Micro-seismic source positioning method and system based on improved particle swarm optimization algorithm

Abstract

The invention discloses a micro-seismic source positioning method and system based on an improved particle swarm optimization algorithm, and the method comprises the steps: arranging n sensors in an open-pit mine field, solving a minimum value of a target function in a definition domain range based on the improved particle swarm optimization algorithm, and determining the coordinates of a source; according to the improved particle swarm optimization algorithm, mixed chaotic mapping combining Logistic mapping and Tent mapping is introduced in the initial population generation stage, so that a more regular and uniform initial population is generated; the particle fitness is sequenced after each iteration is completed, the population with the fitness lower than an average value is called as a dominant population, the population with the fitness higher than the average value is called as an inferior population, a self-adaptive weight strategy is adopted for the dominant population during next iteration, and a Levy flight combined self-adaptive weight strategy is adopted for the inferior population. According to the invention, the stability and precision of the positioning algorithm are effectively improved, and a new idea is provided for micro-seismic positioning.

Description

Technical Field

[0001] This invention belongs to the field of open-pit mine safety prediction technology, and relates to a microseismic source location method and system, specifically to a microseismic source location method and system based on an improved particle swarm optimization algorithm with chaotic mapping initialization and hybrid update strategies. Background Technology

[0002] During mining operations, rocks are prone to fracture, leading to landslides, collapses, and other accidents, causing personal injury and property damage (References 1-3). The formation, propagation, and connection of microcracks in rock masses are often precursors to rock fracturing. Under external stress, the formation of microcracks within rocks is usually accompanied by the excitation of stress waves or elastic waves, which are effective acoustic emission signal sources. By analyzing these acoustic emission signals, the development of rock cracks can be studied, and the characteristics of precursors to rock mass instability can be explored, thereby preventing potential rock mass dynamic disasters (References 4-5).

[0003] Microseismic monitoring technology utilizes stress waves or elastic waves released during rock deformation and fracturing to monitor the stability of rock mass engineering. As a non-destructive method for monitoring the stability of underground rock masses, it has been widely used in recent years in fields such as mining, oilfield development, and tunnel construction (References 6-8). This technology collects acoustic emission signals during rock fracturing, identifies and analyzes these signals, and then determines the development trend of microcracks in the rock, providing early warnings when necessary.

[0004] For a long time, improving the accuracy and precision of microseismic source location has been an important research direction for scholars both at home and abroad. Based on different location principles, location methods are mainly divided into two categories: waveform-based methods and time-of-arrival (TOA) methods (Reference 9). TOA methods are favored due to their fast location speed and simple equipment layout. In this method, the location accuracy depends on how the objective function for the source location is solved. To transform the nonlinear problem into a linear problem, Dong et al. simplified the sensor array into a cuboid and obtained analytical solutions for acoustic emission location under two cuboid monitoring networks when the wave velocity structure was unknown (Reference 10). To reduce the influence of anomalous data on the location results, Luo et al. proposed a Bayesian location method that uses variable weights to connect P-wave and S-wave arrival time data (Reference 11), which significantly improves the location accuracy compared to Bayesian location methods using only P-wave or only S-wave. With the popularization of computer technology, iterative algorithms have continuously developed and become mainstream, such as the Geiger method, simplex method, least squares method, and various intelligent optimization algorithms (References 12-14). The classic Geiger location method minimizes the time residual through iteration (Reference 15). Buland et al. optimized it to make it more suitable for source location (Reference 16). Lin et al. combined the Geiger method with linear location methods (Reference 17), improving the efficiency of iterative solutions in the Geiger location method. Prugger et al. introduced the simplex algorithm into source location (Reference 18) and compared it with the then-popular Geiger location algorithm, showing that the simplex algorithm had higher accuracy. Dong et al. eliminated the location error caused by wave velocity deviation in the MS / AE monitoring system and proposed a microseismic source location method that does not require wave velocity prediction (References 19-20). Liao et al. proposed the NM-PSO algorithm (Reference 21), which combines the advantages of the simplex algorithm and particle swarm optimization algorithm, improving the convergence speed and stability of the location algorithm while preventing the model from getting trapped in local optima. Xiao et al. eliminated the location error caused by improper parameter settings in the nonlinear optimal location method by combining the advantages of the nonlinear optimal location method and the PSO algorithm (Reference 22). To reduce the impact of anomalous arrival data on the accuracy of seismic source location, Lagos et al. used VFSA, PSO, and GS algorithms to locate seismic sources during hydraulic fracturing (Reference 23), finding that VFSA and PSO algorithms were more efficient in both 2D and 3D scenarios. Zhou et al. evaluated the impact of PSO and GA algorithms on VFOM performance (Reference 24), showing that PSO can provide better location accuracy and convergence speed compared to GA. Furthermore, some variants or hybrid PSO algorithms, such as HPSO (Reference 25), AEPSO (Reference 26), and hybrid PSO (Reference 27), have also shown improved optimization performance.

