Nonlinear dynamic disaster deep well ground pressure monitoring and early warning method and system

Abstract

The invention provides a nonlinear dynamic disaster deep well ground pressure monitoring and early warning method and system, and relates to the technical field of geological engineering and disaster early warning, and the method comprises the steps: selecting a state variable and a control variable according to an abrupt change theory, a grey theory, a Verhulst model theory, a Verhulst inverse function mode theory and a nonlinear dynamic theory; analyzing the variables and the key parameters to obtain an analysis result; according to the analysis result, predicting the instability time of the rock mass to obtain a prediction result; according to a prediction result, simulating a process from deformation to instability of the rock mass to obtain predicted rock mass instability time; the acoustic emission time sequence, the stress time sequence and the displacement time sequence serve as observation sequences, the dynamic relation between the observation sequences is described, and an inversion theory is used to determine inversion parameters; according to the prediction result, rock mass instability time and inversion parameters are predicted, the stability of the rock mass is evaluated, and an early warning result is obtained. According to the invention, comprehensive evaluation and early warning of the stability state of the rock mass can be realized.

Description

Technical Field

[0001] The present invention relates to the field of geological engineering and disaster early warning technology, and in particular to a method and system for monitoring and early warning of deep well ground pressure of nonlinear dynamic disasters. Background Art

[0002] With the continuous growth of global demand for mineral resources, mining activities are becoming more frequent and developing deeper, and deep-well mining has become an important trend in the development of mineral resources. However, deep-well mining faces complex geological environments and severe safety challenges, among which ground pressure disasters are one of the main threats to mine safety production. Ground pressure disasters can not only cause tunnel collapse and equipment damage, but also cause serious casualties, bringing huge economic losses and social impacts to mining companies.

[0003] Traditional ground pressure monitoring methods often rely on a single stress monitoring or displacement monitoring method, which is difficult to fully reflect the stability of the rock mass. At the same time, due to the high nonlinearity, fuzziness and uncertainty of the rock mass stress deformation process, traditional linear analysis methods are unable to cope with complex ground pressure problems. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a deep well ground pressure monitoring and early warning method and system for nonlinear dynamic disasters, which comprehensively utilizes a variety of advanced theories and technical means such as nonlinear scientific theory, mutation theory, grey theory, Verhulst model, etc., and realizes comprehensive evaluation and early warning of the stability state of the rock mass through real-time collection and comprehensive analysis of multi-dimensional monitoring data such as acoustic emission signals, stress changes and displacement changes.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] In a first aspect, a method for monitoring and early warning of deep well ground pressure for nonlinear dynamic disasters is provided, the method comprising:

[0007] Acquire acoustic emission signals, stress change data and displacement change data;

[0008] Preprocessing the collected raw data to obtain preprocessed data;

[0009] The key parameters are obtained by extracting the acoustic emission event rate, acoustic emission amplitude, stress change and displacement change from the pre-processed data;

[0010] According to the mutation theory, state variables and control variables are selected, and the variables and key parameters are analyzed to obtain the analysis results;

[0011] According to the analysis results, the instability time of the rock mass is predicted to obtain the prediction result;

[0012] According to the prediction results, the process from rock mass deformation to instability is simulated to obtain the predicted rock mass instability time;

[0013] The acoustic emission time series, stress time series and displacement time series are taken as observation sequences, the dynamic relationship between the observation sequences is described, and the inversion theory is used to determine the inversion parameters;

[0014] Based on the prediction results, predicted rock mass instability time and inversion parameters, the stability of the rock mass is evaluated to obtain early warning results.

[0015] Furthermore, the acoustic emission signal, stress change data and displacement change data are obtained, including:

[0016] Configure the data acquisition software based on the installation of acoustic emission sensors, strain gauges and displacement sensors;

[0017] The acoustic emission signal, stress change data and displacement change data are monitored in real time through the data acquisition software to obtain the acoustic emission signal, stress change data and displacement change data.

[0018] Furthermore, the acoustic emission event rate, acoustic emission amplitude, stress change and displacement change are extracted from the preprocessed data to obtain key parameters, including:

[0019] Setting a threshold of the preprocessed acoustic emission signal, and identifying an acoustic emission event according to the threshold of the acoustic emission signal to obtain an acoustic emission event rate;

[0020] The preprocessed acoustic emission signal is segmented, and the peak amplitude of each signal segment is obtained to obtain the acoustic emission amplitude;

[0021] Determine the stress reference value according to the preprocessed stress data, and calculate the stress change at each time point according to the reference value to obtain the stress change;

[0022] According to the preprocessed displacement data, a reference value of the displacement is determined, and the displacement change at each time point is calculated by using the reference value to obtain the displacement change.

[0023] Furthermore, according to the mutation theory, state variables and control variables are selected, and the variables and key parameters are analyzed to obtain analysis results, including:

[0024] According to the catastrophe theory, state variables and control variables are selected. The state variables are the deformation degree or stress state of the rock mass, and the control variables are the external factors that affect the system behavior, such as loading conditions and material properties.

[0025] According to the state variables and control variables, through X = a0 + a1t + a2t 2 +a3t 3 +a4t 4The state variables are expressed as a function of time, where a i is the unknown coefficient, t is the time, and X is the state variable;

[0026] As a function of time, Perform variable substitution and set To obtain the equilibrium equation Z 3 +aZ+b=06;

[0027] According to the equilibrium equation, the bifurcation set equation is calculated as 4a 3 +27b 2 = 0, to obtain the discriminant D = 4a 3 +27b 2 , a i is the coefficient in the Taylor expansion, b i The transformed coefficients, q is the adjustment term used to simplify the Taylor expansion, a and b are control variables, D is the bifurcation discriminant, ΔZ and Δt are used to describe the sudden change of state variables and their corresponding time changes;

[0028] The analytical stability is obtained according to the discriminant to obtain the analytical results.

