A method for mine earthquake location by combining chaos mapping with t-distributed variable optimization

Abstract

The application discloses a mine earthquake positioning method based on a fusion of a chaos mapping and t-distribution variation, and a mine earthquake positioning method based on a fusion of a chaos mapping and t-distribution variation.The steps are as follows: 1, setting up a microseismic monitoring station, monitoring and collecting data in real time; 2, judging according to a mine earthquake triggering condition through the received vibration wave signal in step 1; when the triggering condition is reached, it is determined that a mine earthquake occurs, and step 3 is performed; if the triggering condition is not reached, the monitoring is continuously performed; 3, when it is determined that a mine earthquake occurs, the vibration wave signal received by the triggered microseismic monitoring station is intercepted, and the station number and P-wave arrival time information are introduced into the mine earthquake positioning method based on a fusion of a chaos mapping and t-distribution variation to determine the position of the earthquake source. Through the method, the population initialization of the Tent chaos mapping and the adaptive t-distribution variation mechanism are improved, and the mine earthquake positioning precision and the positioning result stability are improved.

Description

Technical Field

[0001] This invention belongs to the field of mine seismic monitoring and location, and particularly relates to a dung beetle-optimized seismic location method that integrates chaotic mapping and t-distribution variation. Background Technology

[0002] Mining tremors are earthquakes induced by mining activities. Their main cause is the disruption of underground rock structure stability during mining operations, leading to energy release, ground vibrations, or underground damage, thus triggering disasters.

[0003] Mine-induced tremors are accompanied by rock mass fracturing and energy release, generating seismic waves. Microseismic monitoring stations can monitor these waves in real time and locate the events using various positioning algorithms. With the development of geophysics, scholars both domestically and internationally, based on seismological research, have proposed various positioning algorithms applicable to large-scale earthquakes, capable of accurately calculating the time, spatial location, and intensity of earthquakes. Compared to conventional earthquakes, mine-induced tremors are typically small-scale, generally confined to a few to tens of square kilometers within the mining area, and occurring at depths typically less than 1500 meters.

[0004] Despite the various optimization methods proposed by scholars for mine seismic location, the current location accuracy and stability still have certain shortcomings. In particular, traditional location methods have defects such as limited search space, easy to get trapped in local optima, and low robustness, which lead to decreased location accuracy, unstable location results, and difficulty in meeting the needs of actual field applications. Summary of the Invention

[0005] The purpose of this invention is to provide a dung beetle-optimized seismic location method that integrates chaotic mapping and t-distribution variation, thereby solving the technical problems of unstable location results and low location accuracy in the existing technology.

[0006] To achieve the above objectives, the technical solution adopted in this invention is: a dung beetle-optimized mine seismic location method that integrates chaotic mapping and t-distribution variation, the steps of which are as follows: Step 1: Set up microseismic monitoring stations for real-time monitoring and data acquisition. The specific method involves jointly deploying microseismic monitoring stations on the surface, in deep boreholes, and underground for real-time monitoring and data acquisition. A combined surface-well-ground deployment approach is used, with stations deployed at surface, deep borehole, and underground locations. During deployment, stations are avoided from being located on the same straight line or plane, creating a three-dimensional monitoring network that surrounds the underground mining area, forming a complete three-dimensional monitoring network. The acquired data includes station numbers, station coordinate information, seismic wave velocity, population size in the dung beetle-optimized seismic location method, and search dimensions.

[0007] Step 2: Based on the vibration wave signal received in Step 1, determine the mine vibration triggering conditions; when the triggering conditions are met, it is determined that a mine vibration has occurred, and proceed to Step 3; if the triggering conditions are not met, continue monitoring. The triggering conditions for a mine earthquake are: a. Results of determining the short and long time windows of seismic waves: The ratio of the average energy of the short time window to the average energy of the long time window is greater than the threshold; b. Threshold value: The maximum amplitude of the energy wave is greater than 100mV; c. Number of stations triggered per unit time: The number of stations triggered is greater than or equal to 4; When a, b, and c all meet the criteria, it is determined that a mine tremor has occurred.

