Microseismic event positioning and tracking method and system

Abstract

The embodiment of the present invention provides a method and system for locating and tracking microseismic events. Through the embodiment of the present invention, a technical idea for locating and tracking microseismic events based on the second-order differential grid wave velocity algorithm and AI algorithm debugging is provided. In detail, the embodiment of the present invention achieves high-precision positioning and tracking of microseismic events through refined algorithm design and multi-step optimization strategy, significantly improving positioning accuracy and tracking efficiency. Such a design provides a strong guarantee for safety and efficiency in many fields such as geological engineering, mining, and tunnel construction.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence and data analysis technology, and in particular to a microseismic event positioning and tracking method and system. Background Art

[0002] In the fields of geological engineering, mining, tunnel construction, etc., the positioning and tracking of microseismic events is crucial to ensure engineering safety and improve production efficiency. However, traditional microseismic event positioning methods often have problems such as inaccurate positioning and low tracking efficiency, which cannot meet the needs of actual engineering applications. Especially when dealing with microseismic events under complex geological conditions, the positioning error is large, which seriously affects the accurate assessment of rock mass or structural stability. Summary of the invention

[0003] The present invention at least provides a method and system for locating and tracking microseismic events.

[0004] The technical solution of the present invention is achieved through at least some of the following embodiments.

[0005] A microseismic event locating and tracking method is applied to a microseismic event locating and tracking system, the method comprising: performing microseismic event locating and tracking on past microseismic waveform data in an AI algorithm debugging sample set by using a second-order differential grid wave velocity algorithm to obtain past microseismic event locating and tracking labels corresponding to each past microseismic waveform data, wherein the AI ​​algorithm debugging sample set includes the past microseismic waveform data and a priori microseismic event locating and tracking annotations corresponding to the past microseismic waveform data; determining a debugging error update coefficient corresponding to the past microseismic waveform data based on a past microseismic waveform state vector of the past microseismic waveform data and the past microseismic event locating and tracking labels, the The past microseismic waveform state vector includes at least one of the microseismic event type and the prior annotation number of the prior microseismic event positioning and tracking annotation corresponding to the past microseismic waveform data, the microseismic event type includes a non-noise disturbance type and a noise disturbance type, and the prior annotation number represents the number of past microseismic waveform data with the prior microseismic event positioning and tracking annotation in the AI ​​algorithm debugging sample set; based on the debugging error update coefficient, the past microseismic event positioning and tracking label, and the prior microseismic event positioning and tracking annotation, the algorithm positioning and tracking error of the past microseismic waveform data is determined; based on the algorithm positioning and tracking error, the second-order difference grid wave velocity algorithm is optimized.

[0006] A microseismic event positioning and tracking system includes: a processor, a memory and a network interface; the processor is connected to the memory and the network interface; the network interface is used to provide a data communication function, the memory is used to store program codes, and the processor is used to call the program codes to execute the above method.

[0007] A computer-readable storage medium stores a computer program, which executes a microseismic event location and tracking method when running.

[0008] A computer program product includes a computer program or a computer executable instruction. When the computer program or the computer executable instruction is executed by a processor, a microseismic event location and tracking method is implemented.

[0009] According to an embodiment of the present invention, a technical idea for positioning and tracking microseismic events based on a second-order differential grid wave velocity algorithm and AI algorithm debugging is proposed. In detail, the embodiment of the present invention achieves high-precision positioning and tracking of microseismic events through refined algorithm design and multi-step optimization strategy, significantly improving positioning accuracy and tracking efficiency. Such a design provides a strong guarantee for safety and efficiency in many fields such as geological engineering, mining, and tunnel construction.

[0010] It should be understood that the above general description and the following detailed description are merely exemplary and explanatory, and are not intended to limit the technical solutions of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A schematic diagram of a microseismic event positioning and tracking system provided by an embodiment of the present invention is shown.

[0012] Figure 2 A flow chart of a microseismic event positioning and tracking method provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0013] Figure 1 A schematic diagram of a microseismic event location and tracking system provided by an embodiment of the present invention is shown, and the microseismic event location and tracking system 100 includes a processor 110, a memory 120, and a network interface 130. The processor 110 is connected to the memory 120 and the network interface 130. Further, the network interface 130 is used to provide a data communication function, the memory 120 is used to store program codes, and the processor 110 is used to call the program codes to execute the microseismic event location and tracking method.

[0014] Figure 2 is a schematic diagram showing a flow chart of a method for locating and tracking microseismic events that can be implemented in an embodiment of the present invention. The method for locating and tracking microseismic events can be implemented by Figure 1 The microseismic event location and tracking system 100 shown implements a microseismic event location and tracking method illustratively including steps 110 to 130 .

[0015] Step 110: Use a second-order differential grid wave velocity algorithm to perform microseismic event location tracking on past microseismic waveform data in the AI ​​algorithm debugging sample set to obtain past microseismic event location tracking labels corresponding to each past microseismic waveform data.

[0016] Among them, the AI ​​algorithm debugging sample set includes the past microseismic waveform data and the prior microseismic event positioning and tracking annotations corresponding to the past microseismic waveform data.

[0017] Step 120: Determine a debugging error update coefficient corresponding to the past microseismic waveform data based on the past microseismic waveform state vector of the past microseismic waveform data and the past microseismic event location tracking tag.

[0018] Among them, the past microseismic waveform state vector includes at least one of the microseismic event type and the prior annotation number of the past microseismic waveform data corresponding to the prior microseismic event positioning and tracking annotations, the microseismic event type includes non-noise disturbance type and noise disturbance type, and the prior annotation number represents the number of past microseismic waveform data with the prior microseismic event positioning and tracking annotations in the AI ​​algorithm debugging sample set.

[0019] Step 130, determining the algorithm positioning and tracking error of the past microseismic waveform data based on the debugging error update coefficient, the past microseismic event positioning and tracking label, and the prior microseismic event positioning and tracking annotation; and optimizing the second-order difference grid wave velocity algorithm based on the algorithm positioning and tracking error.

[0020] The microseismic event location and tracking system in the embodiment of the present invention plays a vital role in many fields such as geological engineering, mining, and tunnel construction. It helps engineers and researchers understand the stability of rock masses or structures by monitoring and analyzing small earthquake events, thereby predicting and preventing potential geological disasters. The embodiment of the present invention will use a specific application scenario to introduce in detail how the microseismic event location and tracking system implements steps 110 to 130 to improve the location accuracy and tracking efficiency of microseismic events.

[0021] For example, during the excavation of an underground cavern of a large hydropower station, in order to ensure construction safety and cavern stability, engineers decided to deploy a microseismic event positioning and tracking system to conduct real-time monitoring and positioning and tracking of microseismic events generated during the excavation process.

[0022] In step 110, the microseismic event location and tracking system first processes the past microseismic waveform data in the AI ​​algorithm debugging sample set using a second-order differential grid wave velocity algorithm. The purpose of this step is to locate and track past microseismic events and provide accurate labels for subsequent algorithm debugging.

[0023] It can be understood that the second-order difference grid wave velocity algorithm is the core of microseismic event location. It is based on the rock mass wave velocity model of three-dimensional discrete grids, and estimates the source location of microseismic events by simulating the propagation path and velocity of waves in the rock mass. The algorithm uses the second-order difference form, which can more accurately calculate the wave propagation time, thereby improving the accuracy of positioning.

[0024] Correspondingly, the AI ​​algorithm debugging sample set is a data set containing past microseismic waveform data and its corresponding prior microseismic event location tracking annotations. These data come from monitoring records of previous similar projects or are obtained through experiments. Prior microseismic event location tracking annotations are manually annotated by experienced geological engineers or researchers based on waveform features to indicate the true location of microseismic events.

[0025] Furthermore, the past microseismic waveform data can be processed by the second-order differential grid wave velocity algorithm to obtain the past microseismic event location tracking labels corresponding to each data. The past microseismic event location tracking labels represent the microseismic event positions estimated by the algorithm and will be used for subsequent algorithm debugging and error analysis.

[0026] After obtaining the location tracking tags of past microseismic events, the microseismic event location tracking system needs to determine the debugging error update coefficient based on the state vector and location tracking tags of past microseismic waveform data. Therefore, the purpose of step 120 is to quantify the location error of the algorithm and provide a basis for subsequent algorithm optimization.

[0027] The past microseismic waveform state vector is a vector that describes the characteristics of past microseismic waveform data, including the types of microseismic events and the number of prior annotations. Microseismic event types are divided into non-noise disturbance types and noise disturbance types, which are used to distinguish microseismic events caused by geological activities such as rock mass rupture and noise caused by external interference (such as mechanical vibration). The number of prior annotations indicates the number of past microseismic waveform data with prior microseismic event location tracking annotations in the AI ​​algorithm debugging sample set, which is used to reflect the richness of the data.