[0005] Although previous researchers have made numerous improvements to the particle swarm optimization algorithm and achieved excellent optimization performance, in practical geotechnical engineering applications, due to complex geological conditions, the initial population distribution becomes particularly important. A uniformly distributed population significantly enhances the algorithm's optimization ability. Furthermore, choosing an appropriate location update strategy can also help the algorithm escape local optima and strengthen its later optimization capabilities.

[0006] References [Document 1]Chen Y, Xiao P, Li P, et al. Formation mechanism of rockburstin deep tunnel adjacent to faults: Implication from numerical simulation and microseismic monitoring[J]. Journal of Central South University, 2022, 29(12): 4035-4050. [Document 2]Lasocki S, Orlecka-Sikora B. Seismic hazard assessment undercomplex source size distribution of mining-induced seismicity[J].Tectonophysics, 2008, 456(1-2): 28-37. [Document 3] Mansurov V A. Prediction of rockbursts by analysis of induced seismicity data[J]. International Journal of Rock Mechanics and MiningSciences, 2001, 38(6): 893-901. [Reference 4] Zhao J, Jiang Q, Pei S, et al. Microseismicity and focal mechanism of blasting-induced block falling of intersecting chamber of large underground cavern under high geostress[J]. Journal of Central South University, 2023, 30(2): 542-554. [Reference 5] Zhao J S, Jiang Q, Lu J F, et al. Rock fracturing observation based on microseismic monitoring and borehole imaging: In situ investigation in a large underground cavern under high geostress[J]. Tunnelling and Underground Space Technology, 2022, 126: 104549. [Reference 6] Li B, Ding Q, Xu N, et al. Characteristics of microseismic b-value associated with rock mass large deformation in underground powerhouse caverns at different stress levels[J]. Journal of Central South University, 2022, 29(2): 693-711. [Reference 7] Ma C, Li T, Zhang H, et al. A method for numerical simulation based on microseismic information and the interpretation of hard rock fracture[J]. Journal of Applied Geophysics, 2019, 164: 214-224. [Reference 8] Zhao Y, Yang T, Zhang P, et al. Inversion of seepage channels based on mining-induced microseismic data[J]. International Journal of Rock Mechanics and Mining Sciences, 2020, 126: 104180. [Reference 9] Peng P, Jiang Y, Wang L, et al. Microseismic event location by considering the influence of the empty area in an excavated tunnel[J]. Sensors, 2020, 20(2): 574. [Reference 10] Dong L, Li X, Zhou Z, et al. Three-dimensional analytical solution of acoustic emission source location for cuboid monitoring network without pre-measured wave velocity[J]. Transactions of Nonferrous Metals Society of China, 2015, 25(1): 293-302. [Reference 11] Luo Z, Shang X, Wang Y, et al. P-and S-wave arrival time combined Bayesian location method for a microseismic event[J]. Journal of Central South University, 2023, 30(11): 3808-3820. [Reference 12] Li, Tao, Chen, Bingrui, Zhu, Xinhao, Wang, Xu, and Mingxing Xie. "A New Location Method Without Pre-Measuring Wave Velocity Based on Particle Swarm Optimization." Paper presented at the ISRM Regional Symposium - 11th Asian Rock Mechanics Symposium, Beijing, China, October 2021. [Reference 13] Chakraborty S, Sharma S, Saha A K, et al. A novel improved whale optimization algorithm to solve numerical optimization and real-world applications[J]. Artificial Intelligence Review, 2022: 1-112. [Reference 14] Cherki I, Chaker A, Djidar Z, et al. A sequential hybridization of genetic algorithm and particle swarm optimization for the optimal reactive power flow[J]. Sustainability, 2019, 11(14): 3862. [Reference 15] Geiger L. Probability method for the determination of earthquake epicentres from the arrival time only[J]. Bull. St. Louis Univ., 1912, 8: 60. [Literature 16] Buland R. The mechanics of locating earthquakes[J]. Bulletin of the Seismological Society of America, 1976, 66(1): 173-187. [Literature 17] Lin F, Li S L, Xue Y L, et al. Microseismic sources location methods based on different initial values[J]. Chinese Journal of Rock Mechanics and Engineering, 2010, 29(5): 996-1002. [Literature 18] Prugger A F, Gendzwill D J. Microearthquake location: An nonlinear approach that makes use of a simplex step** procedure[J]. Bulletin of the Seismological Society of America, 1988, 78(2): 799-815. [Literature 19] Dong L, Li X. A microseismic / acoustic emission source location method using arrival times of PS waves for unknown velocity system[J]. International Journal of Distributed Sensor Networks, 2013, 9(10): 307489. [Literature 20] Dong L, Sun D, Li X, et al. Theoretical and experimental studies of localization methodology for AE and microseismic sources without pre-measured wave velocity in mines[J]. IEEE access, 2017, 5: 16818-16828. [Reference 21] Liao Z, Feng T, Yu W, et al. Microseismic source location method and application based on NM-PSO algorithm[J]. Applied Sciences, 2022, 12(17): 8796. [Reference 22] Xiao, Y., Liu, Wj., Wang, Hn. et al. Research on a nonlinear hybrid optimal PSO microseismic positioning method. Appl. Geophys. (2024). [Reference 23] Lagos S R, Velis D R. Microseismic event location using global optimization algorithms: An integrated and automated workflow[J]. Journal of Applied Geophysics, 2018, 149: 18 - 24. [Reference 24] Zhou J, Shen X, Qiu Y, et al. Improving the efficiency of microseismic source locating using a heuristic algorithm-based virtual field optimization method[J]. Geomechanics and Geophysics for Geo-Energy and Geo-Resources, 2021, 7(3): 89. [Reference 25] Clerc M, Kennedy J. The particle swarm-explosion, stability, and convergence in a multidimensional complex space[J]. IEEE transactions on Evolutionary Computation, 2002, 6(1): 58 - 73. [Document 26] Ratnaweera A, Halgamuge SK, Watson H C. Self-organizinghierarchical particle swarm optimizer with time-varying accelerationcoefficients[J]. IEEE Transactions on evolutionary computation, 2004, 8(3):240-255. [Literature 27]Koessler E, Almomani A. Hybrid particle swarm optimization and pattern search algorithm[J]. Optimization and Engineering, 2021, 22(3):1539-1555. Summary of the Invention To address the technical problems of particle swarm optimization (PSO) algorithms in microseismic location, such as sensitivity to initial values ​​and difficulty in balancing the local and global search ratios, this invention provides a microseismic source location method and system based on an improved PSO optimization algorithm using chaotic mapping initialization and a hybrid update strategy.

