Intelligent sensing experiment platform for mechanical properties of coal rock

Abstract

The invention relates to the technical field of mechanical property testing, in particular to a coal rock mechanical property intelligent sensing experiment platform which comprises a pressure data monitoring module, a dynamic behavior simulation module, a fatigue characteristic evaluation module and a performance evaluation adjustment module. According to the invention, through pressure monitoring and intelligent data processing, the capability of capturing microscopic changes in the coal rock is improved, a continuous pressure reading sequence and trend analysis are generated in real time, instant identification of microcracks and stress concentration is allowed, and real-time simulation and prediction of dynamic behaviors are allowed, so that the response time is optimized; the method also improves the accuracy of future coal rock behavior prediction, ensures the scientific decision support of mining safety, continuous strength test and fatigue data analysis to further deeply understand the failure mode of the material, clearly identifies the fatigue accumulation point, ensures the accuracy and adaptability of the result through cross validation, realizes the real-time adjustment of prediction parameters, and improves the prediction efficiency. Therefore, the method adapts to constantly changing mining conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical property testing, and in particular to an intelligent perception experimental platform for mechanical properties of coal and rock. Background Art

[0002] Mechanical properties testing technology involves measuring the behavior and characteristics of materials under external forces. The research focus in this field includes but is not limited to mechanical properties such as elasticity, plasticity, hardness, compressive strength and tensile strength of materials. There are various testing methods, from simple compression tests to complex fatigue tests and fracture mechanics analysis, all of which are aimed at evaluating the performance and durability of materials under specific environmental and load conditions.

[0003] Among them, the intelligent perception experimental platform for coal and rock mechanical properties is a platform used to measure and analyze the mechanical behavior of coal and rock samples under different environmental and stress conditions. It uses a variety of sensors and data processing algorithms to achieve accurate monitoring and intelligent analysis of the mechanical properties of coal and rock. Its main purpose is to evaluate the stability and safety of coal and rock before mining, help engineers design safer mining plans, and provide data support in scientific research to promote in-depth research on coal and rock mechanics.

[0004] When dealing with mechanical property tests of coal and rock materials, existing technologies lack the ability to monitor changes in the internal structure of materials in real time, which is particularly critical in the assessment of coal and rock stability and safety. They rely on discrete and static testing methods, which limits the ability to respond to and predict the dynamic behavior of coal and rock. For example, in long-term fatigue tests, the lack of real-time data processing leads to untimely identification of fatigue accumulation points, which can easily cause safety risks in actual mining operations. Existing methods cannot achieve rapid data feedback and adjustment in the analysis of material failure modes, and perform insufficiently in engineering applications that require rapid response. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings existing in the prior art and to propose an intelligent perception experimental platform for coal and rock mechanical properties.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: an intelligent perception experimental platform for coal and rock mechanical properties, the platform comprising:

[0007] The pressure data monitoring module is based on a pressure sensor, uses a signal amplifier to adjust the pressure signal strength, screens periodic and non-periodic pressure change data, forms a continuous pressure reading sequence, and performs preliminary trend analysis on real-time pressure data to identify microcracks and stress concentration areas to obtain structural change indicators;

[0008] The dynamic behavior simulation module processes the time series data based on the structural change index, analyzes the dynamic characteristics of coal-rock behavior and structural response, simulates and predicts future coal-rock behavior, evaluates the response mode under differential pressure conditions, and obtains behavior response characteristic information;

[0009] The fatigue characteristic evaluation module continuously applies test force or pressure based on the behavioral response characteristic information, analyzes the periodic and non-periodic changes of the coal rock fatigue data, compares it with the failure threshold, identifies the fatigue accumulation point, and combines the historical failure data to predict the failure time and fatigue stage of the coal rock, and obtains the fatigue failure prediction record;

[0010] The performance evaluation and adjustment module evaluates the mechanical properties and fatigue life of the coal and rock materials based on the fatigue failure prediction records and utilizes a cross-validation method, and adjusts the prediction parameters of the coal and rock performance according to the evaluation results to obtain mechanical behavior evaluation results.