[0029] Furthermore, based on the analysis results, the rock mass instability time is predicted to obtain the prediction results, including:

[0030] According to the analysis results, the original monitoring data of the rock mass is obtained, which is usually time series data, such as displacement, stress or deformation data of the rock mass.

[0031] pass The original data is accumulated once, where X (0) (i)>0, i is the number of monitoring data with the same interval time, X (0) (i) is the original monitoring data, and (i) is the one-time accumulated generated data, which is used to reduce the randomness of the data.

[0032] Generate data based on one accumulation, through Z (1) (i) = (X (1) (i)+X (1) (i-1))÷2 calculates the mean data, Z (1) (i) is the mean generated data,;

[0033] By defining the differential equation, Solve once to generate data X (1) The expression of (t) is Among them, when t takes a sequence number, the above formula becomes: Where i = 1, 2, ..., n; a^ and u are expressed as By defining the matrix B and vector Y, And according to the matrix B and vector Y, through generate and u parameters, is the attenuation coefficient of the system, and u is the external input or driving term;

[0034] Generate data and mean data by one accumulation, through X (0) (i) = X (1) (i)-X (1) (i-1) and Get the predicted value of the original data;

[0035] According to the predicted value, the critical value S is set, and the The time t of the expected failure event is calculated to obtain the predicted rock mass instability time, where S is the critical value and Δt is the time interval of the data.

[0036] Furthermore, based on the prediction results, the process of rock mass deformation and instability is simulated to obtain the predicted rock mass instability time, including:

[0037] According to the prediction results, the original monitoring data sequence is transformed by accumulation and fitting into the Verhulst first-order whitening nonlinear differential equation to obtain the Verhulst nonlinear differential dynamic prediction equation for rock mass instability and ground pressure disaster.

[0038] The nonlinear differential dynamic prediction equation is solved to predict the instability time, and the predicted instability time is converted into actual time to obtain the actual predicted instability time.

[0039] Furthermore, the acoustic emission time series, stress time series and displacement time series are taken as observation sequences to describe the dynamic relationship between the observation sequences, and the inversion theory is used to determine the inversion parameters, including:

[0040] Collect acoustic emission time series, stress time series and displacement time series data;

[0041] Analyze the correlation between acoustic emission, stress and displacement time series to obtain analysis results;

[0042] Based on the analysis results, the dynamic relationship between the observation sequences is established to obtain the characteristics of the parameter data;

[0043] According to the characteristics of the data, the observed data are compared with the model prediction results, and the model parameters are continuously adjusted through an iterative optimization process until a certain convergence criterion is met or a predetermined number of iterations is reached to determine the inversion parameters.

[0044] In the second aspect, a nonlinear dynamic disaster deep well ground pressure monitoring and early warning system comprises:

[0045] The acquisition module is used to acquire acoustic emission signals, stress change data and displacement change data; preprocess the collected raw data to obtain preprocessed data; extract the acoustic emission event rate, acoustic emission amplitude, stress change and displacement change from the preprocessed data to obtain key parameters; select state variables and control variables according to the catastrophe theory, analyze the variables and key parameters to obtain analysis results;

[0046] The processing module is used to predict the instability time of the rock mass according to the analysis results to obtain the prediction results; according to the prediction results, simulate the process of the rock mass deformation to instability to obtain the predicted rock mass instability time; use the acoustic emission time series, stress time series and displacement time series as observation sequences, describe the dynamic relationship between the observation sequences, and use the inversion theory to determine the inversion parameters; evaluate the stability of the rock mass according to the prediction results, the predicted rock mass instability time and the inversion parameters to obtain the early warning results.

[0047] According to a third aspect, a computing device includes:

[0048] one or more processors;

[0049] The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described.

[0050] In a fourth aspect, a computer-readable storage medium stores a program, and when the program is executed by a processor, the method described is implemented.

[0051] The above solution of the present invention includes at least the following beneficial effects:

[0052] By comprehensively applying nonlinear scientific theories, the nonlinear, fuzzy and uncertain characteristics of rock masses can be described more accurately, thereby improving the accuracy of ground pressure disaster warnings. By real-time monitoring of acoustic emission signals, stress changes and displacement changes, and extracting key parameters for analysis, early warnings can be given before rock mass instability occurs, providing mining companies with sufficient time to take preventive measures and avoid catastrophic accidents. Combining a variety of technical means such as acoustic emission monitoring, stress monitoring and displacement monitoring to form complementary advantages, the stability state of the rock mass can be more comprehensively reflected through multi-parameter comprehensive analysis. Based on scientific methods such as nonlinear dynamic models and mutation theory, a systematic early warning mechanism is established, and scientific analysis of monitoring data is carried out, avoiding the limitations of a single method and improving the scientificity and systematicity of the early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a flow chart of a method for monitoring and early warning of deep well ground pressure for nonlinear dynamic disasters provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0054] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in a form and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0055] like Figure 1 As shown, an embodiment of the present invention provides a method for monitoring and early warning of deep well ground pressure of nonlinear dynamic disasters, and the method comprises the following steps:

[0056] Step 11, obtaining acoustic emission signals, stress change data, and displacement change data;

[0057] Step 12, preprocessing the collected raw data to obtain preprocessed data;

[0058] Step 13, extracting the acoustic emission event rate, acoustic emission amplitude, stress change and displacement change from the preprocessed data to obtain key parameters;

[0059] Step 14, according to the mutation theory, select state variables and control variables, analyze the variables and key parameters to obtain analysis results;

[0060] Step 15, predicting the rock mass instability time according to the analysis results to obtain a prediction result;

[0061] Step 16, based on the prediction results, simulating the process of rock mass deformation developing to instability, so as to obtain the predicted rock mass instability time;

[0062] Step 17, taking the acoustic emission time series, stress time series and displacement time series as observation sequences, describing the dynamic relationship between the observation sequences, and using inversion theory to determine inversion parameters;

[0063] Step 18, based on the prediction results, the predicted rock mass instability time and the inversion parameters, the stability of the rock mass is evaluated to obtain an early warning result.