[0008] Step 3: When it is determined that a mine tremor has occurred, the seismic wave signal received by the triggering microseismic monitoring station is intercepted, and the station number and P-wave arrival time information are imported into the dung beetle optimized mine tremor location method to determine the location of the seismic source. Step 3.1) Initialize algorithm parameters and population distribution; Step 3.1.1) Initialize basic algorithm parameters: Set the dung beetle population size N, the maximum number of iterations T, and the search space dimension. ; Step 3.1.2) Population initialization based on improved Tent chaotic map: The initial population position is generated using the Tent map to avoid uneven distribution of the initial population in the search space. The calculation formula is as follows: (1) in, This represents the chaotic variable generated in the i-th iteration. This represents the chaotic variable generated in the next iteration. The generated chaotic sequence is mapped to the source search region by representing a uniformly distributed random number within the interval [0,1]. Complete the initial population construction; Step 3.2) Calculate the objective function value and identify key individuals; Step 3.2.1) Construct the Time Difference of Arrival (TDOA) objective function: Using the monitoring station locations and P-wave observation time data, construct an objective function based on the sum of squared residuals. This is used to measure the degree of match between the search point and the actual earthquake source: (2) in, When observed by station i, v is the wave velocity. The moment of the earthquake; Step 3.2.2) Evaluate individual fitness: Calculate the objective function value for all dung beetle individuals in the population, and identify and update the current global optimum position based on the fitness value. and the worst position in the world ; Step 3.3) Location update based on dung beetle foraging behavior and adaptive mutation; Step 3.3.1) Update the rolling ball and obstacle behavior position: Simulate the dung beetle's ball-rolling behavior guided by light, adjusting the position through a controlled perturbation coefficient; if an obstacle is encountered, use a tangent function to simulate dancing behavior to re-perceive direction, calculated as follows: (3) in, This represents the position of the i-th dung beetle individual in the (t+1)-th iteration. This represents the position of the i-th dung beetle individual at the t-th iteration. This is the deviation coefficient. Let represent the position of the i-th dung beetle individual at the t-th iteration, k be the perturbation factor, and b be a constant; Step 3.3.2) Perform reproductive and stealing behavior updates: Simulate the reproductive behavior of dung beetles near the optimal region and their stealing behavior from other individuals; reduce the search radius through a dynamic boundary R, ​​so that the population moves toward the global optimum in the later stages of iteration. High-efficiency aggregation; Step 3.3.3) Introduce an adaptive t-distribution mutation mechanism: To avoid the algorithm getting stuck in local optima in the later stages of iteration, the current global optimum position is... Adaptive t-distribution perturbation is applied, utilizing the dynamic degrees of freedom of the t-distribution as it changes with the number of iterations to balance the algorithm's global exploration and local exploitation capabilities. (4) in, These are t-distributed random numbers with the number of iterations as their degrees of freedom. This is an adaptive step size factor; Step 3.3.4) Boundary check and optimal retention: Perform boundary constraint processing on the updated position. If the fitness of the mutated individual is better than that of the original optimal individual, then perform position replacement. Step 3.4) Iterate through optimization and output the positioning results; Step 3.4.1) Iterative update: Repeat steps 3.2) and 3.3). The population gradually approaches the real earthquake source coordinates through global coverage of chaotic initialization and dynamic optimization of t-distribution variation. Step 3.4.2) Output the optimal solution: When the maximum number of iterations is reached, stop the calculation and output the position information of the globally optimal individual. and the time of its occurrence This serves as the final result for locating the earthquake source.

[0009] The beneficial effects of this invention are as follows: This invention utilizes seismic waves generated from various angles by microseismic monitoring stations distributed at multiple locations on the ground, in deep boreholes, and underground. Then, it employs a dung beetle-optimized seismic location method that integrates chaotic mapping and t-distribution variation. By improving the population initialization of the Tent chaotic mapping and the adaptive t-distribution variation mechanism, the accuracy and stability of seismic location results are enhanced. Attached Figure Description

[0010] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0011] The specific embodiments of the present invention will now be described in detail with reference to examples and accompanying drawings.