[0028] Furthermore, the location tracking labels of past microseismic events can be compared with the location tracking annotations of prior microseismic events, the difference between the two can be calculated, and the debugging error update coefficient can be determined based on the size of the difference and the state vector of the past microseismic waveform. The debugging error update coefficient reflects the positioning error level of the algorithm in the current state and will be used for subsequent algorithm optimization.

[0029] After determining the debugging error update coefficient, the microseismic event location and tracking system can analyze the algorithm location and tracking error based on the debugging error update coefficient, past microseismic event location and tracking labels, and prior microseismic event location and tracking annotations, and optimize the second-order difference grid wave velocity algorithm accordingly.

[0030] In step 130, the microseismic event location tracking system evaluates the positioning accuracy of the algorithm by calculating statistical indicators such as the average error and maximum error between the location tracking labels of past microseismic events and the prior microseismic event location tracking annotations. At the same time, the microseismic event location tracking system also analyzes the spatial distribution characteristics of the error to identify in which areas or conditions the algorithm is prone to large positioning errors. Then, the microseismic event location tracking system optimizes the second-order difference grid wave velocity algorithm based on the error analysis results. The optimization method may include adjusting the resolution of the grid, improving the parameters of the wave velocity model, and adopting more advanced numerical calculation methods. The goal of the optimization is to reduce the positioning error and improve the accuracy and stability of the algorithm.

[0031] Furthermore, the optimized algorithm needs to be iteratively debugged and verified on new microseismic waveform data. The microseismic event location tracking system uses this data as input, locates and tracks microseismic events again, and compares the differences between the location results and the prior annotations. Through multiple iterative debugging and verification, the microseismic event location tracking system can gradually improve the performance of the algorithm until satisfactory positioning accuracy is achieved.

[0032] It can be seen that by applying the above content, the microseismic event location and tracking system can monitor and locate the microseismic events generated during the excavation of the underground cavern of the hydropower station in real time and accurately, providing engineers with valuable geological information. This information helps engineers to understand the stress state and fracture process inside the rock mass in a timely manner, so as to formulate more scientific and reasonable construction plans and safety measures.

[0033] In other application scenarios, for example, in mining, the microseismic event location tracking system can be used to identify and predict geological disasters such as rock bursts and mine earthquakes; in tunnel construction, the microseismic event location tracking system can help engineers monitor the stability of tunnel surrounding rocks; in earthquake research, the microseismic event location tracking system can provide important data support for earthquake prediction and earthquake engineering. Therefore, as an advanced geological monitoring technology, the microseismic event location tracking system will play an increasingly important role in ensuring engineering safety and improving production efficiency.

[0034] To further understand the above technical solution, each step in step 110 to step 130 is described in detail below.

[0035] In the microseismic event location and tracking system, the second-order differential grid wave velocity algorithm and past microseismic waveform data are two key elements that constitute the core of the embodiment of the present invention. They not only determine the accuracy of system location and tracking, but also directly affect the effect of subsequent algorithm optimization and engineering application.

[0036] 1. Detailed explanation of the second-order difference grid wave velocity algorithm

[0037] 1. Algorithm principle

[0038] The second-order difference grid wave velocity algorithm is a numerical calculation method based on discrete grids and difference equations. It is used to simulate the propagation process of seismic waves in rock mass and estimate the source location of microseismic events. The algorithm divides the continuous rock mass space into several discrete grid units, each of which is assigned a specific wave velocity value to reflect the heterogeneity of the rock mass. Then, by solving the second-order difference form of the wave equation, the propagation time and path of the seismic wave in the grid are calculated, and the location of the microseismic event is finally achieved.

[0039] 2. Algorithm implementation steps

[0040] (1) Grid division: First, the monitoring area is divided into several regular rectangular or triangular grid cells according to its shape and size. The resolution of the grid (i.e., cell size) should be selected according to actual needs to ensure a balance between calculation accuracy and efficiency.

[0041] (2) Wave velocity model construction: Based on geological exploration data and empirical knowledge, a specific wave velocity value is assigned to each grid cell. These wave velocity values ​​can be obtained through experimental measurement, numerical simulation, or expert experience to reflect the physical and mechanical properties of the rock mass.

[0042] (3) Discretization of wave equations: The partial differential equations describing the propagation of seismic waves are converted into difference equations. This process can be achieved by using methods such as Taylor series expansion, where the derivative terms in the equations are replaced by the function value difference quotients at the grid nodes. For the second-order difference format, the second-order time and space discretization accuracy needs to be considered to ensure the stability and accuracy of the calculation results.

[0043] (4) Iterative solution: Based on the differential equation and initial conditions (such as the source location, initial waveform, etc.), the propagation time and path of the seismic wave in the grid are solved through iterative calculation. This process requires setting an appropriate time step and number of iterations to ensure the convergence and accuracy of the calculation results.

[0044] (5) Earthquake source location: Based on the difference between the theoretical arrival time and the actual arrival time of the seismic wave at each grid node, the source location of the microseismic event is determined through a search algorithm (such as the least squares method, the fast march method, etc.). This step is the core part of the algorithm and is also the key to accurately locating microseismic events.

[0045] 3. Algorithm Advantages

[0046] High precision: The second-order difference grid wave velocity algorithm can more accurately simulate the propagation process of seismic waves by considering the discrete accuracy of the second order in time and the second order in space, thereby improving the accuracy of earthquake source location.

[0047] Flexibility: The algorithm can flexibly adjust the grid resolution and wave velocity model parameters according to the actual situation and needs of the monitoring area to adapt to rock structures of different complexities.

[0048] Wide applicability: It is not only suitable for engineering fields such as underground cavern excavation of hydropower stations, but can also be widely used in mining, tunnel construction, earthquake monitoring and other fields.

[0049] 2. Past Microseismic Waveform Data

[0050] 1. Past microseismic waveform data refers to the seismic waveform data and related information collected by the microseismic monitoring system in the past period of time. These data come from monitoring records in actual engineering or experimental processes and have the following characteristics:

[0051] Real-time: Data is generated during real-time monitoring and can reflect the real-time changes in the stress state and fracture process inside the rock mass;

[0052] Diversity: Due to the differences in monitoring areas and monitoring conditions, past microseismic waveform data can contain multiple types of seismic waveform features, such as P waves, S waves, surface waves, etc.

[0053] Noise interference: In the actual monitoring process, due to the influence of factors such as external environmental noise and instrument noise, the past microseismic waveform data may contain certain noise interference components;

[0054] Limitedness: Due to factors such as monitoring time and monitoring range, past microseismic waveform data may be limited in quantity and may not fully cover all possible earthquake waveform characteristics.

[0055] 2. Data processing and analysis methods

[0056] In order to extract valuable information from past microseismic waveform data and use it for algorithm debugging and optimization of microseismic event location and tracking systems, a series of data processing and analysis work is required. These works mainly include the following aspects:

[0057] (1) Data preprocessing: including noise filtering, waveform correction, time alignment and other steps to reduce noise interference and improve data quality. Noise filtering can be achieved by using digital filters and other methods; waveform correction can be adjusted by comparing with the standard waveform template; time alignment requires ensuring that the waveform data collected by different monitoring stations are consistent in time;

[0058] (2) Feature extraction: Extract key feature information from the preprocessed waveform data, such as amplitude, frequency, phase, arrival time, etc. This feature information will be used in subsequent tasks such as waveform classification and earthquake source location. Feature extraction methods can be implemented using a variety of technical means such as statistical analysis, signal processing, and machine learning;

[0059] (3) Waveform classification and recognition: Classify and recognize waveforms based on the extracted feature information to distinguish different types of seismic waveform features (such as P waves, S waves, etc.) and non-seismic waveform features (such as noise, etc.). Waveform classification and recognition methods can be implemented using a variety of machine learning algorithms such as template matching, support vector machine (SVM), neural network, etc.

[0060] (4) Source location and tracking: Based on waveform classification and recognition results and numerical calculation methods such as the second-order difference grid wave velocity algorithm, the source location and tracking of microseismic events are achieved. This step requires comprehensive analysis and judgment based on multiple data sources and information fusion technology to improve positioning accuracy and reliability.