[0011] The present invention is improved in that the steps of forming the continuous pressure reading sequence are specifically:

[0012] Based on the pressure sensor, the original pressure data is monitored and collected in real time, and the pressure signal intensity is adjusted using the signal amplifier to obtain the adjusted pressure data set;

[0013] Based on the adjusted pressure data set, the periodic and non-periodic pressure change data are screened, using the formula:

[0014] P t+1 =α·P t +(1-α)·P avg ;

[0015] Perform data smoothing to obtain a continuous pressure reading sequence, where P t+1 Represents the pressure reading at the next time point, P t Represents the pressure reading at the current time point, P avg is the average pressure value and α is the smoothing coefficient.

[0016] The present invention is improved in that the step of obtaining the structural change index is specifically as follows:

[0017] Performing preliminary trend analysis on the real-time pressure data, monitoring trends and mutation points in the pressure data, and obtaining trend analysis information;

[0018] Based on the trend analysis information, the formula is used:

[0019]

[0020] Microcracks and stress concentration areas are identified, and the structural change index is obtained, where K is the stress intensity factor, σ is the stress value in the stress concentration area, a is the crack length, Y is the geometric shape factor, and ν is the material Poisson's ratio.

[0021] The present invention is improved in that the analysis steps of the dynamic characteristics of coal-rock behavior and structural response are specifically as follows:

[0022] Based on the structural change indicator, performing time series data processing, including filtering and denoising, to obtain cleaned time series data;

[0023] Applying a state space model to process the cleaned time series data, mining hidden dynamic states and behavior patterns in the data, and obtaining model analysis results;

[0024] Based on the analysis results of the model, the formula is adopted:

[0025]

[0026] Predict the behavior and structural response of coal rock under different environmental and pressure conditions, and obtain the dynamic characteristic data of coal rock, where RS(t) represents the behavior response of coal rock at time point t, SQ(t) is the current state of coal rock obtained from the state space model, and α S is the weight factor, β S is the adjustment coefficient, γ S is the growth rate of the exponential function.

[0027] The present invention is improved in that the step of acquiring the behavior response characteristic information is specifically as follows:

[0028] Based on the dynamic characteristics, the differential behaviors and response patterns are grouped, each type of behavior pattern is analyzed, and its performance under future differential pressure conditions is predicted and simulated to obtain preliminary simulation prediction results;

[0029] Based on the preliminary simulation prediction results, the behavior changes of coal and rock in the simulation results are evaluated, the response patterns of coal and rock under various pressure conditions are identified, and the behavior response characteristic information is obtained.

[0030] The present invention is improved in that the step of identifying the fatigue accumulation point is specifically:

[0031] Based on the behavior response characteristic information, continuous mechanical testing is performed on the coal rock, periodic test force or pressure is applied, and stress and strain data of the coal rock under various loads are recorded in real time to obtain a stress response data set;

[0032] Analyzing the stress response data set, converting the time domain data into frequency domain data using Fourier transform, and identifying periodic and non-periodic changes in the data to obtain initial fatigue point analysis results;

[0033] Based on the initial fatigue point analysis results, the formula is used:

[0034]

[0035] Calculate the deviation degree of multiple points from the failure threshold and determine the key fatigue accumulation point, where PH fatigue Indicates the fatigue accumulation percentage, SH i represents the stress value at the i-th data point, SH thresh represents the failure threshold, N h Represents the total number of data points.

[0036] The present invention is improved in that the steps of obtaining the fatigue failure prediction record are specifically as follows:

[0037] Based on the identified fatigue accumulation points, the stress response of the coal rock under the same test conditions is continuously monitored to obtain an extended stress data set;

[0038] Based on the extended stress data set, the trends and potential fatigue behaviors in the data are evaluated and combined with the historical failure data to predict the potential failure time and fatigue stage of the coal rock and obtain the fatigue failure prediction record.

[0039] The present invention is improved in that the steps of obtaining the mechanical behavior evaluation results are specifically as follows:

[0040] Based on the fatigue failure prediction records, the mechanical properties and fatigue life of the coal-rock material are evaluated by using a cross-validation method to obtain verification data of the coal-rock material;

[0041] Based on the coal rock material verification data, the prediction parameters of coal rock performance are adjusted, including modifying material characteristic parameters and fatigue parameters, using the formula:

[0042] E mech =K adj ×P orig +B factor ;

[0043] Get the mechanical behavior evaluation result E mech , where K adj is the coefficient adjusted according to the cross-validation results, P orig is the original mechanical properties parameter, B factor is the bias factor.

