the true benchmark rating label. The physical consistency constraint loss is determined based on the predicted score relationship of sample pairs in the training batch. The sample pairs are selected based on the physical performance relationship of the input features of the sample pairs, and the sample with better physical performance should have a predicted score no lower than that of the sample with worse physical performance.

6. The laboratory static load evaluation method for the surrounding rock control effect of coal mine sprayed support material according to claim 5, characterized in that, The physical consistency constraint loss is determined in the following manner: For all selected sample pairs, the hinge loss is calculated based on the prediction score relationship of each sample pair, and the physical consistency constraint loss is determined based on the hinge loss of all sample pairs. Specifically, for a sample pair consisting of a sample with better physical performance and a sample with worse physical performance, based on the input features, the prediction score of the sample with better physical performance and the prediction score of the sample with worse physical performance are calculated respectively. If the prediction score of the sample with better physical performance is less than or equal to the prediction score of the sample with worse physical performance, the hinge loss is calculated based on the prediction scores of the sample with better physical performance and the sample with worse physical performance. Otherwise, the hinge loss of the sample pair is zero.

7. A laboratory static load evaluation device for the surrounding rock control effect of sprayed support materials in coal mines, characterized in that, include: The data acquisition module is used to acquire static load test datasets of the sprayed sample group and the untreated sample group. The datasets include load, strain and acoustic emission signals collected through static load tests. The parameter calculation module is used to calculate multi-dimensional evaluation indicators characterizing the synergistic control effect of the sprayed material and the surrounding rock based on the static load test dataset. The effect evaluation module is used to input the multi-dimensional evaluation indicators into a trained neural network model to obtain a comprehensive judgment result on the rock control effect of the sprayed support material; wherein, the neural network model outputs the comprehensive judgment result on the rock control effect of the sprayed support material by verifying the physical consistency of the changes between different multi-dimensional evaluation indicators.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements a laboratory static load evaluation method for the surrounding rock control effect of coal mine sprayed support material as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements a laboratory static load evaluation method for the surrounding rock control effect of coal mine sprayed support materials as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements a laboratory static load evaluation method for the surrounding rock control effect of coal mine sprayed support materials as described in any one of claims 1 to 6.

Images