[0007] The technical solution adopted by the method of the present invention is: a microseismic source localization method based on an improved particle swarm optimization algorithm, comprising the following steps: Step 1: Deploy n sensors in the open-pit mine, where the sensor coordinates are... , ; Step 2: Based on the pre-determined P-wave velocity, an improved particle swarm optimization algorithm is used to solve for the minimum value of the objective function within the defined domain, thereby determining the source coordinates. ; The improved particle swarm optimization algorithm introduces a hybrid chaotic mapping combining Logistic and Tent mappings in the initial population generation stage to generate a more regular and uniform initial population. After each iteration, the particle fitness is sorted, and the population with fitness below the average is called the dominant population, while the population with fitness above the average is called the suboptimal population. In the next iteration, an adaptive weighting strategy is adopted for the dominant population, and a Levy flight combined with an adaptive weighting strategy is adopted for the suboptimal population.

[0008] As a preliminary option, the improved particle swarm optimization algorithm is specifically implemented including the following sub-steps: Step 2.1: Determine the objective function to be optimized; Step 2.2: Initialize the parameters of the particle swarm, including the particle swarm size, particle dimensions, and number of iterations; Step 2.3: Generate the initial positions of the particles using Logistic-Tent mapping; Step 2.4: Calculate the fitness value of each particle and divide the particle swarm into two parts based on the average fitness value; Step 2.5: Compare the particle's current fitness value with the best historical value. If the updated fitness value is better, update the particle's optimal position and optimal fitness value; otherwise, leave them unchanged. Step 2.6: Select the particle with the best individual fitness value among all particles in the current population and compare it with the historical global best value. If the individual fitness value is better, then update the historical global best value; otherwise, leave it unchanged. Step 2.7: Update the particles and update the inertia weights; Step 2.8: Repeat steps 2.4 to 2.7 until the iteration stopping condition is met or the maximum number of iterations is reached, and output the global optimal solution.