[0044] Compared with the prior art, the advantages and positive effects of the present invention are:

[0045] In the present invention, the ability to capture microscopic changes inside coal and rock is improved through pressure monitoring and intelligent data processing. The real-time generated continuous pressure reading sequence and trend analysis allow instant identification of microcracks and stress concentration. The real-time simulation and prediction of dynamic behavior not only optimizes the response time, but also improves the accuracy of predicting future coal and rock behavior, ensuring scientific decision-making support for mining safety. Continuous strength testing and fatigue data analysis further deepen the understanding of material failure modes and clearly identify fatigue accumulation points. The accuracy and adaptability of the results are ensured through cross-validation, and real-time adjustment of prediction parameters is achieved to adapt to changing mining conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a flow chart of the platform of the present invention;

[0047] Figure 2 is a flow chart of forming a continuous pressure reading sequence in the present invention;

[0048] Figure 3 A flowchart for obtaining the structural change index in the present invention;

[0049] Figure 4 It is a flow chart for analyzing the dynamic characteristics of coal rock behavior and structural response in the present invention;

[0050] Figure 5 is a flow chart for obtaining behavior response characteristic information in the present invention;

[0051] Figure 6 A flow chart for identifying fatigue accumulation points in the present invention;

[0052] Figure 7 A flowchart for obtaining fatigue failure prediction records in the present invention;

[0053] Figure 8 The figure is a flow chart for obtaining the mechanical behavior evaluation results in the present invention. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0055] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0056] Example

[0057] See also Figure 1 The present invention provides a technical solution: the coal rock mechanical properties intelligent perception experimental platform includes:

[0058] The pressure data monitoring module is based on a pressure sensor, uses a signal amplifier to adjust the pressure signal strength, screens periodic and non-periodic pressure change data, forms a continuous pressure reading sequence, and performs preliminary trend analysis on real-time pressure data to identify microcracks and stress concentration areas to obtain structural change indicators;

[0059] The dynamic behavior simulation module processes time series data based on structural change indicators, analyzes the dynamic characteristics of coal-rock behavior and structural response, simulates and predicts future coal-rock behavior, evaluates the response mode under differential pressure conditions, and obtains behavioral response characteristic information;

[0060] The fatigue characteristic evaluation module continuously applies test force or pressure based on the behavioral response characteristic information, analyzes the periodic and non-periodic changes of coal rock fatigue data, compares it with the failure threshold, identifies the fatigue accumulation point, and combines historical failure data to predict the failure time and fatigue stage of coal rock, and obtain fatigue failure prediction records;

[0061] The performance evaluation and adjustment module evaluates the mechanical properties and fatigue life of coal and rock materials based on fatigue failure prediction records and uses the cross-validation method. It adjusts the prediction parameters of coal and rock performance according to the evaluation results to obtain mechanical behavior evaluation results.

[0062] Structural change indicators include stress concentration indicators, crack distribution diagrams, and trend analysis data. Behavioral response characteristic information includes behavioral pattern classification information, prediction accuracy evaluation results, and dynamic pressure diagrams. Fatigue failure prediction records include periodic change diagrams, cumulative fatigue points, and failure stage determination results. Mechanical behavior evaluation results include performance attenuation curves, life estimation results, and adjusted test parameters.

[0063] See also Figure 2 , the steps for forming a sequence of continuous pressure readings are as follows:

[0064] Based on the pressure sensor, the original pressure data is monitored and collected in real time, and the pressure signal intensity is adjusted using the signal amplifier to obtain the adjusted pressure data set;

[0065] Real-time monitoring and collection of raw pressure data. During the pressure monitoring process, the sensor positions need to be reasonably arranged to cover the key areas where stress changes occur. Each sensor needs to be calibrated to reduce the interference of the external environment on data collection. During the monitoring process, the sensor converts the collected pressure signal into an electrical signal. The strength of the electrical signal is determined by the pressure value. The electrical signal is recorded at a fixed frequency. The sampling rate setting needs to ensure the continuity of the pressure change. The recorded signal is stored as a raw pressure data file for subsequent analysis to complete the initial collection and monitoring process of pressure data.