[0064] In the embodiment of the present invention, the early warning system can capture the dynamic changes of the rock mass in real time, improve data quality, and reduce the influence of noise and error: by extracting the acoustic emission event rate, acoustic emission amplitude, stress change and displacement change of the preprocessed data, directly linking the key indicators of the stability state of the rock mass, it is helpful to quickly identify the precursor information of rock instability; using the mutation theory, it can scientifically and systematically analyze the nonlinear change process of the rock mass, improve the accuracy and scientificity of the early warning analysis, and provide timely and accurate early warning information for mining enterprises, so that they have enough time to take preventive measures to reduce the possibility and loss of disasters; by simulating the deformation process of the rock mass, the mechanism and development trend of rock instability can be more intuitively understood, providing a scientific basis for early warning and emergency response, optimizing the model parameters through the inversion theory, improving the adaptability and accuracy of the early warning model, and making the early warning results closer to the actual situation; comprehensively evaluating the stability state of the rock mass, providing a comprehensive early warning report for mining enterprises, and guiding them to formulate scientific and reasonable safety production plans.

[0065] In a preferred embodiment of the present invention, the above step 11 includes:

[0066] Configure the data acquisition software based on the installation of acoustic emission sensors, strain gauges and displacement sensors;

[0067] The acoustic emission signal, stress change data and displacement change data are monitored in real time through the data acquisition software to obtain the acoustic emission signal, stress change data and displacement change data.

[0068] In the embodiment of the present invention, through the real-time monitoring function of the data acquisition software, the mining enterprise can instantly grasp the dynamic changes of the rock mass. Professional data acquisition software can automatically process the original data, reduce human errors, and improve the accuracy and reliability of the data; install acoustic emission sensors, stress gauges and displacement sensors to achieve comprehensive monitoring of different physical quantities of the rock mass; the automation characteristics of the data acquisition software greatly reduce the burden of manual monitoring and improve monitoring efficiency. At the same time, by integrating intelligent algorithms, the software can automatically analyze data, identify abnormal patterns, and realize intelligent early warning.

[0069] In a specific embodiment of the present invention, an acoustic emission sensor, a stress gauge and a displacement sensor are selected according to the monitoring requirements. The acoustic emission sensor should have high sensitivity and be able to capture weak acoustic emission signals; the stress gauge should be able to accurately measure the stress changes of the rock mass; the displacement sensor should have high resolution and stability and be able to accurately measure the displacement changes of the rock mass. Conduct an on-site survey of the mine to understand the geological structure, mining progress and potential dangerous areas, determine the installation location of the sensor according to the survey results, usually select the rock deformation sensitive area, stress concentration area or potential damage area, install it according to the sensor manual, select data acquisition software, ensure that the software can support the access of multiple sensors, and have real-time data acquisition, storage and analysis functions. Install the software on the monitoring center or a designated computer. Connect the sensor to the data acquisition software by wire or wireless. Perform corresponding interface configuration in the software according to the interface type of the sensor, set the sampling rate, range, filter and other parameters of the sensor in the data acquisition software to ensure the accuracy and effectiveness of data acquisition. At the same time, set the data storage path and format for subsequent data processing and analysis, open the data acquisition software, and start the real-time monitoring function. The software will automatically receive data from the sensor and perform preliminary processing, and display the acoustic emission signal, stress change data and displacement change data in real time on the software interface. At the same time, the software will automatically store this data in the set format and path for subsequent analysis.

[0070] In a preferred embodiment of the present invention, the above step 12 includes:

[0071] Clean the collected raw data, remove outliers, obtain cleaned data, and perform data conversion, data standardization / normalization and filtering on the cleaned data to obtain processed data;

[0072] In the embodiment of the present invention, the data cleaning process can remove noise, errors and outliers in the original data, ensuring that the subsequent analysis is based on an accurate and reliable data set; data standardization / normalization processing enables features of different dimensions and magnitudes to be compared and calculated on the same scale, which is crucial for many machine learning algorithms; filtering processing can remove high-frequency noise and unnecessary fluctuations in the data, making the data smoother and facilitating subsequent data analysis. At the same time, reducing redundant information in the data can reduce computational complexity and improve processing efficiency; data that has been cleaned, converted and filtered is closer to the actual situation and can more accurately reflect the actual state of the rock mass.

[0073] In a specific embodiment of the present invention, each data point is compared with a defined standard, outliers are identified, the identified outliers are directly deleted, outliers are replaced with mean, median, mode or other reasonable values, all measurements are in the same unit to avoid calculation errors caused by inconsistent units, and data is converted from the original format to a numerical format suitable for analysis. For time series data, time zone conversion or format adjustment may be required to ensure the accuracy and consistency of timestamps. The data is scaled to fall into a small specific interval: the data is scaled to fall into the [0,1] interval, but the distribution shape of the data is not necessarily changed; the selected standardization / normalization method is applied to each feature to obtain a new feature value. The data is filtered to remove noise or unnecessary frequency components.