[0061] 3. The role of data in algorithm debugging and optimization

[0062] Past microseismic waveform data plays a vital role in the algorithm debugging and optimization process of the microseismic event location and tracking system. It is specifically manifested in the following aspects:

[0063] (1) Providing training samples: Past microseismic waveform data can be used as a training sample set for AI algorithms for model training and parameter adjustment. Through repeated iterative training, the AI ​​model can learn the characteristic representation and classification rules of different types of seismic waveforms, thereby improving the accuracy and efficiency of waveform classification and recognition;

[0064] (2) Verify algorithm performance: Applying AI algorithms to past microseismic waveform data for testing and verification can evaluate the algorithm's performance indicators such as classification accuracy, recall rate, F1 score, etc., and identify problems and deficiencies in the algorithm to provide a basis and direction for subsequent optimization and improvement;

[0065] (3) Guiding algorithm optimization: By analyzing the characteristic distribution and classification results of past microseismic waveform data, the differences and similarities between different types of earthquake waveforms can be discovered, and the optimization and improvement of the algorithm can be guided accordingly, such as adjusting the feature extraction method, improving the classifier design, etc., to improve the performance and robustness of the algorithm;

[0066] (4) Support for practical applications: AI algorithms that have been fully trained and verified can be applied to microseismic event location and tracking systems in actual projects to achieve real-time monitoring and early warning functions. Prior knowledge and experience information in past microseismic waveform data can also provide important reference and support for practical applications.

[0067] 3. In the field of microseismic monitoring, the classification of microseismic events is a basic and important task. According to the causes and characteristics of the events, microseismic events can be divided into non-noise disturbance types and noise disturbance types. This classification not only helps to better understand the nature of microseismic phenomena, but also improves the accuracy and reliability of monitoring in practical applications. The following will introduce the principles and beneficial effects of these two classifications in detail.

[0068] 1. Types of non-noise disturbances

[0069] Non-noise disturbance types refer to microseismic events caused by natural or human factors such as rock fracture and stress release. These events usually have clear physical mechanisms and energy release processes, and can reflect the stress state and damage inside the rock mass. Non-noise disturbance types of microseismic events mainly include the following:

[0070] Rock fracture microseismic: A microseismic event caused by the fracture of underground rock mass under stress. This event is particularly common in engineering activities such as mining, tunneling, oil and gas extraction, etc. Rock fracture microseismic has obvious waveform characteristics, short duration, and usually small amplitude, but it is enough to be captured by a highly sensitive microseismic monitoring system;

[0071] Stress release microseism: After the underground rock mass is subjected to stress for a long time, when the stress accumulates to a certain extent, a sudden stress release will occur, resulting in a microseismic event. This event is particularly significant in areas with frequent seismic activity and is also one of the important means of studying earthquake precursors;

[0072] Artificially induced microseismic events: In some engineering activities, microseismic events are artificially induced to monitor the integrity and stability of the rock mass, such as through blasting, hydraulic fracturing, etc. These artificially induced microseismic events are controllable and repeatable, and are an important means of evaluating the properties of the rock mass.

[0073] 2. Types of noise disturbance

[0074] Noise disturbance refers to interference signals generated by external environmental factors or the monitoring equipment itself. These signals are not caused by changes in the internal stress state of the rock mass, so they are useless or even harmful to microseismic monitoring. Microseismic events of noise disturbance mainly include the following:

[0075] Transportation noise: vibration and noise signals generated by road traffic, railway transportation, etc. These signals may be transmitted to the monitoring area through the surface and interfere with microseismic monitoring;

[0076] Construction noise: At construction sites, vibration and noise signals generated by activities such as excavation, blasting, and piling may also interfere with microseismic monitoring;

[0077] Industrial activity noise: Vibration and electromagnetic interference signals generated by industrial activities such as factory machine operation and power transmission may also affect microseismic monitoring;

[0078] Natural background noise: including noise signals generated by natural phenomena such as wind, rain, and lightning, as well as background noise generated by geological tectonic activities themselves. Although these noise signals cannot be completely eliminated, their impact can be reduced through data processing technology;

[0079] Internal noise of equipment: The internal noise signals such as electronic noise and mechanical vibration that may exist in the monitoring equipment itself will also affect the microseismic monitoring results.

[0080] It can be understood that by classifying microseismic events into non-noise disturbance types and noise disturbance types, it is possible to more accurately identify real microseismic events caused by changes in the internal stress state of the rock mass and eliminate the influence of external interference signals. This helps to improve the accuracy and reliability of microseismic monitoring and provide a more reliable basis for engineering safety and geological disaster early warning.

[0081] In the process of microseismic monitoring, the collected raw data needs to be processed and analyzed in a series of ways to extract useful microseismic information. By classifying microseismic events, different data processing methods and technical means can be used for different types of events. For example, for microseismic events of non-noise disturbance types, waveform matching, arrival time inversion and other methods can be used for precise positioning and parameter estimation; while for microseismic events of noise disturbance types, filtering, denoising and other methods can be used to reduce their impact. This classification processing method can optimize the data processing process and improve processing efficiency and accuracy.

[0082] By classifying microseismic events, we can more clearly understand the characteristics and laws of different types of events, providing strong support for technological innovation and development. For example, for microseismic events of non-noise disturbance types, we can conduct more in-depth research to explore their physical mechanisms and energy release processes; while for microseismic events of noise disturbance types, we can study more effective denoising technologies and methods. These studies will not only help promote the development and innovation of microseismic monitoring technology, but also provide new ideas and methods for scientific research in related fields.

[0083] In engineering practice, microseismic monitoring technology is widely used in the fields of underground cavern excavation in hydropower stations, mining, tunneling, oil and gas extraction, etc. By classifying, processing and analyzing microseismic events, the stability and integrity of the rock mass can be more accurately evaluated, providing a scientific basis for engineering design and construction. For example, in the mining process, by monitoring rock fracture microseismic events, potential safety hazards such as roof falls and spalling can be discovered in a timely manner; in the tunneling process, by monitoring stress release microseismic events, the stability and support effect of the tunnel can be evaluated. These applications not only help to ensure engineering safety and quality, but also improve engineering efficiency and economic benefits.

[0084] 4. Debugging Error Update Coefficient

[0085] The debugging error update coefficient is a key parameter used to quantify the algorithm positioning error and guide the algorithm optimization. In the microseismic event positioning and tracking system, this coefficient is determined based on the state vector and positioning tracking label of the past microseismic waveform data.

[0086] The past microseismic waveform state vector contains information describing the characteristics of past microseismic waveform data, such as the types of microseismic events (non-noise disturbance types and noise disturbance types) and the number of prior annotations (the number of past microseismic waveform data with prior microseismic event location tracking annotations). The types of microseismic events are used to distinguish between microseismic events caused by geological activities and noise caused by external interference, which helps the algorithm to process different types of microseismic events more accurately. The number of prior annotations reflects the richness of the data, which can help the algorithm make appropriate adjustments when processing different amounts of data.

[0087] By comparing the location tracking labels of past microseismic events (the locations of microseismic events estimated by the algorithm) and the prior location tracking annotations of microseismic events (the real locations manually annotated by experienced geological engineers or researchers), the difference between the two can be calculated. This difference reflects the positioning error level of the algorithm in its current state.

[0088] The debugging error update coefficient can be determined based on the difference between the positioning tracking label and the prior annotation, as well as the information in the past microseismic waveform state vector. The debugging error update coefficient is used to guide the subsequent algorithm optimization, which reflects the degree to which the algorithm needs to be adjusted in order to reduce the positioning error.

[0089] 5. Algorithm positioning and tracking error is an important indicator to measure the performance of microseismic event positioning and tracking system, which reflects the accuracy of the algorithm in locating microseismic events.

[0090] Error calculation: The algorithm positioning and tracking error is obtained by calculating the difference between the positioning and tracking labels of past microseismic events and the prior microseismic event positioning and tracking annotations. Commonly used error statistical indicators include average error, maximum error, etc. These indicators can help comprehensively evaluate the positioning accuracy of the algorithm.

[0091] Error analysis: In addition to calculating the statistical indicators of the error, it is also necessary to analyze the spatial distribution characteristics of the error. This helps to identify in which areas or conditions the algorithm is prone to large positioning errors, thereby providing targeted guidance for subsequent algorithm optimization.

[0092] Algorithm optimization: Based on the error analysis results, the second-order difference grid wave velocity algorithm can be optimized. The goal of optimization is to reduce positioning errors and improve the accuracy and stability of the algorithm. Optimization methods can include adjusting the resolution of the grid, improving the parameters of the wave velocity model, and using more advanced numerical calculation methods.