[0009] As a preliminary selection, the objective function described in step 2.1 is: ; in, For the epicenter to reach the first The distance between the sensors; and For the P wave to reach the first , The actual time of each sensor For the first , The actual time difference of arrival of the P-waves collected by each sensor; For the first , Theoretical time difference between individual sensors Wave speed; For the P wave to reach the first The time of each sensor, The moment the earthquake source was generated.

[0010] As a preliminary step, in step 2.3, a hybrid chaotic mapping combining the Logistic mapping and the Tent mapping is used to generate a more regular and uniform initial population, thereby improving the quality of the initial solution. The mathematical descriptions of the Logistic map, the Tent map, and the mixed chaotic map are as follows: ; ; ; in, , It is a randomly generated sequence of numbers. It is the multiplier of the chaotic mapping.

[0011] As a preliminary selection, in step 2.4, the particle fitness is sorted, and the population with fitness below the average value is called the dominant population, and the population with fitness above the average value is called the suboptimal population. An adaptive weighting strategy is used for the dominant population, and a Levy flight combined with an adaptive weighting strategy is used for the suboptimal population.

[0012] As a preliminary choice, the adaptive weighting strategy is a linearly decreasing adaptive weighting strategy, as follows: ; in, As the initial inertia weight, To achieve the inertia weight at the maximum number of iterations, This represents the current iteration number. This represents the maximum number of iterations.

[0013] As a preliminary option, the Levy flight combined with an adaptive weighting strategy incorporates the Levy flight strategy into the particle position update formula, and the updated particle position is: ; in, They represent the first The particle in the first Position and velocity in the next iteration; For the first The particle in the first The optimal position in the next iteration; α is the step size scaling factor used to control the range of the random search; It is a dot product; It follows the parameter as Levy distribution, ; and These are the individual fitness value and the average fitness value of the particle, respectively.

[0014] The technical solution adopted by the system of this invention is: a microseismic source localization system based on an improved particle swarm optimization algorithm, comprising: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the microseismic source localization method based on the improved particle swarm optimization algorithm as described in any one of claims 1 to 7.

[0015] Compared with existing technologies, this invention proposes an improved particle swarm optimization algorithm based on the time difference of arrival theory, which is based on chaotic mapping initialization and hybrid update strategy. On the one hand, the initial position of the particles is generated by using Logistic-Tent mapping to obtain a uniformly distributed initial population. On the other hand, the position update method of the particles is improved by combining adaptive weights and Levy flight strategy, so as to avoid the disadvantage of the particle swarm optimization algorithm being prone to getting trapped in local optima. This effectively improves the stability and accuracy of the positioning algorithm and provides a new approach for microseismic positioning. Attached Figure Description

[0016] The technical solutions of the present invention will be further illustrated below using embodiments and specific implementation methods. In addition, some accompanying drawings are used in the description of the technical solutions. Those skilled in the art can obtain other drawings and the intent of the present invention from these drawings without any creative effort.

[0017] Figure 1 This is a flowchart of a method according to an embodiment of the present invention; Figure 2 These are three types of mapping scatter plots according to embodiments of the present invention; Figure 3 Scatter plots of the three mappings in this embodiment of the invention; Figure 4 This is a diagram showing the positioning error of microseismic event A3 in the experiment of this embodiment of the invention; Figure 5 This is the localization convergence curve of microseismic event A3 in the experiment of this embodiment of the invention. Detailed Implementation

[0018] To facilitate understanding and implementation of the present invention by those skilled in the art, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0019] Please see Figure 1 This embodiment provides a microseismic source localization method based on an improved particle swarm optimization algorithm, comprising the following steps: Step 1: Deploy n sensors in the open-pit mine, where the sensor coordinates are... , ; Step 2: Based on the pre-determined P-wave velocity, an improved particle swarm optimization algorithm is used to solve for the minimum value of the objective function within the defined domain, thereby determining the source coordinates. ; In geotechnical engineering, P-waves are often chosen for source location because they have the fastest propagation speed among seismic waves and their first arrival time is easily identifiable (Jiang C, Liu C, Shang X. Double event joint location method considering P-wave arrival time system errors[J]. Soil Dynamics and Earthquake Engineering, 2021, 149: 106890.). The most widely used method for source location is the typical MS location (TMSL) method. This method assumes that the rock strata between the source and each sensor are isotropic and homogeneous media, and that the P-wave propagation speed is... It is known and remains unchanged. The coordinates of the epicenter are... The coordinates of each sensor are The time of origin of the earthquake was Therefore, it is possible to calculate the arrival time of the P wave. Time for each sensor:

[0020] In the formula, This indicates the propagation speed of the P-wave; For the epicenter to reach the first The distance between the sensors can be obtained from the spatial distance formula:

[0021] No. , The theoretical time difference between the sensors is:

[0022] For each calculated time difference, the actual time difference can be determined from the arrival time of the P-wave recorded by the sensor:

[0023] In the formula: For the first , The actual time difference of each sensor and For the P wave to reach the first , The actual time of each sensor. Actual time difference. Time difference from theory The degree of deviation of the difference describes the error value of the earthquake source location, from which an objective function can be determined. When the objective function is equal to 0 or infinitely close to 0, the calculated source coordinates are... The difference from the actual coordinates is not significant, that is:

[0024] In the TMSL method, the measurement of P-wave propagation velocity is crucial. When the predicted wave velocity differs significantly from the actual wave velocity in the detection area, the location result will have a large error. To address the error caused by predicted wave velocity in traditional location methods, a source location method that does not require prior wave velocity measurement has been proposed. Its objective function is:

[0025] In this method, the coordinates of the earthquake source and the propagation velocity of the P-wave are unknown. To ensure accurate calculation results, the number of effective sensors, n ≥ 4, must be guaranteed. By solving for the minimum value of the objective function within the domain, the coordinates of the earthquake source and the propagation velocity of the P-wave can be determined.

[0026] Particle Swarm Optimization (PSO) algorithm, proposed by Kennedy et al. in 1995, is an intelligent optimization algorithm that simulates the foraging process of flocks of birds in nature. In this algorithm, each particle in the swarm represents a potential solution in the solution space. By iteratively updating the individual optimal solutions and the historical global optimal solutions, the algorithm finds the optimal solution for the entire solution space. Due to its excellent global optimization capabilities, it is often used to solve complex problems such as nonlinear equation systems.

[0027] Assume the search space has the following dimensions: population Composed of _ particles, then the _ ... The position of each particle can be represented as... The particle's flight speed is ;No. The optimal position of each particle is Let pbest be the position of the best position in the entire population. Let gbest be the particle's position and velocity; then the particle's position and velocity can be updated according to the following formula:

[0028]

[0029] in, As a weighted average, it balances the search ratio between individuals and the overall population; , These are individual learning factors and group learning factors, respectively. , for The random number between the two values ​​increases the randomness of the search; They represent the first The particle in the first Position and velocity in the next iteration; For the first The particle in the first The optimal position in the next iteration; For the first The global optimal position in the next iteration.

[0030] In one implementation, the improved particle swarm optimization algorithm specifically includes the following sub-steps: Step 2.1: Determine the objective function to be optimized. ; Step 2.2: Initialize the parameters of the particle swarm, including the particle swarm size, particle dimensions, and number of iterations; Step 2.3: Generate the initial positions of the particles using Logistic-Tent mapping; Similar to many swarm optimization algorithms, uneven initial population distribution directly affects the optimization results, leading to slower convergence and even causing the algorithm to get trapped in local optima. To address this issue, one implementation introduces a hybrid chaotic mapping combining the Logistic and Tent maps during the initial population generation stage. This generates a more regular and uniform initial population, thereby improving the quality of the initial solution. The mathematical descriptions of the Logistic, Tent, and hybrid mappings are as follows:

[0031]

[0032]

[0033] in, , It is a randomly generated sequence of numbers. It is the multiplier of the chaotic mapping; it should be noted that the Logistic mapping and the Tent mapping are topologically conjugate mappings, when At this time, the system will enter a short-period state, causing particles to be distributed regularly in the same area, so it is generally not taken. .

[0034] Scatter plots of the three mappings are as follows Figure 3As shown in the figure, the Logistic mapping has strong spatial ergodicity, but it has clustered and blank areas; the Tent mapping has good correlation and a relatively uniform probability density distribution, but it is prone to decaying into a periodic sequence in the later stages of iteration; the Logistic-Tent mapping combines the advantages of both, generating particles with a very uniform distribution and good random distribution capability. Therefore, this embodiment uses the Logistic-Tent mapping to generate the initial population.

[0035] Step 2.4: Calculate the fitness value of each particle and divide the particle swarm into two parts based on the average fitness value; In one implementation, to further optimize the local solution capability of the PSO algorithm, the particle fitness is sorted after each iteration. Particles with fitness below the average are designated as the dominant population, and those with fitness above the average are designated as the suboptimal population. An adaptive weighting strategy is used for the dominant population, while a hybrid strategy combining Levy flight and adaptive weighting is used for the suboptimal population.