[0012] The steps of the dung beetle-optimized mine seismic location method, which integrates chaotic mapping and t-distribution variation, are as follows: Step 1) Real-time monitoring is carried out by deploying microseismic monitoring stations in combination with ground, deep borehole and underground stations; A combined well-ground deployment approach is adopted, with microseismic monitoring stations deployed at the surface, deep boreholes, and underground locations. The stations are arranged to avoid being on the same straight line or plane, ensuring a three-dimensional monitoring network that maximizes the coverage of the underground mining area and forms a complete three-dimensional monitoring network. This network monitors energy waves within its coverage area in real time, focusing on areas with shock hazard to improve early warning capabilities and location accuracy for potential mine tremors.

[0013] There should be more than 4 microseismic monitoring stations. The more stations there are within a certain range, the higher the positioning accuracy. For a range of 5km×5km×1km, we generally take 6-50 stations.

[0014] Step 2) Import the dung beetle-optimized mine seismic location method that integrates chaotic mapping and t-distribution variation, and the initial information of microseismic monitoring stations into the system; Import the station number, station coordinate information, seismic wave velocity, population size, search dimension, and maximum number of iterations in the dung beetle-optimized seismic location method into the system for algorithm calculation.

[0015] Step 3) Monitor seismic wave signals in real time; The seismic wave signals received by various microseismic monitoring stations are monitored in real time to determine the triggering conditions for mine tremors. These conditions include the results of the long and short time windows of the seismic waves (whether the ratio of the average energy of the short time window to the average energy of the long time window is greater than the threshold), the threshold value (the maximum amplitude of the energy wave is greater than 100mV), and the number of stations triggered per unit time (the number of stations triggered is greater than or equal to 4). When all the triggering conditions are met, it is determined that a mine tremor has occurred; if the triggering conditions are not met, monitoring continues.

[0016] Step 4) When it is determined that a mine tremor has occurred, import the station number that meets the triggering conditions and the arrival time of the P-wave; When a mine tremor is detected, the seismic wave signal received by the triggering microseismic monitoring station is intercepted, and the station number and P-wave arrival time information are imported into the dung beetle-optimized mine tremor location method that integrates chaotic mapping and t-distribution variation to further determine the location of the seismic source.

[0017] Step 5) Initialize algorithm parameters and population distribution; Step 5.1) Initialize basic algorithm parameters: Set the dung beetle population size, maximum number of iterations, and search space dimension. ; Step 5.2) Population initialization based on improved Tent chaotic map: The Tent map is used to generate initial population positions with high ergodicity and uniformity, avoiding initial population clustering in space. The calculation formula is as follows: (1) in, This represents the chaotic variable generated in the i-th iteration. This represents the chaotic variable generated in the next iteration. The generated chaotic sequence is mapped to the source search region by representing a uniformly distributed random number within the interval [0,1]. Complete the initial population construction; Step 6) Calculate the objective function value and identify key individuals; Step 6.1) Constructing the Time Difference of Arrival (TDOA) Objective Function: Using the monitoring station locations and P-wave observation time data, construct an objective function based on the sum of squared residuals. This is used to measure the degree of match between the search point and the actual earthquake source: (2) in, When observed by station i, v is the wave velocity. The moment of the earthquake; Step 6.2) Evaluate individual fitness: Calculate the objective function value for all dung beetle individuals in the population, and identify and update the current global optimum position based on the fitness value. and the worst position in the world ; Step 7) Position update based on dung beetle foraging behavior and adaptive mutation; Step 7.1) Update the rolling ball and obstacle behavior position: Simulate the dung beetle's ball-rolling behavior guided by light, adjusting the position using a controlled perturbation coefficient; if an obstacle is encountered, use a tangent function to simulate dancing behavior to re-perceive direction, calculated as follows: (3) in, This represents the position of the i-th dung beetle individual in the (t+1)-th iteration. This represents the position of the i-th dung beetle individual at the t-th iteration. This is the deviation coefficient. Let represent the position of the i-th dung beetle individual at the t-th iteration, k be the perturbation factor, and b be a constant; Step 7.2) Perform reproductive and stealing behavior updates: Simulate the dung beetle's reproductive behavior near the optimal region and its stealing behavior from other individuals. Reduce the search radius using a dynamic boundary R, ​​allowing the population to move towards the global optimum in the later stages of iteration. High-efficiency aggregation; Step 7.3) Introduce an adaptive t-distribution mutation mechanism: To avoid the algorithm getting stuck in local optima in the later stages of iteration, the current global optimum position is... Adaptive t-distribution perturbation is applied, utilizing the dynamic degrees of freedom of the t-distribution as it changes with the number of iterations to balance the algorithm's global exploration and local exploitation capabilities. (4) in, These are t-distributed random numbers with the number of iterations as their degrees of freedom. It is an adaptive step size factor that decreases as iterations increase to improve positioning accuracy in later stages; Step 7.4) Boundary check and optimal retention: Perform boundary constraint processing on the updated position. If the fitness of the mutated individual is better than that of the original optimal individual, then perform position replacement. Step 8) Iterate through optimization and output the positioning results; Step 8.1) Iterative update: Repeat steps 6) and 7). The population gradually approaches the true source coordinates through global coverage of chaotic initialization and dynamic optimization of t-distribution variation. 8.2) Output the optimal solution: When the maximum number of iterations is reached, stop the calculation and output the position information of the globally optimal individual. and the time of its occurrence This serves as the final result for locating the earthquake source.