[0074] In a preferred embodiment of the present invention, the above step 13 extracts the acoustic emission event rate, acoustic emission amplitude, stress change and displacement change from the pre-processed data to obtain key parameters, including:

[0075] Setting a threshold of the preprocessed acoustic emission signal, and identifying an acoustic emission event according to the threshold of the acoustic emission signal to obtain an acoustic emission event rate;

[0076] The preprocessed acoustic emission signal is segmented, and the peak amplitude of each signal segment is obtained to obtain the acoustic emission amplitude;

[0077] Determine the stress reference value according to the preprocessed stress data, and calculate the stress change at each time point according to the reference value to obtain the stress change;

[0078] According to the preprocessed displacement data, a reference value of the displacement is determined, and the displacement change at each time point is calculated by using the reference value to obtain the displacement change.

[0079] In the embodiment of the present invention, by setting a reasonable threshold to identify acoustic emission events, background noise and real rock mass fracture signals can be accurately distinguished. The peak amplitude of the acoustic emission signal is directly related to the energy release of the crack extension inside the rock mass. By calculating the peak amplitude of each signal segment, the intensity of rock mass fracture can be quantified, the baseline value of stress can be determined, and the stress change at each time point can be calculated, which can intuitively reflect the dynamic change of stress on the rock mass. By calculating the baseline value of displacement and the displacement change at each time point, the overall deformation trend and local damage of the rock mass can be evaluated.

[0080] In a specific embodiment of the present invention, a suitable threshold is selected based on historical data or statistical analysis. This threshold is used to distinguish between background noise and meaningful acoustic emission signals. The pre-processed acoustic emission signal is traversed. When the signal amplitude exceeds the set threshold, it is considered that an acoustic emission event has occurred, and the start time and end time of the event, as well as the signal characteristics during the event, are recorded. The number of acoustic emission events occurring within a given time window is counted, and the number of events is divided by the length of the time window to obtain the acoustic emission event rate. The pre-processed acoustic emission signal is divided into several signal segments of equal length, and for each signal segment, its maximum value, that is, the peak amplitude, is found; the peak amplitude and time position of each signal segment are recorded. The average stress value within the time period is selected as the reference value, the pre-processed stress data is traversed, and for each time point, the difference between the current stress value and the reference value, that is, the stress change, is calculated, and the stress change at each time point is recorded. The average displacement value within the time period is selected as the reference value, the pre-processed displacement data is traversed, and for each time point, the difference between the current displacement value and the reference value, that is, the displacement change, is calculated. The displacement change at each time point is recorded.

[0081] In a preferred embodiment of the present invention, the above step 14, according to the catastrophe theory, selects state variables and control variables, analyzes the variables and key parameters to obtain analysis results, including:

[0082] According to the catastrophe theory, state variables and control variables are selected. The state variables are the deformation degree or stress state of the rock mass, and the control variables are the external factors that affect the system behavior, such as loading conditions and material properties.

[0083] According to the state variables and control variables, through X = a0 + a1t + a2t 2 +a3t 3 +a4t 4 The state variables are expressed as a function of time, where a i is the unknown coefficient, t is the time, and X is the state variable;

[0084] As a function of time, Perform variable substitution and set To obtain the equilibrium equation Z 3 +aZ+b=0;

[0085] According to the equilibrium equation, the bifurcation set equation is calculated as 4a 3 +27b 2 = 0, to obtain the discriminant D = 4a 3 +27b 2 , a i is the coefficient in the Taylor expansion, b iThe transformed coefficients, q is the adjustment term used to simplify the Taylor expansion, a and b are control variables, D is the bifurcation discriminant, ΔZ and Δt are used to describe the sudden change of state variables and their corresponding time changes;

[0086] The analytical stability is obtained according to the discriminant to obtain the analytical results.

[0087] In the embodiment of the present invention, the mutation theory provides a systematic method based on physical principles to predict the instability behavior of rock mass; the early warning results obtained by this method are more scientific and reliable. By real-time monitoring of state variables and control variables and applying mutation theory for analysis, early warning signals can be found in time before rock mass instability, providing valuable response time for mining enterprises. The mutation theory is not only used for prediction, but also helps to understand the internal mechanism of rock mass instability; according to the analysis results of the mutation theory, the layout of monitoring points and monitoring frequency can be optimized to improve monitoring efficiency and reduce costs.