[0036] Inertia weights reflect a particle's ability to inherit previous velocities. Larger values ​​should be assigned in the early iterations to broaden the global search range, enabling the algorithm to quickly find the global optimum. In later iterations, the weights should be appropriately reduced to improve the algorithm's local search capability and avoid getting trapped in local optima. To better balance the algorithm's global and local search capabilities, one implementation employs a linearly decreasing adaptive weight strategy, as follows:

[0037] in, As the initial inertia weight, To achieve the inertia weight at the maximum number of iterations, This represents the current iteration number. This represents the maximum number of iterations. Generally, the inertia weight ranges from [0.4, 0.9], which ensures that the algorithm has strong global search capabilities in the early stages and more accurate local searches in the later stages of iteration.

[0038] When the particle swarm optimization algorithm is in a stagnant state, particles in the population often cluster, leading to a decrease in population diversity and the algorithm getting stuck in a local optimum. Many scholars have optimized the parameters of the particle swarm optimization algorithm itself. These improvement strategies have improved the convergence speed and accuracy of the algorithm to varying degrees, but none of them have fundamentally solved the problem of particles getting stuck in a local optimum. Levy flight is a probability distribution model proposed by the French mathematician Levy, which is a non-Gaussian stochastic process. Its flight trajectory consists of frequent short-distance movements and occasional long-distance movements. The activity trajectories of many insects and even humans in nature conform to Levy flight. In one implementation, the Levy flight strategy is introduced into the particle position update formula, so that when the particle gets stuck in a local optimum, it can jump out in time and continue to search in a new area, increasing the probability of finding the optimal solution. Therefore, formula (8) can be improved to: (13) in, They represent the first The particle in the first Position and velocity in the next iteration; For the first The particle in the first The optimal position in the next iteration; α is the step size scaling factor used to control the range of the random search; It is a dot product; It follows the parameter as Levy distribution, ; and These are the individual fitness value and the average fitness value of the particle, respectively.

[0039] While Levy flight can help particles escape local optima, it doesn't guarantee that the new solution found will be better than the original one. Therefore, a portion of the population with fitness values ​​above average uses the Levy flight strategy to update their positions, while those with fitness values ​​below average still use the adaptive weighting strategy.

[0040] Step 2.5: Compare the particle's current fitness value with the best historical value. If the updated fitness value is better, update the particle's optimal position and optimal fitness value; otherwise, leave them unchanged. Step 2.6: Select the particle with the best individual fitness value among all particles in the current population and compare it with the historical global best value. If the individual fitness value is better, then update the historical global best value; otherwise, leave it unchanged. Step 2.7: Update the particles further according to formulas (7) and (13), and update the inertia weights according to formula (12); Step 2.8: Repeat steps 2.4 to 2.7 until the iteration stopping condition is met or the maximum number of iterations is reached, and output the global optimal solution.

[0041] This embodiment also provides a microseismic source localization system based on an improved particle swarm optimization algorithm, including: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, enable the one or more processors to implement the microseismic source localization method based on the improved particle swarm optimization algorithm.

[0042] This embodiment also provides a non-volatile computer-readable storage medium containing a computer program, which, when executed by one or more processors, causes the processors to perform the microseismic source localization method based on the improved particle swarm optimization algorithm.

[0043] This embodiment also provides a computer program product, including computer program instructions, which, when run on a computer, cause the computer to execute the microseismic source localization method based on the improved particle swarm optimization algorithm.

[0044] The invention will be further illustrated below through specific experiments.

[0045] To verify the accuracy and effectiveness of the algorithm proposed in this invention, a local slope in the western part of an open-pit copper mine in Jiangxi Province was selected as the test site. The test site included 12 microseismic sensors and 10 known seismic sources, arranged as follows: Figure 3 As shown, the specific sensor coordinates are shown in Table 1, and the earthquake source coordinates are shown in Table 2.

[0046] Table 1 Sensor Coordinate Table

[0047] Table 2 Source Coordinates

[0048] During the experiment, known coordinate points were struck sequentially to generate microseismic signals, and the arrival time of the P-wave at the sensor was recorded. With the wave velocity as the true value, different particle initial position generation methods and improved PSO algorithms with different optimization strategies were used to perform three-dimensional localization of the seismic source in the experimental area to compare the performance of the improved algorithms. To ensure the accuracy of the experimental results, each microseismic event was calculated 10 times using the optimized algorithm. After removing the maximum and minimum values, the average error of the remaining 8 runs was taken as the absolute error value of the seismic source.