[0018] Example 1: A coal mine has reached a mining depth of nearly 1000 meters, and has experienced multiple mine-induced seismic events during the mining process, with vibrations clearly felt on the surface. To enhance monitoring, the mine has installed seismic monitoring stations using a combined underground and surface deployment method. The underground stations use fiber optic transmission of ground GPS signals for time synchronization, while the surface and deep borehole stations rely on GPS for unified time synchronization.

[0019] Step 1: Deploy microseismic monitoring stations to monitor mine seismic signals in real time; Microseismic monitoring stations are deployed at ground level, deep boreholes, and underground locations. During the deployment process, multiple stations are avoided from being in the same straight line or plane to ensure a three-dimensional structure and to fully surround the underground mining area as much as possible, forming a three-dimensional deployment network. The stations are located as follows: T1 underground station (6738m, 7965m, -842m), T2 underground station (8126m, 7406m, -1034m), T3 underground station (7386m, 7795m, -863m), T4 underground station (7790m, 8277m, -962m), T5 deep hole station (6690m, 8177m, -441m), T6 deep hole station (7802m, 7336m, -510m), and T7 surface station (7792m, 7825m, 41m). After the deployment is completed, the energy waves in the coverage area will be monitored in real time, and areas assessed as having a risk of impact will be subject to key monitoring.

[0020] Step 2: Import the dung beetle optimization localization method that integrates chaotic mapping and t-distribution variation, along with the initial information of the microseismic monitoring stations, into the system; Import microseismic monitoring station numbers T1-T7, location coordinates, seismic wave velocity V=3850m / s into the system. The population size of the dung beetle-optimized mine seismic location method, which integrates chaotic mapping and t-distribution variation, is 100, the search dimension is 3, and the maximum number of iterations is 1000.

[0021] Step 3: Real-time monitoring of seismic wave signals; The seismic wave signals received by microseismic monitoring stations distributed in various locations are monitored in real time to determine the triggering conditions of mine tremors. These conditions include the results of the short-time window method for determining the mine tremor wavelength (the triggering conditions of the long-short-time window method), the threshold value (the maximum amplitude of the energy wave is greater than 100mV), and the number of stations triggered per unit time (the number of stations triggered is greater than or equal to 4). When the triggering conditions are met, it is determined that a mine tremor has occurred; if the triggering conditions are not met, monitoring continues.

[0022] Step 4: When it is determined that a mine tremor has occurred, import the microseismic monitoring station number, coordinate information, and P-wave arrival time information into the system; The stations perform real-time detection of vibration wave signals. After the mine earthquake simulation experiment begins, all stations receive high-energy wave signals, which indicates that a mine earthquake has occurred. The station coordinates and P-wave arrival time information are then input into the dung beetle-optimized mine earthquake location method that integrates chaotic mapping and t-distribution variation. The P-wave arrival time information is shown in Table 1.