[0066] Based on the adjusted pressure data set, filter the periodic and non-periodic pressure change data, using the formula:

[0067] P t+1 =α·P t +(1-α)·P avg ;

[0068] Perform data smoothing to obtain a continuous pressure reading sequence, where P t+1 Represents the pressure reading at the next time point, which is a predicted or estimated value, P t Represents the pressure reading at the current time point, representing the latest monitored pressure value, P avg is the pressure average, which is used to indicate the average level of pressure readings in the recent period, and α is the smoothing coefficient, which is used to adjust the impact of historical data on the current value;

[0069] In the real-time monitoring data, the latest pressure P t is 150000Pa, and the recent average pressure P avg The pressure is 145000Pa, and the smoothing coefficient α is selected as 0.6. The calculation is performed to reflect the influence of past data on new data. First, the calculation is:

[0070] α·P t =0.6 150000 = 90000;

[0071] Then calculate:

[0072] (1-α)·P avg =0.4 145000 = 58000;

[0073] Optimal calculation:

[0074] P t+1 =90000+58000=148000;

[0075] The result shows that the pressure prediction at the next time point after smoothing is 148,000 Pa, which means that the pressure reading is relatively stable, reducing the fluctuations caused by sudden changes.

[0076] See also Figure 3 , the steps to obtain the structural change index are as follows:

[0077] Conduct preliminary trend analysis on real-time pressure data, monitor trends and mutation points in pressure data, and obtain trend analysis information;

[0078] By collecting continuous pressure data in areas where coal and rock pressure is concentrated, the data sampling frequency is strictly set to avoid information loss and redundancy. The collected data is recorded and transmitted in real time. The pressure change rate is obtained by calculating the difference between continuous data points in a short period of time, and the acceleration is further calculated through the second derivative of the pressure rate to identify the mutation point. Subsequently, the continuous data is processed in segments through the sliding window technology, and the characteristic indicators such as the mean, variance and peak value of the pressure change are statistically analyzed in each segment. The indicators can reveal the change of trend. Through automated logical judgment, significant mutation points are marked and stored for subsequent in-depth analysis to obtain trend analysis information.

[0079] Based on the trend analysis information, the formula is used:

[0080]

[0081] Identify microcracks and stress concentration areas, and obtain structural change indicators, where K is the stress intensity factor, a measure of the stress state at the crack tip, used to evaluate the severity of the crack under given stress, crack length and material properties, σ is the stress value in the stress concentration area, indicating the magnitude of the normal stress at the current crack location, a is the crack length, which refers to the straight-line distance from the crack starting point to the tip, Y is the geometric shape factor, considering how the geometry and crack location affect the stress concentration around the crack, and is determined based on the specific shape and location of the crack, ν is the material Poisson's ratio, which indicates the ratio of lateral contraction to longitudinal extension of the material when subjected to stress;

[0082] If the crack length a in the coal rock is 0.1m, the geometric shape factor Y is 1.0, the Poisson's ratio ν is 0.25, and the stress σ is 10MPa, substitute it into the formula for calculation. First calculate

[0083]

[0084] Calculate the denominator:

[0085] 1-ν 2 =1-0.25 2 =0.9375;

[0086] Finally calculate K:

[0087]

[0088] The results show that the stress intensity factor is It shows that under given stress and crack length, the structural risk of coal rock is relatively low, but it still needs regular monitoring to prevent structural failure.

[0089] See also Figure 4 , the analysis steps of the dynamic characteristics of coal-rock behavior and structural response are as follows:

[0090] Based on the structural change indicators, time series data processing is performed, including filtering and denoising, to obtain cleaned time series data;

[0091] In the filtering process, a high-pass filter is used to remove low-frequency interference signals, and a low-pass filter is used to weaken high-frequency noise to ensure that the main frequency components related to coal and rock behavior in the data are retained. Then, mean smoothing is applied to the filtered signal to eliminate residual sudden outliers. The sliding window standard deviation threshold method is used in the denoising process. The window length is set to the number of sampling points in a fixed period, and the sampling values ​​that exceed the standard deviation range are eliminated. The impurity information in the time series is cleaned through layer-by-layer processing to finally obtain the cleaned time series data.

[0092] Apply the state space model to process the cleaned time series data, mine the hidden dynamic states and behavior patterns in the data, and obtain the model analysis results;

[0093] The main trend term, periodic term and random term of the time series are extracted through data decomposition, and the components are input into the state space model as state variables. The model structure is set as a group of dynamic equations containing observable and unobservable states according to the dynamic behavior characteristics of coal and rock. The Bayesian estimation method is used to iteratively optimize the model parameters, and the hidden dynamic state is further calculated. The characteristic data of coal and rock behavior and structural response are obtained by model solution. The data fully describes the dynamic changes of coal and rock in the time dimension and reveals the potential behavior patterns, thus obtaining the model analysis results.