[0093] In summary, the debugging error update coefficient and algorithm positioning and tracking error play a vital role in the algorithm optimization and performance improvement of the microseismic event positioning and tracking system. Through continuous iterative debugging and verification, the performance of the algorithm can be gradually improved, and the positioning accuracy and tracking efficiency of microseismic events can be improved, thereby providing a strong guarantee for the safety and efficiency of many fields such as geological engineering, mining, and tunnel construction.

[0094] The embodiment of the present invention significantly improves the positioning accuracy and tracking efficiency of microseismic events by implementing a series of innovative technical steps. Specifically, the embodiment of the present invention first uses the second-order differential grid wave velocity algorithm to process the past microseismic waveform data in the AI ​​algorithm debugging sample set, realizes high-precision positioning and tracking of microseismic events, and generates corresponding past microseismic event positioning and tracking tags. This step not only provides an accurate benchmark for subsequent algorithm debugging, but also greatly enriches the data set of microseismic events.

[0095] Furthermore, the embodiment of the present invention cleverly determines the debugging error update coefficient based on the state vector and positioning tracking tags of past microseismic waveform data. The introduction of this coefficient enables the algorithm to dynamically adjust its positioning strategy according to the different types of microseismic events and the number of prior annotations, thereby more effectively reducing the positioning error.

[0096] In addition, the embodiment of the present invention also comprehensively evaluates the positioning and tracking error of the algorithm based on the debugging error update coefficient, the positioning and tracking labels of past microseismic events, and the positioning and tracking annotations of prior microseismic events, and optimizes the second-order difference grid wave velocity algorithm accordingly. This optimization process not only improves the accuracy and stability of the algorithm, but also makes the algorithm more adaptable to new data and has better positioning effect.

[0097] In summary, the technical solution of the embodiment of the present invention significantly improves the positioning accuracy and tracking efficiency of microseismic events through refined algorithm design and multi-step optimization strategy, providing a strong guarantee for the safety and efficiency in many fields such as geological engineering, mining, tunnel construction, etc.

[0098] In some examples, the second-order differential grid wave velocity algorithm is obtained through dual-cycle debugging; the optimizing the second-order differential grid wave velocity algorithm based on the algorithm positioning tracking error includes: in a first debugging cycle, optimizing the second-order differential grid wave velocity algorithm based on a first algorithm positioning tracking error, the first algorithm positioning tracking error being an algorithm positioning tracking error corresponding to past microseismic waveform data in a first AI algorithm debugging sample set, and the past microseismic waveform data in the first AI algorithm debugging sample set having non-associated prior microseismic event positioning tracking annotations; in a second debugging cycle, optimizing the second-order differential grid wave velocity algorithm based on a second algorithm positioning tracking error based on the second-order differential grid wave velocity algorithm obtained through debugging in the first debugging cycle, the second algorithm positioning tracking error being an algorithm positioning tracking error corresponding to past microseismic waveform data in a second AI algorithm debugging sample set, and the past microseismic waveform data in the second AI algorithm debugging sample set having at least one prior microseismic event positioning tracking annotation.

[0099] It can be understood that the core of the embodiment of the present invention is to combine the second-order differential grid wave velocity algorithm and AI algorithm debugging to achieve high-precision positioning and tracking of microseismic events through fine processing of past microseismic waveform data.

[0100] In the specific implementation, the second-order difference grid wave velocity algorithm is obtained through a unique two-cycle tuning process. This process includes two tuning cycles, each cycle has its own specific goals and optimization strategies.

[0101] In the first debugging cycle, the system mainly optimizes the second-order differential grid wave velocity algorithm based on the first algorithm positioning and tracking error. The first algorithm positioning and tracking error in the embodiment of the present invention refers to the algorithm positioning and tracking error corresponding to the past microseismic waveform data in the first AI algorithm debugging sample set. In particular, the past microseismic waveform data in the first AI algorithm debugging sample set have non-correlated prior microseismic event positioning and tracking annotations. This means that there is no obvious correlation between these annotations and actual microseismic events, which increases the difficulty of positioning and tracking. However, through the processing of the second-order differential grid wave velocity algorithm, the system can still effectively locate and track these microseismic events, and generate corresponding past microseismic event positioning and tracking labels.

[0102] In this step, the second-order difference grid wave velocity algorithm demonstrates its powerful processing capabilities. It extracts key feature information through a detailed analysis of microseismic waveform data and locates microseismic events based on this information. At the same time, the algorithm also considers different types of microseismic events, such as non-noise disturbances and noise disturbances, as well as the influence of the number of prior annotations, so as to dynamically adjust its positioning strategy to reduce positioning errors.

[0103] After completing the first debugging cycle, the system obtains a preliminarily optimized second-order differential grid wave velocity algorithm. However, this is not the final result. In order to further improve positioning accuracy and tracking efficiency, the system needs to be optimized in the second debugging cycle.

[0104] In the second debugging cycle, the system is further optimized based on the second algorithm positioning and tracking error on the basis of the second-order differential grid wave velocity algorithm obtained by debugging in the first debugging cycle. The second algorithm positioning and tracking error in the embodiment of the present invention refers to the algorithm positioning and tracking error corresponding to the past microseismic waveform data in the second AI algorithm debugging sample set. Unlike the first debugging cycle, the past microseismic waveform data in the second AI algorithm debugging sample set has at least one prior microseismic event positioning and tracking annotation. This means that there is a certain correlation between these annotations and actual microseismic events, thereby providing more useful information for positioning and tracking.

[0105] In the second debugging cycle, the system made full use of these associated prior annotation information and optimized the second-order difference grid wave velocity algorithm more finely. By adjusting the algorithm parameters and formulating the optimization strategy, the system successfully reduced the positioning error and improved the positioning accuracy and tracking efficiency.

[0106] After two-cycle debugging, the second-order differential grid wave velocity algorithm has been significantly optimized and improved. At this point, the system has the ability to locate and track microseismic events with high precision. However, in order to ensure the stability and reliability of the algorithm, the system still needs to undergo a series of tests and verifications.

[0107] During the testing and verification phase, the system selected several representative microseismic events as test samples and used the optimized second-order differential grid wave velocity algorithm to locate and track these samples. By comparing and analyzing the actual microseismic events, the system successfully verified the positioning accuracy and tracking efficiency of the algorithm. At the same time, the system also tested the adaptability of the algorithm under different geological conditions. The results showed that the algorithm can maintain high positioning accuracy and stability under various complex geological conditions.

[0108] In this way, through double-cycle debugging and verification of multiple test samples, high positioning accuracy and stability can be maintained under various complex geological conditions. This feature enables the system to play a role in a wider range of application scenarios.

[0109] In the next step, in the first debugging cycle, the debugging error update coefficient corresponding to the past microseismic waveform data is determined based on the past microseismic waveform state vector of the past microseismic waveform data and the past microseismic event positioning and tracking label, including: based on the microseismic event type of the past microseismic waveform data and the past microseismic event positioning and tracking label, the first debugging error update coefficient corresponding to the past microseismic waveform data is determined, and the first debugging error update coefficient is used to adjust the algorithm positioning and tracking error of the noise disturbance type; based on the prior annotation number of the prior microseismic event positioning and tracking annotation and the first classification keyword number in the first AI algorithm debugging sample set, the past microseismic event positioning and tracking annotation is determined. a second debugging error update coefficient corresponding to the waveform data, the number of the first classification keywords represents the total number of prior annotation keywords contained in the first AI algorithm debugging sample set, and the second debugging error update coefficient has a first quantitative relationship with the number of prior annotations; the algorithm positioning and tracking error of the past microseismic waveform data is determined based on the debugging error update coefficient, the past microseismic event positioning and tracking tags, and the prior microseismic event positioning and tracking annotations, including: determining the first algorithm positioning and tracking error of the past microseismic waveform data based on the first debugging error update coefficient, the second debugging error update coefficient, the past microseismic event positioning and tracking tags, and the prior microseismic event positioning and tracking annotations.

[0110] In some more detailed embodiments, first, a first debugging cycle is entered. In this stage, the main task of the microseismic event location tracking system is to determine the debugging error update coefficient corresponding to the past microseismic waveform data based on the past microseismic waveform state vector of the past microseismic waveform data and the past microseismic event location tracking label.

[0111] The past microseismic waveform state vector is a vector containing at least one of the microseismic event type and the prior annotation number of the past microseismic event location tracking annotation. The microseismic event type is an important feature that can help the system distinguish different types of microseismic events, such as non-noise disturbances and noise disturbances. Non-noise disturbances are usually caused by geological activities or human activities, while noise disturbances may be caused by environmental factors or equipment noise. These two types of microseismic events have different characteristics when positioning and tracking, so different processing strategies are required.