[0088] In a specific embodiment of the present invention, physical quantities that can reflect the current state of the rock mass, such as the degree of deformation or stress state of the rock mass, are selected to identify and quantify external factors that affect the stability of the rock mass, such as loading conditions, material properties, and geological conditions. The displacement changes, changes in the number of acoustic emission events, and changes in stress during rock deformation are all time-sensitive. The parameter variables (X, Y, Z) can be regarded as functions of time T, that is, X = f(t). Therefore, they can be expanded using the Taylor formula as follows: X = f(t) = a0+a1t+a2t 2 +…+a n t n , where a0, a1, … a n is the unknown coefficient. In actual engineering research, it can be approximately written as: X = f(t) = a0 + a1t + a2t 2 +a3t 3 +a4t 4 The unknown coefficients in the formula can be obtained through regression analysis. According to the function of time, the variables are replaced and t = Z t -q, Find: Where b0 = a4q 4 -a3q 3 +a2q 2 -a1q+a0,b1=-4a4q 3 +3a3q 2 -2a2q+a1,b2=6a4q 2 -3a3q+2a2, b4=a4, when further variable substitution is made, or Assume b4>0, then Among them: C is the shear term, which is meaningless for mutation analysis; a and b are: When C is not considered, that is, the cusp mutation model with Z as the state variable and a and b as the control variables, according to the principle of mutation theory, The equilibrium surface equation is: Z 3 +aZ+b=0, which gives the relationship between the state variable Z and the control variables a, b, which can be expressed as a surface graph. The set of folding or sharp turning points of the surface graph is called the singularity set, and its projection on the ab plane is called the bifurcation set. The bifurcation set equation is: 4a 3 +27b 2 =0, let D = 4a 3 +27b 2 , the bifurcation set is a half cubic parabola with a sharp point at (0, 0). The point (a, b) (D = 0) on the bifurcation set corresponds to the unstable state (critical state) of the system, and the system may suddenly change from one equilibrium state to another. At the same time, the bifurcation set divides the control variable plane into two areas: in the larger area, D>0, the system is in a stable state; in the smaller area, D<0, the system has three equilibrium states, two of which are stable and one is unstable. This means that when D<0, the rock mass may become unstable. The criteria for judging the unstable point are: From the above analysis, it can be concluded that there are two situations in which the rock mass system undergoes mutation: ①D=0; ②D<0, and 3Z 2 +a<0. The mutation referred to by the right branch of the bifurcation set (b>0) means that the mathematical structure of the system (the number and stability of equilibrium states) has a mutation, while the value of the state variable Z has not jumped. For the prediction of rock mass instability and ground pressure disasters in mines, it is meaningful to cross the left branch of the bifurcation set (b<0). At this time, the corresponding point is in an unstable state and the Z value jumps. Now determine the state variable values ​​corresponding to each point on the left branch of the bifurcation set (b<0). Under the condition that equation (2.15) holds, when a=0, equation (2.15) has three zero roots, Z1=Z2=Z3=0; when a<0, equation (2.14) has three real roots, which are: The state variable Z mutates when crossing the bifurcation set. According to the described substitution, the corresponding time difference before and after the rock mass instability is: The time difference Δt between the critical state (D = 0) and the destruction of the rock mass can be calculated. The sum of Δt and the duration of the critical state is the instability time of the rock mass. 2 +a<0, the system is in the bifurcation set, and the failure time calculated by the above method is generally greater than the actual failure time. Assuming b4<0, the state of the rock mass is exactly the opposite of the results discussed above.

[0089] In a preferred embodiment of the present invention, the above step 15 predicts the rock mass instability time according to the analysis results to obtain the prediction result, including:

[0090] According to the analysis results, the original monitoring data of the rock mass is obtained, which is usually time series data, such as displacement, stress or deformation data of the rock mass.

[0091] pass The original data is accumulated once, where X (0) (i)>0, i is the number of monitoring data with the same interval time, X (0) (i) is the original monitoring data, and (i) is the one-time accumulated generated data, which is used to reduce the randomness of the data.

[0092] Generate data based on one accumulation, through Z (1) (i) = (X (1) (i)+X (1) (i-1))÷2 calculates the mean data, Z (1) (i) is the mean generated data;

[0093] By defining the differential equation, Solve once to generate data X (1) The expression of (t) is Among them, when t takes a sequence number, the above formula becomes: Where i = 1, 2, ..., n; a^ and u are expressed as By defining the matrix B and vector Y, And according to the matrix B and vector Y, through generate and u parameters, is the attenuation coefficient of the system, and u is the external input or driving term;

[0094] Generate data and mean data by one accumulation, through X (0) (i) = X (1) (i)-X (1) (i-1) and Get the predicted value of the original data;

[0095] According to the predicted value, the critical value S is set, and the The time t of the expected failure event is calculated to obtain the predicted rock mass instability time, where S is the critical value and Δt is the time interval of the data.

[0096] In the embodiment of the present invention, the randomness of the data is reduced through cumulative generation, so that the prediction result is more accurate, does not require a large amount of sample data, has good adaptability to small samples and poor information systems, the parameter solution process is relatively simple, the amount of calculation is small, it is suitable for real-time online prediction, the principle is intuitive, and through real-time monitoring and prediction, an early warning signal can be issued before the rock mass becomes unstable, providing mining companies with valuable time to take preventive measures.

[0097] In a specific embodiment of the present invention, the original time series data of the rock mass are obtained from the monitoring system. These data may be displacement, stress or deformation data. Assume that these data are arranged in chronological order, and each time point corresponds to a data point. Usually, a one-time accumulation data generation method is adopted to perform a certain generation transformation on the original data. The formula can be expressed as: Among them, i is the number of monitoring data with the same interval time, X (1) (i) is the data generated by one accumulation, and the number in the bracket in the upper right corner indicates the number of accumulations; X (0) (i) is the non-negative cumulative monitoring data before, and the mean generated data is calculated as follows: Z (1) (i) = [X (1) (i)+X (1) (i-1)] / 2, let the differential equation be but is called the derivative of x, and x is background value; a, b are parameters, and X (1) The first-order linear differential equation for (t): Solving the equation gives: When t takes a sequence number, the above formula becomes: The coefficients a and u can be

[0098] It is found by the following formula: in: Result: X (0) (i) = X (1) (i)-X (1) (i-1) and The critical value S can be substituted into the above formula to obtain the damage event t: Among them, Δt is the monitoring data time interval. After the prediction results are obtained, the accuracy of the model should be tested. If the error is large, the residual can be re-established for the error sequence to make corrections.

[0099] In a preferred embodiment of the present invention, the above step 16 simulates the process of rock mass deformation developing to instability according to the prediction result to obtain the predicted rock mass instability time, including:

[0100] According to the prediction results, the original monitoring data sequence is transformed by accumulation and fitting into the Verhulst first-order whitening nonlinear differential equation to obtain the Verhulst nonlinear differential dynamic prediction equation for rock mass instability and ground pressure disaster.