[0094] Based on the model analysis results, the formula is adopted:

[0095]

[0096] Predict the behavior and structural response of coal rock under different environmental and pressure conditions, and obtain the dynamic characteristic data of coal rock, where RS(t) represents the behavior response of coal rock at time point t, SQ(t) is the current state of coal rock obtained from the state space model, including the physical properties, chemical properties of coal rock or comprehensive indicators obtained by multiple sensors, such as temperature, pressure, etc., α S is a weight factor used to adjust the contribution of SQ(t) to the behavioral response, β S is an adjustment factor used to describe time-related behavioral changes, such as the gradual strengthening or weakening trend of coal rock behavior, γ S It is the growth (or decay) rate of an exponential function, which determines the speed of change in coal rock behavior and reflects the sensitivity of coal rock to changes in specific environmental pressures or conditions;

[0097] At a specific measurement time point t, the current state of the coal rock SQ(t) obtained from the state space model is 150, which represents the comprehensive state index of the coal rock, including physical and chemical properties, and the weight factor α is set S is 0.5, and the adjustment coefficient β S is 20, and the growth rate is γ S is 0.05, substitute it into the formula for calculation, first calculate the exponential term:

[0098] e γSt =e 0.05×10 =e 0.5 ≈1.64872;

[0099] Calculate RS(t):

[0100] RS(t)=0.5×150+20×1.64872≈75+32.9744=107.9744;

[0101] The results show that the coal rock behavior response at time point t is 107.9744, and the value reflects the behavior intensity and response mode of the coal rock under given states and conditions.

[0102] See also Figure 5 ,The specific steps for obtaining behavioral response characteristic information are:

[0103] Based on dynamic characteristics, the differential behaviors and response patterns are grouped, each type of behavior pattern is analyzed, and its performance under future differential pressure conditions is predicted and simulated to obtain preliminary simulation prediction results;

[0104] Through preliminary clustering analysis of the data, different behavior patterns are classified into groups with similar dynamic characteristics. In the specific analysis, based on the physical properties and behavioral characteristics of coal and rock, the state parameters such as pressure, displacement, and strain are quantified and input into the clustering model, and the Euclidean distance of each group of data is calculated, and then the state points with high similarity are divided into the same group. After that, the behavioral characteristics of each group are extracted and counted, including the mean, variance and change trend within each group, and the response characteristics of coal and rock in the differential behavior patterns are further analyzed. Through simulation experiments or data mapping methods, combined with dynamic characteristics and pressure gradients, the future performance of each type of behavior pattern is predicted to generate preliminary simulation prediction results.

[0105] Based on the preliminary simulation prediction results, evaluate the behavior changes of coal and rock in the simulation results, identify the response patterns of coal and rock under various pressure conditions, and obtain the behavior response characteristics information;

[0106] The deformation and crack propagation behaviors of coal and rock in the simulation are monitored, and the response information of coal and rock under pressure conditions in the simulation data, including the number of cracks, crack propagation speed and deformation rate, is collected. The data are divided into different stages in chronological order, and the characteristic data of each stage are summarized. The key parameters reflecting the response mode of coal and rock are extracted, such as the velocity threshold of crack growth, the growth characteristics of deformation speed and the peak change range under stress. The coal and rock behavior patterns are classified based on the parameters. By analyzing the relationship between pressure conditions and coal and rock responses one by one, the response law of coal and rock is extracted, and the influence pattern of changes in pressure conditions on coal and rock behavior is summarized. Finally, the dynamic response characteristics of coal and rock under various pressure conditions are obtained, including the critical characteristics of crack propagation, the relationship between deformation rate and pressure, etc.