[0112] The past microseismic event location tracking tag is obtained by the system after processing the past microseismic waveform data through the second-order difference grid wave velocity algorithm. It represents the location information of the past microseismic events in time and space. This tag is an important basis for the subsequent determination of the debugging error update coefficient.

[0113] In the first debugging cycle, the system first determines the first debugging error update coefficient corresponding to the past microseismic waveform data based on the microseismic event types and the positioning and tracking tags of the past microseismic events. The main function of this coefficient is to adjust the algorithm positioning and tracking error of the noise disturbance type. Since noise disturbances usually have greater randomness and uncertainty, their positioning and tracking errors are also larger. By introducing the first debugging error update coefficient, the system can dynamically adjust the algorithm according to the characteristics of the noise disturbance to reduce its positioning and tracking error.

[0114] In addition to the types of microseismic events, the number of prior annotations of the prior microseismic event location tracking annotations is also an important factor in determining the debugging error update coefficient. The number of prior annotations represents the number of past microseismic waveform data with prior microseismic event location tracking annotations in the first AI algorithm debugging sample set. This number reflects the richness of the useful information contained in the sample set. Based on the number of prior annotations and the number of first-category keywords in the first AI algorithm debugging sample set, the system can determine the second debugging error update coefficient corresponding to the past microseismic waveform data.

[0115] The number of first classification keywords in the embodiment of the present invention is a parameter that characterizes the total number of prior annotation keywords contained in the first AI algorithm debugging sample set. It reflects the diversity and complexity of the annotations in the sample set. The second debugging error update coefficient has a first quantitative relationship with the number of prior annotations, which means that as the number of prior annotations increases, the second debugging error update coefficient will also change accordingly. This quantitative relationship enables the system to make more refined adjustments to the algorithm based on the useful information in the sample set.

[0116] After determining the first debugging error update coefficient and the second debugging error update coefficient, the system can determine the first algorithm positioning and tracking error of the past microseismic waveform data based on these two coefficients, the past microseismic event positioning and tracking labels, and the prior microseismic event positioning and tracking annotations. This error is an important indicator for evaluating the positioning and tracking performance of the algorithm, and is also an important basis for the subsequent optimization algorithm.

[0117] Through the processing of the first debugging cycle, the system has obtained a preliminary optimized second-order differential grid wave velocity algorithm. However, this is not the final result. In order to further improve the positioning accuracy and tracking efficiency, the system needs to be optimized in the second debugging cycle.

[0118] In the second debugging cycle, the system will further optimize the positioning and tracking error of the second algorithm based on the second-order differential grid wave velocity algorithm debugged in the first debugging cycle. The second algorithm positioning and tracking error in the embodiment of the present invention refers to the algorithm positioning and tracking error corresponding to the past microseismic waveform data in the second AI algorithm debugging sample set. Unlike the first debugging cycle, the past microseismic waveform data in the second AI algorithm debugging sample set has at least one prior microseismic event positioning and tracking annotation. This means that there is a certain correlation between these annotations and actual microseismic events, thereby providing more useful information for positioning and tracking.

[0119] In the second debugging cycle, the system made full use of these associated prior annotation information and optimized the second-order difference grid wave velocity algorithm more finely. By adjusting the algorithm parameters and formulating the optimization strategy, the system successfully reduced the positioning error and improved the positioning accuracy and tracking efficiency.

[0120] After double-cycle debugging, the second-order differential grid wave velocity algorithm has been significantly optimized and improved. At this point, the microseismic event location and tracking system has the ability to locate and track microseismic events with high precision. However, in order to ensure the stability and reliability of the algorithm, the system still needs to undergo a series of tests and verifications.

[0121] During the testing and verification phase, the system selected several representative microseismic events as test samples and used the optimized second-order differential grid wave velocity algorithm to locate and track these samples. By comparing and analyzing the actual microseismic events, the system successfully verified the positioning accuracy and tracking efficiency of the algorithm. At the same time, the system also tested the adaptability of the algorithm under different geological conditions. The results showed that the algorithm can maintain high positioning accuracy and stability under various complex geological conditions.

[0122] In summary, the microseismic event positioning and tracking system of the embodiment of the present invention realizes high-precision positioning and tracking of microseismic events by combining the second-order differential grid wave velocity algorithm and AI algorithm debugging. It not only solves the problems existing in the traditional microseismic event positioning method, but also provides a strong guarantee for the safety and efficiency in many fields such as geological engineering, mining and tunnel construction. In detail, firstly, through dual-cycle debugging and refined algorithm optimization strategies, the positioning accuracy of microseismic events is significantly improved; secondly, by optimizing the algorithm and processing flow, the tracking efficiency of microseismic events is effectively improved; thirdly, the adaptability of the algorithm under different geological conditions is enhanced, so that the system can play a role in a wider range of application scenarios; finally, it provides a strong guarantee for the safety and efficiency of related fields.

[0123] In some preferred embodiments, in the second debugging cycle, the debugging error update coefficient corresponding to the past microseismic waveform data is determined based on the past microseismic waveform state vector of the past microseismic waveform data and the past microseismic event positioning and tracking label, including: based on the microseismic event type of the past microseismic waveform data and the past microseismic event positioning and tracking label, determining the first debugging error update coefficient corresponding to the past microseismic waveform data, the first debugging error update coefficient is used to adjust the algorithm positioning and tracking error of the noise disturbance type; based on the number of past classification keywords of the past microseismic waveform data, the number of prior annotations of the prior microseismic event positioning and tracking annotations, and the number of second classification keywords in the second AI algorithm debugging sample set, determining the second debugging error update coefficient corresponding to the past microseismic waveform data. Then, the algorithm positioning and tracking error of the past microseismic waveform data is determined based on the debugging error update coefficient, the past microseismic event positioning and tracking labels, and the prior microseismic event positioning and tracking annotations, including: determining the second algorithm positioning and tracking error of the past microseismic waveform data based on the first debugging error update coefficient, the second debugging error update coefficient, the past microseismic event positioning and tracking labels, and the prior microseismic event positioning and tracking annotations.

[0124] In the implementation process of the microseismic event location and tracking system, a series of fine debugging and optimization are required to ensure the accuracy and efficiency of the system. Especially when dealing with complex and changeable microseismic waveform data, the system needs to go through multiple debugging cycles to gradually reduce the positioning and tracking error and improve the overall positioning accuracy. Next, it will be analyzed in detail how the microseismic event location and tracking system determines the debugging error update coefficient corresponding to the past microseismic waveform data based on the past microseismic waveform state vector of the past microseismic waveform data and the past microseismic event location and tracking label in the second debugging cycle, and determines the algorithm location and tracking error accordingly.

[0125] In the second debugging cycle, the primary task of the microseismic event positioning and tracking system is to determine the first debugging error update coefficient corresponding to the past microseismic waveform data based on the microseismic event types and the past microseismic event positioning and tracking labels of the past microseismic waveform data. The main function of this coefficient is to adjust the algorithm positioning and tracking error of the noise disturbance type. In practical applications, microseismic events are often accompanied by various noise disturbances, such as environmental noise, equipment noise, etc. These noises will interfere with the positioning of microseismic events and cause positioning errors. Therefore, by introducing the first debugging error update coefficient, the microseismic event positioning and tracking system can dynamically adjust the algorithm according to the characteristics of the noise disturbance to reduce the positioning error caused by noise.

[0126] In addition to considering the types of microseismic events and the location and tracking labels of past microseismic events, the system also needs to comprehensively consider the number of past classification keywords in the past microseismic waveform data, the number of prior annotations of the prior microseismic event location and tracking annotations, and the number of second classification keywords in the second AI algorithm debugging sample set to determine the second debugging error update coefficient corresponding to the past microseismic waveform data. Here, the number of past classification keywords refers to the number of keywords with specific meanings contained in the past microseismic waveform data, which can reflect certain characteristics or attributes of microseismic events. The number of prior annotations characterizes the number of past microseismic waveform data with prior microseismic event location and tracking annotations. These annotations provide the system with additional useful information and help to more accurately locate microseismic events. The number of second classification keywords refers to the total number of keywords contained in the second AI algorithm debugging sample set, which reflects the diversity and complexity of the sample set.

[0127] By comprehensively considering these three factors, the microseismic event location tracking system can more comprehensively understand the characteristics of past microseismic waveform data and determine the second debugging error update coefficient accordingly. This coefficient works together with the first debugging error update coefficient to fine-tune the algorithm to further reduce the positioning error.