[0023] Table 1. Coordinates of Microseismic Monitoring Stations and P-wave Arrival Times

[0024] Step 5: Initialize the parameters of the dung beetle optimization mine seismic location method; (1) Initialize the population size to 100, the search dimension to 3, the maximum number of iterations to 1000, and the initial position and velocity of each population; (2) The Time Difference of Arrival (TDOA) is defined as the objective function to evaluate the spatial relationship between particles and stations. Dynamic optimization is performed by improving the population initialization and adaptive t-distribution mutation mechanism of the Tent chaotic mapping to optimize the accuracy of source localization.

[0025] Step 6: Calculate the objective function values ​​for all populations and determine the optimal individual location. Taking the first iteration as an example; (1) Calculate the difference matrix of microseismic monitoring stations; A difference matrix is ​​constructed using the location coordinates of microseismic monitoring stations and P-wave arrival time data. This matrix measures the distance difference between the station and the seismic source. It is calculated using an auxiliary objective function. (2) Calculate the particle position difference matrix; For each particle location, the distance difference between it and the microseismic monitoring station is calculated to obtain the difference matrix. Then the difference matrix and By comparing the results, the fitness of the particles relative to the target seismic source position can be determined; (3) Calculate the objective function value; By calculating the TDOA (Temporal Directional Occurrence Address), the objective function value for mine seismic location is constructed. Tent mapping is used to generate initial population locations with high ergodicity and uniformity, avoiding spatial clustering of the initial population and thus enhancing the accuracy and robustness of the TDOA calculation. The formula for calculating the objective function value is: (5) in, Matrix difference , This is the median of the difference matrix.

[0026] (4) Select the particle position with the smallest fitness value as the optimal particle position. The value is (7638, 7983, -677). Step 7: Continuously update the optimal population location; Repeat step 6 to continuously update the optimal population location.

[0027] Step 8: Determine the epicenter; As the number of iterations increases, the algorithm dynamically optimizes the source location objective function, gradually bringing the search results closer to the actual source location. When the maximum number of iterations (1000) is reached, the source location calculation is complete, and the optimal virtual source location is output. The final epicenter location is (7414m, 7958m, -709m) on the roof above the goaf. Analysis suggests that this seismic event was mainly caused by the fracturing of the hard, thick roof at -709m above the goaf due to mining activities. The location is consistent with the hard fine sandstone layer above the mining area, and the fracturing of this layer due to mining provided the energy for this seismic event.