[0107] See also Figure 6 , the steps for identifying fatigue accumulation points are as follows:

[0108] Based on the behavioral response characteristic information, continuous mechanical testing is performed on the coal rock, periodic test force or pressure is applied, and the stress and strain data of the coal rock under various loads are recorded in real time to obtain a stress response data set;

[0109] Prepare test samples to ensure that the volume, density and moisture content of coal rock samples meet consistent standards, and calibrate the test equipment to provide stable test force. Use a cyclic loading method to apply pressure of different frequencies and amplitudes to the coal rock. During the test, stress sensors and strain gauges are used to record the stress and strain data of the coal rock in real time. The collected real-time data is stored in the data acquisition system. At the same time, the physical changes of the coal rock under the load are recorded. By setting a variety of loading modes, such as constant load, linear incremental load and nonlinear variable load, etc., mechanical tests are repeated multiple times to ensure that the data covers a variety of pressure conditions, and the stress and strain data characteristics of the coal rock under each loading mode are extracted. The differences in the data are compared and analyzed to form a complete stress response data set.

[0110] Analyze the stress response data set, use Fourier transform to convert time domain data into frequency domain data, identify periodic and non-periodic changes in the data, and obtain the initial fatigue point analysis results;

[0111] The stress and strain values ​​at each time point in the data set are sampled, and the outliers and missing values ​​in the original data are removed and supplemented. The sampling point data are sorted according to the time sequence, and the stress and strain change trends in different time periods are calculated in segments. Then, the time domain data are converted into frequency domain data using data processing methods, and the characteristic parameters in the frequency domain data are extracted, including the frequency distribution and amplitude distribution of periodic signals. The time series characteristics of the data are combined to identify the periodic and non-periodic changes, and the change law of the frequency domain characteristic data is calculated according to the time period. The fatigue point positions and the corresponding stress and strain value change trends that appear in the frequency domain data are marked to obtain the initial analysis results of the coal rock fatigue points.

[0112] Based on the initial fatigue point analysis results, the formula is used:

[0113]

[0114] Calculate the deviation degree of multiple points from the failure threshold and determine the key fatigue accumulation point, where PH fatigue Indicates the fatigue accumulation percentage, indicating the overall situation of fatigue damage accumulation of coal and rock during the test process. i Represents the stress value at the i-th data point, which is the coal rock stress data recorded by the real-time monitoring equipment under different test forces or pressures. thresh Indicates the failure threshold, which is used to determine when coal and rock fail. N h Represents the total number of data points, that is, the total number of data points collected during the entire test;

[0115] Collected 2 data points N h =2(total number of data points), SH thresh=200 stress (failure threshold,) SH 1 = 250 stress (stress value of the first data point), SH 2 = 220 stress (the stress value of the second data point), calculate the squared deviation ratio of the first data point:

[0116]

[0117] Compute the ratio of squared deviations for the second data point:

[0118]

[0119] Calculate the total fatigue accumulation percentage:

[0120] PH fatigue =(12.5+2.0)×100%=14.5%;

[0121] The results show that under the conditions of two data points, the fatigue accumulation percentage is 14.5%, indicating that the degree of fatigue damage accumulation exhibited by the coal rock during the test requires further monitoring and analysis.

[0122] See also Figure 7 , the specific steps for obtaining fatigue failure prediction records are:

[0123] Based on the identified fatigue accumulation points, the stress response of coal and rock under the same test conditions is continuously monitored to obtain an extended stress data set;

[0124] First of all, it is necessary to ensure the accuracy and reliability of the monitoring equipment, adjust the sensors used for monitoring to ensure the continuity and consistency of the data, conduct regular calibration of coal and rock samples to simulate the same test conditions, continuously apply periodic test forces to ensure that the loading force is consistent each time, record the stress response of the coal and rock in each test, synchronize the collected data in time to ensure the integrity of the data set, obtain multiple stress response data of coal and rock under the same conditions through repeated tests, and integrate the data to form an extended stress data set.

[0125] Based on the extended stress data set, the trends and potential fatigue behaviors in the data are evaluated, and combined with the historical failure data, the potential failure time and fatigue stage of the coal and rock are predicted to obtain fatigue failure prediction records;

[0126] Clean and process the data, remove outliers and noise, smooth the data to ensure the accuracy of the analysis, analyze the main trends in the data through statistical analysis methods, identify the periodic and non-periodic change characteristics in the data set, compare and analyze with historical failure data, evaluate the potential failure risk of coal and rock samples through trend analysis technology, use statistical prediction models to predict the potential failure time of coal and rock based on current stress response data, determine the stage of fatigue failure by analyzing the evolution of fatigue stages, record the prediction results, and form a fatigue failure prediction record.