[0128] After determining the first debugging error update coefficient and the second debugging error update coefficient, the microseismic event location tracking system can determine the second algorithm location tracking error of the past microseismic waveform data based on these two coefficients, the past microseismic event location tracking labels, and the prior microseismic event location tracking annotations. This error is an important indicator for evaluating the algorithm location tracking performance and an important basis for subsequent optimization algorithms. By continuously reducing the second algorithm location tracking error, the microseismic event location tracking system can gradually improve the location accuracy and tracking efficiency of microseismic events.

[0129] During the specific implementation process, the microseismic event location and tracking system uses advanced algorithms and technical means to conduct in-depth analysis and processing of past microseismic waveform data. First, the system preprocesses the past microseismic waveform data, such as denoising and filtering, to improve the quality of the data. Then, the system uses the second-order differential grid wave velocity algorithm to perform positioning processing on the preprocessed data to obtain preliminary positioning results. Next, the system combines the AI ​​algorithm to further optimize and adjust the positioning results to reduce the positioning error. In this process, the system makes full use of useful information such as past microseismic waveform state vectors, past microseismic event location and tracking labels, and prior microseismic event location and tracking annotations to fine-tune and optimize the algorithm.

[0130] Through the processing and optimization of the second debugging cycle, the microseismic event location and tracking system can significantly improve the location accuracy and tracking efficiency of microseismic events. This not only provides a strong guarantee for the safety and efficiency of many fields such as geological engineering, mining and tunnel construction, but also provides important data support and technical means for scientific research and technological innovation in related fields.

[0131] In summary, the microseismic event location and tracking system determines the debugging error update coefficient by comprehensively considering multiple factors of past microseismic waveform data in the second debugging cycle, and determines the algorithm location and tracking error accordingly. In this way, the location accuracy and tracking efficiency of microseismic events are improved.

[0132] In another preferred embodiment, the second debugging error update coefficient corresponding to the past microseismic waveform data is determined based on the number of past classification keywords possessed by the past microseismic waveform data, the number of prior annotations of the prior microseismic event positioning and tracking annotations, and the number of second classification keywords in the second AI algorithm debugging sample set, including: on the basis that the number of past classification keywords possessed by the past microseismic waveform data is a set number, the second debugging error update coefficient corresponding to the past microseismic waveform data is determined according to the number of prior annotations of the prior microseismic event positioning and tracking annotations and the number of second classification keywords in the second AI algorithm debugging sample set. ; On the basis that the number of past classification keywords possessed by the past microseismic waveform data is greater than the set number, the a priori learning confidence of each prior microseismic event positioning and tracking annotation is determined according to the a priori annotation number of each prior microseismic event positioning and tracking annotation, and the a priori microseismic event positioning and tracking annotation with the least number of prior annotations is determined as the target a priori microseismic event positioning and tracking annotation; based on the second number of classification keywords and the confidence comparison result between the a priori learning confidence of the target a priori microseismic event positioning and tracking annotation and the a priori learning confidence of other prior microseismic event positioning and tracking annotations, the second debugging error update coefficient corresponding to the past microseismic waveform data is determined.

[0133] For example, in the excavation process of underground caverns in large hydropower stations, the microseismic event positioning and tracking system plays a vital role. Since the excavation of underground caverns involves complex geological structures and a large amount of rock crushing operations, microseismic events occur frequently, posing a serious threat to construction safety and efficiency. Therefore, accurate and real-time positioning and tracking of microseismic events are of great significance for preventing geological disasters and ensuring construction safety. Next, we will analyze in detail how the microseismic event positioning and tracking system can determine the second debugging error update coefficient corresponding to the past microseismic waveform data in the scenario of underground cavern excavation in large hydropower stations based on the number of past classification keywords in the past microseismic waveform data, the number of prior annotations of the prior microseismic event positioning and tracking annotations, and the number of second classification keywords in the second AI algorithm debugging sample set, and optimize the positioning and tracking effect of microseismic events accordingly.

[0134] In this application scenario, the microseismic event location and tracking system will first collect a large amount of past microseismic waveform data. These data contain various characteristic information of microseismic events and are the basis for the system to locate and track. Among them, the number of past classification keywords is an important indicator, which reflects the number of keywords with specific meanings contained in the past microseismic waveform data. These keywords can reveal certain characteristics or attributes of microseismic events, which is of great significance for improving the accuracy of location and tracking.

[0135] In addition to the number of past classification keywords, the number of prior annotations of prior microseismic event location tracking annotations is also an important factor that the system needs to consider. The number of prior annotations characterizes the number of past microseismic waveform data with prior microseismic event location tracking annotations. These annotations provide the system with additional useful information, which helps to locate microseismic events more accurately. During the excavation of underground caverns in large hydropower stations, the number of prior annotations is often very limited due to the complexity of geological conditions and construction environment. Therefore, the system needs to make full use of this limited prior information to improve the accuracy of positioning and tracking.

[0136] In addition, the number of second-category keywords in the second AI algorithm debugging sample set is also a factor that needs to be considered when determining the second debugging error update coefficient. The number of second-category keywords reflects the diversity and complexity of the sample set. In the excavation scenario of underground caverns in large hydropower stations, the types and characteristics of microseismic events are also diverse due to the diversity of geological conditions and construction environments. Therefore, the system needs to ensure that the positioning and tracking algorithm can adapt to different types of microseismic events by considering the number of second-category keywords.

[0137] After determining the number of past classification keywords, the number of prior annotations, and the number of second classification keywords, the microseismic event location and tracking system begins to determine the second debugging error update coefficient corresponding to the past microseismic waveform data. This process is divided into two cases:

[0138] When the number of past classification keywords in the past microseismic waveform data is the set number, the system determines the second debugging error update coefficient corresponding to the past microseismic waveform data based on the prior annotation number of the prior microseismic event positioning and tracking annotations and the number of second classification keywords in the second AI algorithm debugging sample set. In this process, the system comprehensively considers the scarcity of the prior annotation number and the diversity of the second classification keyword number to make fine adjustments to the algorithm. By adjusting the second debugging error update coefficient, the system can gradually reduce the positioning and tracking error and improve the accuracy of positioning and tracking.

[0139] When the number of past classification keywords in the past microseismic waveform data is greater than the set number, the system adopts a more complex strategy to determine the second debugging error update coefficient. First, the system determines the prior learning confidence of each prior microseismic event location tracking annotation based on the number of prior annotations of each prior microseismic event location tracking annotation. The prior learning confidence reflects the system's degree of trust in each prior annotation. In the excavation scenario of underground caverns in large hydropower stations, the reliability of different prior annotations may vary due to the complexity of geological conditions and construction environment. Therefore, the system needs to evaluate the credibility of each prior annotation by calculating the prior learning confidence.

[0140] Next, the system determines the prior microseismic event location tracking annotation with the least prior annotation number as the target prior microseismic event location tracking annotation. This is because when the number of prior annotations is limited, the annotation with the least prior annotation number often contains the most scarce and useful information. By focusing on these annotations, the system can more effectively improve the accuracy of location tracking.

[0141] Finally, the system determines the second debugging error update coefficient corresponding to the past microseismic waveform data based on the number of second-category keywords and the confidence comparison between the prior learning confidence of the target prior microseismic event location tracking annotation and the prior learning confidence of other prior microseismic event location tracking annotations. In this process, the system comprehensively considers the diversity of the number of second-category keywords and the scarcity and credibility of the target prior annotations to make more refined adjustments to the algorithm. Through this strategy, the system can further reduce the positioning and tracking error and improve the accuracy of positioning and tracking.

[0142] In this way, by comprehensively considering factors such as the number of past classification keywords, the number of prior annotations, and the number of second classification keywords, the system can more accurately locate the location and scale of microseismic events, providing more reliable protection for construction safety. By optimizing the algorithm and debugging the error update coefficient, the system can respond to changes in microseismic events more quickly, achieve real-time tracking and early warning, and provide strong support for improving construction efficiency. By considering the diversity of geological conditions and construction environments, the system can better adapt to different types of microseismic events and improve the accuracy and stability of positioning and tracking.

[0143] In some optional embodiments, the microseismic event positioning and tracking of past microseismic waveform data in the AI ​​algorithm debugging sample set is performed through the second-order differential grid wave velocity algorithm to obtain the past microseismic event positioning and tracking labels corresponding to each past microseismic waveform data, including: passing each past microseismic waveform data in the AI ​​algorithm debugging sample set into the event element mining branch in the second-order differential grid wave velocity algorithm to obtain the past event element information corresponding to each past microseismic waveform data generated by the event element mining branch; passing the past event element information into the event positioning and tracking branch in the second-order differential grid wave velocity algorithm to obtain the past microseismic event positioning and tracking labels corresponding to each past microseismic waveform data generated by the event positioning and tracking branch.