[0049] Table 3 shows the location results of some microseismic events under different improvement strategies. -PSO, -PSO, -PSO, -PSO stands for standard PSO, chaotic mapping-based PSO, hybrid optimization strategy PSO, and chaotic mapping-hybrid optimization strategy PSO, respectively. As expected, -PSO significantly outperforms other algorithms in positioning accuracy. In other words, initializing particles through Logistic-Tent mapping and combining adaptive weights and the Levy flight strategy can effectively improve the performance of particle swarm optimization algorithms when applied to microseismic positioning.

[0050] Table 3. Microseismic event location results

[0051] Figure 4 This shows a comparison of the absolute errors of 10 runs of the microseismic event A3. Figure 4 The results shown indicate that The absolute error of the PSO positioning results fluctuated between 7.60m and 469.69m, exhibiting a highly unstable state, indicating that although... -PSO can reach the optimal solution under the best-case scenario, but its stability is poor. In contrast, -PSO, -PSO and -PSO localization results are relatively stable. Among them, -PSO and -PSO, due to its hybrid Levy flight strategy and its abrupt jump characteristics, is less prone to getting trapped in local optima, resulting in higher positioning accuracy compared to... -PSO is higher. But -PSO relative to For PSO, its localization stability is significantly improved, indicating that the Logistic-Tent mapping can effectively improve the distribution of initial particles and alleviate the shortcomings of low stability in particle swarm optimization algorithms.

[0052] Figure 5 The convergence curves of the algorithm for microseismic event A3 under different improvement strategies are shown. As can be seen from the figure, -PSO and The convergence curve of -PSO slowly declines with each iteration, and the iteration curve stalls several times, indicating that the particles are trapped in local optima, severely impacting the algorithm's convergence speed. Especially... - PSO got stuck in a local optimum after the 100th iteration and didn't escape until the end of the run, resulting in a localization error as high as 469.69m. This confirms the importance of a uniformly distributed initial population for the particle swarm optimization algorithm. Similarly, from... Figure 5 The results shown demonstrate that combining chaotic mapping and hybrid optimization strategies... -PSO converges the fastest, quickly converging to the global optimum after 45 iterations. This is because each time it gets stuck in a local optimum, it utilizes the leaping characteristic of the Levy flight strategy to help the algorithm escape local optima and continue searching in other regions of the solution space. Another interesting phenomenon comes from... -PSO iteration curve, although -PSO and -PSO uses the same optimization strategy, but -PSO's optimization speed is significantly better than -PSO, -PSO got stuck in local optima at generations 20 and 45, causing it to converge to the global optimum only at generation 70. However, compared to -PSO and -PSO, -PSO converges relatively quickly.

[0053] This invention addresses the shortcomings of the standard particle swarm optimization algorithm in microseismic source localization by proposing an improved particle swarm optimization algorithm that combines Logistic-Tent chaotic mapping with adaptive weights and a Levy flight strategy. Field tests were conducted on a local slope in the western part of an open-pit copper mine in Jiangxi Province. Comparative analysis of the experimental results leads to the following conclusions: (1) Logistic-Tent mapping is used to initialize particles during the algorithm initialization process, which improves the initial particle distribution. Adaptive weights and Levy flight strategy are introduced to optimize the particle position update formula, which effectively improves the accuracy and stability of the localization algorithm. (2) The standard PSO algorithm exhibits significant instability in source location. The randomness of its initial population distribution leads to large differences in the algorithm's results, making it prone to getting trapped in local optima and resulting in large errors. Experimental data show that the location error of the standard PSO algorithm fluctuates between 7.60m and 469.69m, failing to guarantee the accuracy and stability of source location.

[0054] It should be understood that the embodiments described above are only some, not all, of the embodiments of the present invention. Furthermore, the technical features of the various embodiments or individual embodiments provided by the present invention can be arbitrarily combined to form feasible technical solutions. Such combinations are not constrained by the order of steps and / or structural composition patterns, but must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0055] It should be understood that the above description of the preferred embodiments is quite detailed, but it should not be considered as a limitation on the scope of protection of this invention. Those skilled in the art, under the guidance of this invention, can make substitutions or modifications without departing from the scope of protection of the claims of this invention, and all such substitutions or modifications fall within the scope of protection of this invention. The scope of protection of this invention should be determined by the appended claims.