[0127] See also Figure 8 , the specific steps for obtaining the mechanical behavior evaluation results are:

[0128] Based on fatigue failure prediction records, the mechanical properties and fatigue life of coal and rock materials are evaluated using the cross-validation method to obtain coal and rock material verification data;

[0129] When using the cross-validation method, the data in the fatigue failure prediction records are first divided into a training data set and a test data set to ensure the consistency of the data distribution. Through multiple divisions and cycles, each data subset is used for testing in turn, and the prediction error of the model in each cycle is recorded. The error results of each time are summarized and the average error value is calculated to evaluate the stability and generalization ability of the model. Further, by adjusting different model parameters, the performance of the model under different conditions is calculated, and the optimal model configuration is determined. Finally, the mechanical properties and fatigue life of the coal and rock materials are comprehensively evaluated to obtain the verification data of the coal and rock materials.

[0130] Based on the coal rock material verification data, the prediction parameters of coal rock performance are adjusted, including modifying the material characteristic parameters and fatigue parameters, using the formula:

[0131] E mech =K adj ×P orig +B factor ;

[0132] Get the mechanical behavior evaluation result E mech , where K adj is the coefficient adjusted according to the cross-validation results, P orig is the original mechanical performance parameter, such as elastic modulus, fatigue limit, etc., B factor is a bias factor used to reflect the influence of specific experimental conditions or external factors, such as temperature, humidity or other environmental variables;

[0133] The material characteristic parameters in the verification data are grouped, and the mechanical characteristic parameters such as elastic modulus, shear modulus, Poisson's ratio, etc. are combined with the corresponding fatigue limit and material fatigue life. A multivariate correlation model of material performance is established based on the experimental observation data. Then the fatigue limit parameters are adjusted. By fitting the fatigue life curve in the verification data, the adjustment factor of the fatigue limit parameters is determined. The deviation between the fatigue life point of each group of materials in the verification data and the fitting curve is calculated. The original parameters are corrected based on the adjustment factor and the adjusted prediction parameters are regenerated. Finally, the adjusted material characteristic parameters and fatigue parameters are applied to the mechanical performance model to obtain the modified performance prediction parameters. The following data are collected, and the adjustment coefficient K adj is 1.05, the original mechanical properties parameter P orig is 200MPa (such as elastic modulus, fatigue limit, etc.), bias factor B factor is 10MPa, substitute it into the formula for calculation:

[0134] E mech =1.05×200+10=210+10=220;

[0135] This result shows that the adjusted mechanical behavior evaluation results E mech It is 220MPa, which reflects that under the given adjustment coefficient and bias factor, the predicted mechanical properties of coal rock have been improved accordingly, indicating that the prediction ability of the model is consistent with experimental observations.

[0136] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. Intelligent perception experimental platform for coal and rock mechanical properties, characterized by: The platform includes: The pressure data monitoring module is based on a pressure sensor, uses a signal amplifier to adjust the pressure signal strength, screens periodic and non-periodic pressure change data, forms a continuous pressure reading sequence, and performs preliminary trend analysis on real-time pressure data to identify microcracks and stress concentration areas to obtain structural change indicators; The dynamic behavior simulation module processes the time series data based on the structural change index, analyzes the dynamic characteristics of coal-rock behavior and structural response, simulates and predicts future coal-rock behavior, evaluates the response mode under differential pressure conditions, and obtains behavior response characteristic information; The fatigue characteristic evaluation module continuously applies test force or pressure based on the behavioral response characteristic information, analyzes the periodic and non-periodic changes of the coal rock fatigue data, compares it with the failure threshold, identifies the fatigue accumulation point, and combines the historical failure data to predict the failure time and fatigue stage of the coal rock, and obtains the fatigue failure prediction record; The performance evaluation and adjustment module evaluates the mechanical properties and fatigue life of the coal and rock materials based on the fatigue failure prediction records and utilizes a cross-validation method, and adjusts the prediction parameters of the coal and rock performance according to the evaluation results to obtain mechanical behavior evaluation results.

2. The coal rock mechanical properties intelligent perception experimental platform according to claim 1 is characterized in that: The steps of forming the continuous pressure reading sequence are specifically as follows: Based on the pressure sensor, the original pressure data is monitored and collected in real time, and the pressure signal intensity is adjusted using the signal amplifier to obtain the adjusted pressure data set; Based on the adjusted pressure data set, the periodic and non-periodic pressure change data are screened, using the formula: P t+1 =α·P t +(1-α)·P avg ; Perform data smoothing to obtain a continuous pressure reading sequence, where P t+1 Represents the pressure reading at the next time point, P t Represents the pressure reading at the current time point, P avg is the average pressure value and α is the smoothing coefficient.