[0101] The nonlinear differential dynamic prediction equation is solved to predict the instability time, and the predicted instability time is converted into actual time to obtain the actual predicted instability time.

[0102] In an embodiment of the present invention, the Verhulst model can better describe a system with saturated growth characteristics. The parameters in the model help to understand the physical mechanism of rock instability, and are suitable for rock instability prediction. It can also be extended to the prediction of other systems with similar growth characteristics. Unlike some short-term prediction methods, the Verhulst model supports long-term prediction of rock stability, which helps mining companies to formulate long-term production safety plans. Through real-time monitoring data and feedback from model prediction results, mining companies can make decisions more scientifically, optimize mining plans, and improve production efficiency and safety.

[0103] In a specific embodiment of the present invention, according to the prediction result, the original monitoring data sequence is subjected to an accumulation generation transformation to obtain the original equally spaced monitoring data sequence X (0) (t):X (0) (t) = {X (0) (1),X (0) (2)…X (0) (n)}, perform an AGO transformation on X(0)(t), and get: X (1) (t) = {X (1) (1),X (1) (2)…X (1) (n)}. X(1)(t) is fitted into the Verhulst first-order whitening nonlinear differential equation: Among them: a, b are unknown coefficients, which can be obtained by the least squares method: and

[0104] Y N =[X (0) (2),X (0) (3),…X (0) (n)) T ;according to

[0105] The unknown coefficients are solved to obtain the solution of the nonlinear differential equation: This is the established Verhulst nonlinear differential dynamic prediction model for rock mass instability and ground pressure disasters, where: t0 = 0 is the initial time, if X (0)(t) is a non-equally spaced time series data, it should be converted into an equal-spaced time series before the model is established. The process of rock mass instability is similar to the process of biological reproduction → extinction. Therefore, the critical value a / 2b of the biological model can be used as the critical judgment value of instability (displacement, acoustic emission, stress). Substitute a / 2b into The instability time t is obtained: At the initial time t0=0, then: The actual prediction time is: T = t*Δt, where t is the time series number and Δt is the average interval time of monitoring data.

[0106] In a preferred embodiment of the present invention, the above step 17 uses the acoustic emission time series, stress time series and displacement time series as observation sequences, describes the dynamic relationship between the observation sequences, and uses the inversion theory to determine the inversion parameters, including:

[0107] Collect acoustic emission time series, stress time series and displacement time series data;

[0108] Analyze the correlation between acoustic emission, stress and displacement time series to obtain analysis results;

[0109] Based on the analysis results, the dynamic relationship between the observation sequences is established to obtain the characteristics of the parameter data;

[0110] According to the characteristics of the data, the observed data are compared with the model prediction results, and the model parameters are continuously adjusted through an iterative optimization process until a certain convergence criterion is met or a predetermined number of iterations is reached to determine the inversion parameters.

[0111] In an embodiment of the present invention, by comprehensively considering the time series data of various physical quantities such as acoustic emission, stress and displacement, the stability state of the rock mass can be more comprehensively reflected. The iterative optimization process enables the model parameters to be dynamically adjusted according to the actual observation data. Based on the data analysis and model prediction results, mining enterprises can make more scientific and reasonable decisions, optimize mining plans, collect and analyze time series data in real time, so that the early warning system can promptly detect abnormal changes in the rock mass and provide timely early warning information for the mine; by establishing a more accurate and reliable early warning system, mining enterprises can more effectively manage the risk of ground pressure disasters and improve the overall safety management level.

[0112] In a specific embodiment of the present invention, the data of acoustic emission time series, stress time series and displacement time series are collected to establish a Verhulst inverse function model, which can improve the accuracy of the model warning forecast and reduce the impact of random fluctuations. The Verhulst model is: When t→0, we have That is, when time t tends to infinity, the displacement value of the characteristic curve of the Verhulst prediction model tends to a characteristic value a / b. Therefore, after the variable is exchanged, the characteristic curve of its inverse function model is symmetrical with the Verhulst model curve about the X-axis. When the displacement tends to infinity, time t tends to a certain value. Theoretically, when the deformation of the rock mass, the number of acoustic events, or the stress change value tends to infinity, it means that the rock mass is unstable and damaged. Therefore, this certain value of time t can be used as the prediction time of rock mass instability and damage. The Verhulst first-order whitening nonlinear differential equation is: The solution to this differential equation is: Find the inverse function of the above formula and swap the variables to get: The above formula is the established Verhulst inverse function model for ground pressure disaster warning and forecasting. Among them, t is the time sequence number; a and b are unknown coefficients. When: When it is positive infinity, that is: at0-bt0·t→0. The initial time t0≠0, generally t0≥1, then: t→a / b. When the deformation of the rock tends to infinity, the time t tends to a certain value (a / b). Therefore, T=a / b can be used as the instability prediction time. According to the principle of least squares method, a and b are obtained from the following formula:

[0113] Where: d(k) = 1 / (X (0) (k-t0+1)·k), after t0 is calculated, the predicted time from the starting point of the modeling data to the rock failure is: T=a / b-t0.

[0114] In a preferred embodiment of the present invention, the above step 18 includes:

[0115] Comprehensively analyze the prediction results obtained by the deep well ground pressure monitoring and early warning method of nonlinear dynamic disasters to obtain analysis results, including but not limited to the predicted value of rock mass instability time, the change trend of acoustic emission activities, stress change characteristics and displacement change rate;

[0116] Compare the current analysis results with past rock stability monitoring data under similar geological and mining conditions to obtain comparative results;

[0117] Based on the comparison results, the stability of the rock mass is evaluated to obtain early warning results.