Claims

1. A method for locating microseismic sources based on an improved particle swarm optimization algorithm, characterized in that, Includes the following steps: Step 1: Deploy n sensors in the open-pit mine, where the sensor coordinates are... , ; Step 2: Based on the pre-determined P-wave velocity, an improved particle swarm optimization algorithm is used to solve for the minimum value of the objective function within the defined domain, thereby determining the source coordinates. ; The improved particle swarm optimization algorithm introduces a hybrid chaotic mapping combining Logistic and Tent mappings in the initial population generation stage to generate a more regular and uniform initial population. After each iteration, the particle fitness is sorted, and the population with fitness below the average is called the dominant population, while the population with fitness above the average is called the suboptimal population. In the next iteration, an adaptive weighting strategy is adopted for the dominant population, and a Levy flight combined with an adaptive weighting strategy is adopted for the suboptimal population.

2. The microseismic source localization method based on the improved particle swarm optimization algorithm according to claim 1, characterized in that: The improved particle swarm optimization algorithm described in step 2 is specifically implemented by the following sub-steps: Step 2.1: Determine the objective function to be optimized; Step 2.2: Initialize the parameters of the particle swarm, including the particle swarm size, particle dimensions, and number of iterations; Step 2.3: Generate the initial positions of the particles using Logistic-Tent mapping; Step 2.4: Calculate the fitness value of each particle and divide the particle swarm into two parts based on the average fitness value; Step 2.5: Compare the particle's current fitness value with the best historical value. If the updated fitness value is better, update the particle's optimal position and optimal fitness value; otherwise, leave them unchanged. Step 2.6: Select the particle with the best individual fitness value among all particles in the current population and compare it with the historical global best value. If the individual fitness value is better, then update the historical global best value; otherwise, leave it unchanged. Step 2.7: Update the particles and update the inertia weights; Step 2.8: Repeat steps 2.4 to 2.7 until the iteration stopping condition is met or the maximum number of iterations is reached, and output the global optimal solution.

3. The microseismic source localization method based on the improved particle swarm optimization algorithm according to claim 2, characterized in that, The objective function mentioned in step 2.1 is: ； in, For the epicenter to reach the The distance between the sensors; and For the P wave to reach the first , The actual time of each sensor For the first , The actual time difference of arrival of the P-waves collected by each sensor; For the first , Theoretical time difference between individual sensors This indicates the propagation speed of the P-wave; For the P wave to reach the first The time of each sensor, The moment the earthquake source was generated.

4. The microseismic source localization method based on the improved particle swarm optimization algorithm according to claim 2, characterized in that: In step 2.3, a hybrid chaotic mapping combining the Logistic mapping and the Tent mapping is used to generate a more regular and uniform initial population, thereby improving the quality of the initial solution. The mathematical descriptions of the Logistic map, the Tent map, and the mixed chaotic map are as follows: ； ； ； in, , It is a randomly generated sequence of numbers. It is the multiplier of the chaotic mapping.

5. The microseismic source localization method based on the improved particle swarm optimization algorithm according to claim 2, characterized in that: In step 2.4, the particle fitness is sorted, and the population with fitness below the average is called the dominant population, and the population with fitness above the average is called the suboptimal population. An adaptive weighting strategy is used for the dominant population, and a Levy flight combined with an adaptive weighting strategy is used for the suboptimal population.

6. The microseismic source localization method based on the improved particle swarm optimization algorithm according to claim 5, characterized in that: The adaptive weighting strategy is a linearly decreasing adaptive weighting strategy, as follows: ； in, As the initial inertia weight, To achieve the inertia weight at the maximum number of iterations, This represents the current iteration number. This represents the maximum number of iterations.

7. The microseismic source localization method based on the improved particle swarm optimization algorithm according to claim 5, characterized in that: The Levy flight combined with the adaptive weight strategy introduces the Levy flight strategy into the particle position update formula, and the updated particle position is: ； in, They represent the first The particle in the first Position and velocity in the next iteration; For the first The particle in the first The optimal position in the next iteration; α is the step size scaling factor used to control the range of the random search; It is a dot product; It follows the parameter as Levy distribution, ; and These are the individual fitness value and the average fitness value of the particle, respectively.

8. A microseismic source localization system based on an improved particle swarm optimization algorithm, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the microseismic source localization method based on the improved particle swarm optimization algorithm as described in any one of claims 1 to 7.

9. A non-volatile computer-readable storage medium containing a computer program, characterized in that: When the computer program is executed by one or more processors, the processors perform the microseismic source localization method based on the improved particle swarm optimization algorithm as described in any one of claims 1 to 7.

10. A computer program product comprising computer program instructions, characterized in that: When the computer program instructions are executed on a computer, the computer performs the microseismic source localization method based on the improved particle swarm optimization algorithm as described in any one of claims 1 to 7.

Images