3. The coal rock mechanical properties intelligent perception experimental platform according to claim 1 is characterized in that: The steps for obtaining the structural change index are specifically as follows: Performing preliminary trend analysis on the real-time pressure data, monitoring trends and mutation points in the pressure data, and obtaining trend analysis information; Based on the trend analysis information, the formula is used: Microcracks and stress concentration areas are identified, and the structural change index is obtained, where K is the stress intensity factor, σ is the stress value in the stress concentration area, a is the crack length, Y is the geometric shape factor, and ν is the material Poisson's ratio.

4. The coal rock mechanical properties intelligent perception experimental platform according to claim 1 is characterized in that: The analysis steps of the dynamic characteristics of coal-rock behavior and structural response are specifically as follows: Based on the structural change indicator, performing time series data processing, including filtering and denoising, to obtain cleaned time series data; Applying a state space model to process the cleaned time series data, mining hidden dynamic states and behavior patterns in the data, and obtaining model analysis results; Based on the analysis results of the model, the formula is adopted: Predict the behavior and structural response of coal rock under different environmental and pressure conditions, and obtain the dynamic characteristic data of coal rock, where RS(t) represents the behavior response of coal rock at time point t, SQ(t) is the current state of coal rock obtained from the state space model, and α S is the weight factor, β S is the adjustment coefficient, γ S is the growth rate of the exponential function.

5. The coal rock mechanical properties intelligent perception experimental platform according to claim 1 is characterized in that: The steps for obtaining the behavior response characteristic information are specifically as follows: Based on the dynamic characteristics, the differential behaviors and response patterns are grouped, each type of behavior pattern is analyzed, and its performance under future differential pressure conditions is predicted and simulated to obtain preliminary simulation prediction results; Based on the preliminary simulation prediction results, the behavior changes of coal and rock in the simulation results are evaluated, the response patterns of coal and rock under various pressure conditions are identified, and the behavior response characteristic information is obtained.

6. The coal rock mechanical properties intelligent perception experimental platform according to claim 1 is characterized in that: The steps of identifying fatigue accumulation points are specifically as follows: Based on the behavior response characteristic information, continuous mechanical testing is performed on the coal rock, periodic test force or pressure is applied, and stress and strain data of the coal rock under various loads are recorded in real time to obtain a stress response data set; Analyzing the stress response data set, converting the time domain data into frequency domain data using Fourier transform, and identifying periodic and non-periodic changes in the data to obtain initial fatigue point analysis results; Based on the initial fatigue point analysis results, the formula is used: Calculate the deviation degree of multiple points from the failure threshold and determine the key fatigue accumulation point, where PH fatigue Indicates the fatigue accumulation percentage, SH i represents the stress value at the i-th data point, SH thresh represents the failure threshold, N h Represents the total number of data points.

7. The coal and rock mechanical properties intelligent perception experimental platform according to claim 1 is characterized in that: The steps for obtaining the fatigue failure prediction record are specifically as follows: Based on the identified fatigue accumulation points, the stress response of the coal rock under the same test conditions is continuously monitored to obtain an extended stress data set; Based on the extended stress data set, the trends and potential fatigue behaviors in the data are evaluated and combined with the historical failure data to predict the potential failure time and fatigue stage of the coal rock and obtain the fatigue failure prediction record.

8. The coal and rock mechanical properties intelligent perception experimental platform according to claim 1 is characterized in that: The steps for obtaining the mechanical behavior evaluation results are specifically as follows: Based on the fatigue failure prediction records, the mechanical properties and fatigue life of the coal-rock material are evaluated by using a cross-validation method to obtain verification data of the coal-rock material; Based on the coal rock material verification data, the prediction parameters of coal rock performance are adjusted, including modifying material characteristic parameters and fatigue parameters, using the formula: E mech =K adj ×P orig +B factor ; Get the mechanical behavior evaluation result E mech , where K adj is the coefficient adjusted according to the cross-validation results, P orig is the original mechanical properties parameter, B factor is the bias factor.

Images