Claims

1. A dung beetle-optimized mine seismic location method integrating chaotic mapping and t-distribution variation, characterized in that, The steps are as follows: Step 1: Set up microseismic monitoring stations to monitor and collect data in real time; Step 2: Based on the vibration wave signal received in Step 1, determine the mine vibration triggering conditions; when the triggering conditions are met, it is determined that a mine vibration has occurred, and proceed to Step 3; if the triggering conditions are not met, continue monitoring. Step 3: When it is determined that a mine tremor has occurred, the seismic wave signal received by the triggering microseismic monitoring station is intercepted, and the station number and P-wave arrival time information are imported into the dung beetle optimized mine tremor location method to determine the location of the seismic source. Step 3.1) Initialize algorithm parameters and population distribution; Step 3.1.1) Initialize basic algorithm parameters: Set the dung beetle population size N, the maximum number of iterations T, and the search space dimension. This indicates the search dimensions for the problem of earthquake source location; Step 3.1.2) Population initialization based on improved Tent chaotic map: The initial population position is generated using the Tent map to avoid uneven distribution of the initial population in the search space. The calculation formula is as follows: (1) in, This represents the chaotic variable generated in the i-th iteration. This represents the chaotic variable generated in the next iteration. The generated chaotic sequence is mapped to the source search region by representing a uniformly distributed random number within the interval [0,1]. Complete the initial population construction; Step 3.2) Calculate the objective function value and identify key individuals; Step 3.2.1) Construct the Time Difference of Arrival (TDOA) objective function: Using the monitoring station locations and P-wave observation time data, construct an objective function based on the sum of squared residuals. This is used to measure the degree of match between the search point and the actual earthquake source: (2) in, When observed by station i, v is the wave velocity. The moment of the earthquake; Step 3.2.2) Evaluate individual fitness: Calculate the objective function value for all dung beetle individuals in the population, and identify and update the current global optimum position based on the fitness value. and the worst position in the world ; Step 3.3) Location update based on dung beetle foraging behavior and adaptive mutation; Step 3.3.1) Update the rolling ball and obstacle behavior position: Simulate the dung beetle's ball-rolling behavior guided by light, adjusting the position through a controlled perturbation coefficient; if an obstacle is encountered, use a tangent function to simulate dancing behavior to re-perceive direction, calculated as follows: (3) in, This represents the position of the i-th dung beetle individual in the (t+1)-th iteration. This represents the position of the i-th dung beetle individual at the t-th iteration. This is the deviation coefficient. Let represent the position of the i-th dung beetle individual at the t-th iteration, k be the perturbation factor, and b be a constant; Step 3.3.2) Perform reproductive and stealing behavior updates: Simulate the reproductive behavior of dung beetles near the optimal region and their stealing behavior from other individuals; reduce the search radius through a dynamic boundary R, ​​so that the population moves toward the global optimum in the later stages of iteration. High-efficiency aggregation; Step 3.3.3) Introduce an adaptive t-distribution mutation mechanism: To avoid the algorithm getting stuck in local optima in the later stages of iteration, the current global optimum position is... Adaptive t-distribution perturbation is applied, utilizing the dynamic degrees of freedom of the t-distribution as it changes with the number of iterations to balance the algorithm's global exploration and local exploitation capabilities. (4) in, These are t-distributed random numbers with the number of iterations as their degrees of freedom. This is an adaptive step size factor; Step 3.3.4) Boundary check and optimal retention: Perform boundary constraint processing on the updated position. If the fitness of the mutated individual is better than that of the original optimal individual, then perform position replacement. Step 3.4) Iterate through optimization and output the positioning results; Step 3.4.1) Iterative update: Repeat steps 3.2) and 3.3). The population gradually approaches the real earthquake source coordinates through global coverage of chaotic initialization and dynamic optimization of t-distribution variation. Step 3.4.2) Output the optimal solution: When the maximum number of iterations is reached, stop the calculation and output the position information of the globally optimal individual. and the time of its occurrence This serves as the final result for locating the earthquake source.

2. The dung beetle-optimized mine seismic location method according to claim 1, which integrates chaotic mapping and t-distribution variation, is characterized in that... In step 1, the specific method is as follows: microseismic monitoring stations are jointly deployed on the ground, in deep boreholes, and underground to conduct real-time monitoring and collect data. The well-ground joint deployment method is adopted, and the microseismic monitoring stations are deployed on the ground, in deep boreholes, and underground respectively. During the deployment, the stations are avoided to be located on the same straight line or the same plane, so that the monitoring network forms a three-dimensional structure that surrounds the underground mining area and forms a complete three-dimensional monitoring network.

3. The dung beetle-optimized mine seismic location method according to claim 1, which integrates chaotic mapping and t-distribution variation, is characterized in that... In step 1, the collected data includes the imported station number, station coordinate information, seismic wave velocity, population size in the dung beetle-optimized seismic location method, and search dimension.

4. The dung beetle-optimized mine seismic location method according to claim 1, which integrates chaotic mapping and t-distribution variation, is characterized in that... In step 2, the triggering condition for the mine tremor is: a. Results of determining the short and long time windows of seismic waves: The ratio of the average energy of the short time window to the average energy of the long time window is greater than the threshold; b. Threshold value: The maximum amplitude of the energy wave is greater than 100mV; c. Number of stations triggered per unit time: The number of stations triggered is greater than or equal to 4; When a, b, and c all meet the criteria, it is determined that a mine tremor has occurred.

Images