[0144] During the excavation of underground caverns in large hydropower stations, the microseismic event location and tracking system can accurately locate and track microseismic events through advanced algorithms and technical means, providing a strong guarantee for construction safety and efficiency. Next, we will analyze in detail how the microseismic event location and tracking system uses the second-order differential grid wave velocity algorithm to locate and track microseismic events in the past microseismic waveform data in the AI ​​algorithm debugging sample set, and obtain the past microseismic event location and tracking labels corresponding to each past microseismic waveform data.

[0145] First, we need to understand the basic principle of the second-order difference grid wave velocity algorithm. This is a wave velocity calculation method based on grid division. By performing differential processing on waveform data, the propagation speed and direction of the waveform are obtained. In the microseismic event location and tracking system, the second-order difference grid wave velocity algorithm is used to process past microseismic waveform data to achieve the location and tracking of microseismic events.

[0146] Specifically, the microseismic event location and tracking system passes each past microseismic waveform data in the AI ​​algorithm debugging sample set to the event element mining branch in the second-order differential grid wave velocity algorithm. The main task of this branch is to conduct in-depth analysis and mining of waveform data to extract key element information related to microseismic events. These element information may include the occurrence time, location, scale, and waveform characteristics of microseismic events.

[0147] In the event element mining branch, the system preprocesses the incoming past microseismic waveform data, such as filtering and denoising, to improve the signal-to-noise ratio and reliability of the data. Subsequently, the system extracts features from the processed waveform data, such as calculating the amplitude, frequency, phase and other parameters of the waveform to obtain key feature information of the microseismic event. Finally, based on this feature information, the system generates past event element information corresponding to each past microseismic waveform data.

[0148] After obtaining the past event element information, the microseismic event location and tracking system transmits this information to the event location and tracking branch in the second-order difference grid wave velocity algorithm. The main task of this branch is to locate and track microseismic events based on the event element information using the second-order difference grid wave velocity algorithm.

[0149] In the event location and tracking branch, the system constructs a two-dimensional or three-dimensional grid model based on the time, location and other parameters in the past event element information. Then, the system applies the second-order differential grid wave velocity algorithm on the grid model to calculate the propagation speed and direction of the waveform. Through continuous iteration and optimization, the system can gradually approach the true location and scale of microseismic events and achieve accurate location and tracking of microseismic events.

[0150] Finally, the event location tracking branch generates past microseismic event location tracking tags corresponding to each past microseismic waveform data. These tags contain key information such as the location, scale, and time of occurrence of microseismic events, providing important data support for subsequent microseismic event warnings and disaster assessments.

[0151] In this way, through the second-order difference grid wave velocity algorithm, the system can more accurately extract the key element information related to the microseismic event and accurately locate the microseismic event. This helps construction personnel to more accurately judge the location and scale of the microseismic event, take corresponding safety measures in time, and avoid potential geological disaster risks. Through the event location tracking branch, the system can grasp the changes of microseismic events at any time and provide timely warning and decision support for construction personnel. This helps construction personnel to adjust the construction plan and take countermeasures in time to ensure the safety and smooth progress of the construction. By accurately locating and tracking microseismic events, the system can help construction personnel more accurately judge the stability of geological conditions and construction environment, and provide strong data support for the optimization and adjustment of construction plans. This helps to reduce unnecessary construction waste and duplication of labor and improve construction efficiency and quality.

[0152] In some other exemplary embodiments, the microseismic event locating and tracking method also includes: calling the debugged second-order differential grid wave velocity algorithm to perform microseismic event locating and tracking on the target microseismic waveform monitoring data of each geological monitoring task in the spatiotemporal geological monitoring log, and determining the target locating and tracking features corresponding to the target microseismic waveform monitoring data of each geological monitoring task, wherein the spatiotemporal geological monitoring log includes at least one spatiotemporal geological monitoring data set and target microseismic waveform monitoring data of at least two geological monitoring tasks; and performing monitoring status identification on the spatiotemporal geological monitoring log according to the target locating and tracking features corresponding to the target microseismic waveform monitoring data of each geological monitoring task, and obtaining a status identification result of the spatiotemporal geological monitoring log monitoring.

[0153] During the excavation of underground caverns in large hydropower stations, the microseismic event location and tracking system can not only accurately locate and track past microseismic waveform data, but also effectively process and analyze the target microseismic waveform monitoring data in real-time geological monitoring tasks. Next, we will analyze in detail how the microseismic event location and tracking system can locate and track microseismic events for the target microseismic waveform monitoring data of each geological monitoring task in the spatiotemporal geological monitoring log by calling the debugged second-order differential grid wave velocity algorithm, and identify the monitoring status based on the obtained target location and tracking features, so as to achieve more accurate and efficient geological disaster warning and construction safety management.

[0154] First of all, it is necessary to emphasize again the importance of the spatiotemporal geological monitoring log. This is a comprehensive log that contains at least one spatiotemporal geological monitoring data set and target microseismic waveform monitoring data of at least two geological monitoring tasks. The spatiotemporal geological monitoring data set refers to a data set obtained by continuous monitoring of geological activities within a certain time and space range. It covers various information of geological activities, such as the propagation speed, direction, amplitude, etc. of seismic waves. The geological monitoring task refers to a specific monitoring activity, such as regular microseismic waveform monitoring of a certain area of ​​the underground cavern to obtain geological activity information in the area.

[0155] When processing the spatiotemporal geological monitoring log, the microseismic event location and tracking system will first call the debugged second-order differential grid wave velocity algorithm. This algorithm has been introduced in detail in the previous embodiment. It can conduct in-depth analysis and mining of microseismic waveform data, extract key element information related to microseismic events, such as the occurrence time, location, scale and waveform characteristics of microseismic events, and realize accurate location and tracking of microseismic events.

[0156] When the microseismic event location and tracking system starts processing the target microseismic waveform monitoring data in the spatiotemporal geological monitoring log, it will pass this data into the second-order differential grid wave velocity algorithm. The algorithm will preprocess this data, such as filtering and denoising, to improve the signal-to-noise ratio and reliability of the data. Subsequently, the algorithm will perform feature extraction on the processed waveform data, which is the core part of the algorithm. Feature extraction refers to extracting parameters that can reflect the characteristics of microseismic events from waveform data, such as amplitude, frequency, phase, etc. These parameters are the key to subsequent microseismic event location and tracking.

[0157] After extracting the characteristic information of the target microseismic waveform monitoring data, the second-order differential grid wave velocity algorithm will use this information to locate and track microseismic events. It will build a two-dimensional or three-dimensional grid model on the spatiotemporal geological monitoring data set based on the propagation speed and direction of the waveform. This model divides the geological space into multiple grids, each of which represents a small geological area. Then, the algorithm will iterate and optimize on the grid model, gradually approaching the actual location and scale of the microseismic event by calculating the propagation path and time of the waveform in the grid, and realizing the precise positioning and tracking of the microseismic event.

[0158] After completing the location and tracking of microseismic events, the microseismic event location and tracking system generates target location and tracking features corresponding to the target microseismic waveform monitoring data of each geological monitoring task. These features include key information such as the location, scale, occurrence time, and waveform characteristics of the microseismic event. This information is of great value for subsequent monitoring status identification because it can reflect the status and change trend of geological activities.

[0159] Next, the microseismic event positioning and tracking system identifies the monitoring status of the spatiotemporal geological monitoring log based on the target positioning and tracking features corresponding to the target microseismic waveform monitoring data of each geological monitoring task. The purpose of this step is to determine whether the current geological monitoring status is normal and whether there is a potential risk of geological disasters. To achieve this goal, the system first performs data analysis on the target positioning and tracking features to extract key indicators, such as the frequency and scale changes of microseismic events. The system then compares these indicators with preset thresholds or models. These thresholds and models are pre-set based on historical data and geological knowledge, and they represent the normal range and abnormal standards of geological activities. By comparison, the system can determine whether the current geological monitoring status is within the normal range and whether there is a potential risk of geological disasters.

[0160] If the system finds that an indicator exceeds the preset threshold or the range predicted by the model, it will issue an alarm to remind the construction personnel that there may be a potential risk of geological disasters. At the same time, the system will also generate the status identification results of the spatiotemporal geological monitoring log monitoring. This result includes information such as the current geological monitoring status, potential geological disaster risks, and recommended countermeasures. This result is of great reference value to construction personnel, helping them to better understand the changes in geological conditions, take corresponding safety measures in a timely manner, and ensure the safety and smooth progress of construction.