[0118] In an embodiment of the present invention, by comprehensively analyzing the prediction results and comparing them with historical data, the stability state of the rock mass can be evaluated more accurately. The early warning results provide a scientific decision-making basis for mining enterprises, which helps to reasonably arrange mining plans, strengthen support measures or take other necessary preventive measures; timely detection and early warning of rock instability risks help avoid or reduce the occurrence of ground pressure disasters and ensure the safe conduct of mining production.

[0119] In a specific embodiment of the present invention, let X 、 Y and Z are three different observation sequences, representing stress time series, acoustic emission time series and displacement time series respectively, which together constitute the three observation sequences in the nonlinear dynamic model of rock mass failure and instability ground pressure disaster. Considering the different dimensions and magnitudes of the observation sequences X, Y, and Z, they should be standardized first. Generally, the average value of the observation sequence can be used as the standardization factor. According to the physical properties of the system, f i Assume that it is a nonlinear function. For the ground pressure disaster prediction system, it is set to the general form mentioned above

[66] ~

[68] , and the nonlinear dynamic mathematical model equation is set to the following form:

[0120]

[0121]

[0122] Where: a1, a2, ... a9, b 1, b2…,b9 and c 1,c2…,c9 are constants to be determined by inversion, which can be determined by inversion theory. Substitute the parameters into the above formula to obtain the nonlinear dynamic prediction results of ground pressure monitoring y warning; derive the latest prediction results from the deep well ground pressure monitoring and early warning system for nonlinear dynamic disasters, including but not limited to the predicted value of rock mass instability time, the change trend data of acoustic emission activities, stress change data and displacement change data; check the specific time point or time period of possible rock mass instability given in the prediction report, evaluate its urgency and accuracy, use charts or statistical software to draw the trend of acoustic emission event rate and amplitude, analyze the frequency, intensity and duration of acoustic emission activities, judge the development of internal damage in the rock mass, draw stress change curves, observe the concentration, release and fluctuation of stress; analyze the relationship between stress changes and mining activities, geological structures and other factors, and identify possible stress abnormal areas. Calculate and draw displacement change rate diagrams, observe the overall deformation trend and local deformation characteristics of the rock mass, analyze the relationship between displacement change rate and rock mass stability, and judge whether there are signs of rapid deformation or sudden displacement. Retrieve the rock mass stability monitoring data under similar geological conditions and mining conditions in the past from the database, and align the historical data with the current prediction results in terms of time scale and monitoring parameters. Compare the predicted value of the rock mass instability time with the historical instability records, compare the key parameters such as the change trend of acoustic emission activities, stress change characteristics and displacement change rate, and use statistical methods to quantify the differences and significance of the comparison results. Based on the comparison results, comprehensively evaluate the current stability state of the rock mass; determine whether the rock mass is in a high-risk state, that is, whether it is close to or has reached the critical conditions before historical instability. According to the stability assessment results, clarify the warning level and warning information, and formulate a detailed warning report, including the warning area, warning level, expected instability time range, possible impact range and recommended preventive measures.

[0123] The embodiment of the present invention further provides a nonlinear dynamic disaster deep well ground pressure monitoring and early warning system, comprising:

[0124] The acquisition module is used to acquire acoustic emission signals, stress change data and displacement change data; preprocess the collected raw data to obtain preprocessed data; extract the acoustic emission event rate, acoustic emission amplitude, stress change and displacement change from the preprocessed data to obtain key parameters; select state variables and control variables according to the catastrophe theory, analyze the variables and key parameters to obtain analysis results;

[0125] The processing module is used to predict the instability time of the rock mass according to the analysis results to obtain the prediction results; according to the prediction results, simulate the process of the rock mass deformation to instability to obtain the predicted rock mass instability time; use the acoustic emission time series, stress time series and displacement time series as observation sequences, describe the dynamic relationship between the observation sequences, and use the inversion theory to determine the inversion parameters; evaluate the stability of the rock mass according to the prediction results, the predicted rock mass instability time and the inversion parameters to obtain the early warning results.

[0126] It should be noted that the system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0127] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0128] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0129] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A nonlinear dynamic disaster deep well ground pressure monitoring and early warning method, characterized in that: The method comprises: Acquire acoustic emission signals, stress change data and displacement change data; Preprocessing the collected raw data to obtain preprocessed data; The key parameters are obtained by extracting the acoustic emission event rate, acoustic emission amplitude, stress change and displacement change from the pre-processed data; According to the mutation theory, state variables and control variables are selected, and the variables and key parameters are analyzed to obtain the analysis results; According to the analysis results, the instability time of the rock mass is predicted to obtain the prediction result; According to the prediction results, the process from rock mass deformation to instability is simulated to obtain the predicted rock mass instability time; The acoustic emission time series, stress time series and displacement time series are taken as observation sequences, the dynamic relationship between the observation sequences is described, and the inversion theory is used to determine the inversion parameters; Based on the prediction results, predicted rock mass instability time and inversion parameters, the stability of the rock mass is evaluated to obtain early warning results.

2. The nonlinear dynamic disaster deep well ground pressure monitoring and early warning method according to claim 1 is characterized in that: Acquire acoustic emission signals, stress change data, and displacement change data, including: Configure the data acquisition software based on the installation of acoustic emission sensors, strain gauges and displacement sensors; The acoustic emission signal, stress change data and displacement change data are monitored in real time through the data acquisition software to obtain the acoustic emission signal, stress change data and displacement change data.

3. The method for monitoring and early warning of deep well ground pressure for nonlinear dynamic disasters according to claim 2 is characterized in that: The acoustic emission event rate, acoustic emission amplitude, stress change and displacement change are extracted from the pre-processed data to obtain key parameters, including: Setting a threshold of the preprocessed acoustic emission signal, and identifying an acoustic emission event according to the threshold of the acoustic emission signal to obtain an acoustic emission event rate; The preprocessed acoustic emission signal is segmented, and the peak amplitude of each signal segment is obtained to obtain the acoustic emission amplitude; Determine the stress reference value according to the preprocessed stress data, and calculate the stress change at each time point according to the reference value to obtain the stress change; According to the preprocessed displacement data, a reference value of the displacement is determined, and the displacement change at each time point is calculated by using the reference value to obtain the displacement change.

4. The nonlinear dynamic disaster deep well ground pressure monitoring and early warning method according to claim 3 is characterized in that: According to the mutation theory, state variables and control variables are selected, and the variables and key parameters are analyzed to obtain the analysis results, including: According to the catastrophe theory, state variables and control variables are selected. The state variables are the deformation degree or stress state of the rock mass, and the control variables are the external factors that affect the system behavior, such as loading conditions and material properties. According to the state variables and control variables, through X = a0 + a1t + a2t 2 +a3t 3 +a4t 4 The state variables are expressed as a function of time, where a i is the unknown coefficient, t is the time, and X is the state variable; As a function of time, Perform variable substitution and set To obtain the equilibrium equation Z 3 +aZ+b=0; According to the equilibrium equation, the bifurcation set equation is calculated as 4a 3 +27b 2 = 0, to obtain the discriminant D = 4a 3 +27b 2 , a i is the coefficient in the Taylor expansion, b i The transformed coefficients, q is the adjustment term used to simplify the Taylor expansion, a and b are control variables, D is the bifurcation discriminant, ΔZ and Δt are used to describe the sudden change of state variables and their corresponding time changes; The analytical stability is obtained according to the discriminant to obtain the analytical results.

5. The nonlinear dynamic disaster deep well ground pressure monitoring and early warning method according to claim 4 is characterized in that: According to the analysis results, the instability time of the rock mass is predicted to obtain the prediction results, including: According to the analysis results, the original monitoring data of the rock mass is obtained, which is usually time series data, such as displacement, stress or deformation data of the rock mass; pass The original data is accumulated once, where X (0) (i)>0, i is the number of monitoring data with the same interval time, X (0) (i) is the original monitoring data, (i) is the one-time accumulation generated data, which is used to reduce the randomness of the data; Generate data based on one accumulation, through Z (1) (i) = (X (1) (i)+X (1) (i-1))÷2 calculates the mean data, Z (1) (i) is the mean generated data; By defining the differential equation, Solve once to generate data X (1) The expression of (t) is Among them, when t takes a sequence number, the above formula becomes: Where i = 1, 2, ..., n; a^ and u are expressed as By defining the matrix B and vector Y, And according to the matrix B and vector Y, through generate and u parameters, is the attenuation coefficient of the system, and u is the external input or driving term; Generate data and mean data by one accumulation, through X (0) (i) = X (1) (i)-X (1) (i-1) and Get the predicted value of the original data; According to the predicted value, the critical value S is set, and the The time t of the expected failure event is calculated to obtain the predicted rock mass instability time, where S is the critical value and Δt is the time interval of the data.

6. The method for monitoring and early warning of deep well ground pressure for nonlinear dynamic disasters according to claim 5 is characterized in that: According to the prediction results, the process from rock mass deformation to instability is simulated to obtain the predicted rock mass instability time, including: According to the prediction results, the original monitoring data sequence is transformed by accumulation and fitting into the Verhulst first-order whitening nonlinear differential equation to obtain the Verhulst nonlinear differential dynamic prediction equation for rock mass instability and ground pressure disaster. The nonlinear differential dynamic prediction equation is solved to predict the instability time, and the predicted instability time is converted into actual time to obtain the actual predicted instability time.

7. The nonlinear dynamic disaster deep well ground pressure monitoring and early warning method according to claim 6 is characterized in that: The acoustic emission time series, stress time series and displacement time series are used as observation sequences to describe the dynamic relationship between the observation sequences, and the inversion theory is used to determine the inversion parameters, including: Collect acoustic emission time series, stress time series and displacement time series data; Analyze the correlation between acoustic emission, stress and displacement time series to obtain analysis results; Based on the analysis results, the dynamic relationship between the observation sequences is established to obtain the characteristics of the parameter data; According to the characteristics of the data, the observed data are compared with the model prediction results, and the model parameters are continuously adjusted through an iterative optimization process until a certain convergence criterion is met or a predetermined number of iterations is reached to determine the inversion parameters.

8. A nonlinear dynamic disaster deep well ground pressure monitoring and early warning system, characterized in that: include: An acquisition module, used to acquire acoustic emission signals, stress change data and displacement change data; Preprocessing the collected raw data to obtain preprocessed data; The key parameters are obtained by extracting the acoustic emission event rate, acoustic emission amplitude, stress change and displacement change from the pre-processed data; According to the mutation theory, state variables and control variables are selected, and the variables and key parameters are analyzed to obtain the analysis results; The processing module is used to predict the instability time of the rock mass according to the analysis results to obtain the prediction results; according to the prediction results, simulate the process of the rock mass deformation to instability to obtain the predicted rock mass instability time; use the acoustic emission time series, stress time series and displacement time series as observation sequences, describe the dynamic relationship between the observation sequences, and use the inversion theory to determine the inversion parameters; evaluate the stability of the rock mass according to the prediction results, the predicted rock mass instability time and the inversion parameters to obtain the early warning results.

9. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images