[0161] In this way, the microseismic event location and tracking system realizes the precise location and tracking of the target microseismic waveform monitoring data of each geological monitoring task in the spatiotemporal geological monitoring log in the underground cavern excavation scene of a large hydropower station, and effectively identifies the monitoring state based on the obtained target location and tracking features. By calling the debugged second-order difference grid wave velocity algorithm to process and analyze the target microseismic waveform monitoring data, the microseismic event location and tracking system can more accurately extract the key element information related to the microseismic event and realize the precise location and tracking of the microseismic event. At the same time, since the system can process real-time geological monitoring data, it can greatly improve the accuracy and real-time performance of geological monitoring and provide timely and accurate geological information for construction personnel. By locating, tracking and extracting features of the target microseismic waveform monitoring data, the microseismic event location and tracking system can timely discover potential geological disaster risks and issue an alarm to prompt construction personnel to take corresponding safety measures. This helps to reduce the impact of geological disasters on construction safety and progress, ensure the smooth progress of construction, and also reduce the safety threats of geological disasters to personnel and equipment. By identifying the monitoring status of geological monitoring logs in the time and space domain, the microseismic event location tracking system can help construction personnel better understand the changes in geological conditions and timely optimize and adjust the construction plan. This helps to reduce unnecessary construction waste and duplication of labor and improve construction efficiency and quality. At the same time, since the system can provide accurate geological disaster warning information, it can also help construction personnel avoid potential geological disaster risks, ensure the safety of construction, and further guarantee the quality and progress of the project.

[0162] In some replaceable embodiments, the monitoring status of the spatiotemporal geological monitoring log is identified based on the target positioning tracking features corresponding to the target microseismic waveform monitoring data of each geological monitoring task to obtain the status identification result of the spatiotemporal geological monitoring log monitoring, including: generating the microseismic trend deduction vector of each geological monitoring task in the spatiotemporal geological monitoring data set based on the target positioning tracking features corresponding to the target microseismic waveform monitoring data of each geological monitoring task in the same spatiotemporal geological monitoring data set, and the microseismic event discrimination weight corresponding to the target positioning tracking features, the microseismic event discrimination weight being the event discrimination viewpoint obtained by performing microseismic event positioning and tracking on the target microseismic waveform monitoring data through the second-order differential grid wave velocity algorithm; the monitoring status of the spatiotemporal geological monitoring log is identified based on the microseismic trend deduction vector corresponding to each geological monitoring task to obtain the status identification result of the spatiotemporal geological monitoring log monitoring.

[0163] During the excavation of underground caverns in large hydropower stations, the microseismic event location and tracking system can not only accurately locate and track microseismic waveform data, but also accurately identify the geological monitoring status based on these data, thereby providing a strong guarantee for construction safety and efficiency. Next, we will analyze in detail how the microseismic event location and tracking system can achieve monitoring status identification of geological monitoring logs in the spatiotemporal domain through a series of complex calculations and processing steps, and deduce the beneficial effects of this technical solution.

[0164] First of all, it is necessary to clarify the importance of the spatiotemporal geological monitoring log. This log contains at least one spatiotemporal geological monitoring data set and target microseismic waveform monitoring data of at least two geological monitoring tasks. The spatiotemporal geological monitoring data set refers to a data set obtained by continuous monitoring of geological activities within a certain time and space range, which covers various information of geological activities. The geological monitoring task refers to a specific monitoring activity, such as regular microseismic waveform monitoring of a certain area of ​​the underground cavern to obtain geological activity information in the area.

[0165] In an alternative embodiment, the microseismic event location tracking system generates microseismic trend deduction vectors for each geological monitoring task in the spatiotemporal geological monitoring data set based on the target location tracking features corresponding to the target microseismic waveform monitoring data of each geological monitoring task and the microseismic event discrimination weights corresponding to these features. This step is the key to the system's identification of the geological monitoring status.

[0166] Specifically, the microseismic event location and tracking system first processes the target microseismic waveform monitoring data of each geological monitoring task in the same spatiotemporal geological monitoring data set. These data are obtained by arranging microseismic monitoring instruments at different locations in the underground caverns and recording microseismic waveforms in real time. The system pre-processes these data, such as filtering and denoising, to improve the signal-to-noise ratio and reliability of the data.

[0167] Next, the system calls the debugged second-order differential grid wave velocity algorithm to locate and track the microseismic events on these preprocessed target microseismic waveform monitoring data. This algorithm can deeply analyze and mine microseismic waveform data, extract key element information related to microseismic events, such as the time, location, scale, and waveform characteristics of microseismic events, and achieve accurate positioning and tracking of microseismic events. During the positioning and tracking process, the algorithm will also build a two-dimensional or three-dimensional grid model on the spatiotemporal geological monitoring data set based on the propagation speed and direction of the waveform, and approximate the true location and scale of the microseismic event through iteration and optimization.

[0168] After completing the positioning and tracking of microseismic events, the microseismic event positioning and tracking system generates target positioning and tracking features corresponding to the target microseismic waveform monitoring data of each geological monitoring task. These features include key information such as the location, scale, occurrence time, and waveform characteristics of the microseismic event. At the same time, the system will also perform microseismic event positioning and tracking results on the target microseismic waveform monitoring data based on the second-order differential grid wave velocity algorithm, and assign a microseismic event discrimination weight to each target positioning and tracking feature. This weight represents the importance or reliability of the feature in judging the type and nature of microseismic events.

[0169] With the target positioning and tracking features and the corresponding microseismic event discrimination weights, the microseismic event positioning and tracking system further generates microseismic trend deduction vectors for each geological monitoring task in the spatiotemporal geological monitoring data set. This vector is a mathematical representation that integrates multiple target positioning and tracking features and corresponding weights. It can reflect the trend and change law of microseismic activity of geological monitoring tasks in the spatiotemporal geological monitoring data set. Specifically, the system maps them to a multidimensional vector space according to the type and nature of the target positioning and tracking features, and adjusts the position and direction of these features in the vector space according to the corresponding microseismic event discrimination weights. In this way, each geological monitoring task will correspond to a unique microseismic trend deduction vector, which can accurately describe the microseismic activity of the task in the spatiotemporal geological monitoring data set.

[0170] Finally, the microseismic event location and tracking system identifies the monitoring status of the spatiotemporal geological monitoring logs based on the microseismic trend deduction vectors corresponding to each geological monitoring task. The purpose of this step is to determine whether the current geological monitoring status is normal and whether there is a potential risk of geological disasters. To achieve this goal, the system first performs data analysis on the microseismic trend deduction vectors and extracts key indicators, such as the frequency and scale changes of microseismic events. The system then compares these indicators with preset thresholds or models. These thresholds and models are pre-set based on historical data and geological knowledge, and they represent the normal range and abnormal standards of geological activities. By comparison, the system can determine whether the current geological monitoring status is within the normal range and whether there is a potential risk of geological disasters.

[0171] If the system finds that an indicator exceeds the preset threshold or the range predicted by the model, it will issue an alarm to remind the construction personnel that there may be a potential risk of geological disasters. At the same time, the system will also generate the status identification results of the spatiotemporal geological monitoring log monitoring. This result includes information such as the current geological monitoring status, potential geological disaster risks, and recommended countermeasures. This result is of great reference value to construction personnel, helping them to better understand the changes in geological conditions, take corresponding safety measures in a timely manner, and ensure the safety and smooth progress of construction.

[0172] In this way, the microseismic event positioning and tracking system realizes the precise positioning and tracking of the target microseismic waveform monitoring data of each geological monitoring task in the spatiotemporal geological monitoring log in the underground cavern excavation scenario of a large hydropower station, and generates a microseismic trend deduction vector based on the obtained target positioning and tracking characteristics and microseismic event discrimination weights, thereby realizing accurate identification of the geological monitoring status.

[0173] In some preferred technical schemes, the microseismic trend deduction vector of each geological monitoring task in the same spatiotemporal geological monitoring data set is generated based on the target positioning and tracking features corresponding to the target microseismic waveform monitoring data of each geological monitoring task in the spatiotemporal geological monitoring data set, and the microseismic event discrimination weights corresponding to the target positioning and tracking features, including: on the basis that the target microseismic waveform monitoring data has a non-linked target positioning and tracking feature, the risk prediction coefficient of the target microseismic waveform monitoring data is determined based on the microseismic event discrimination weights corresponding to the target positioning and tracking features; on the basis that the target microseismic waveform monitoring data has at least two target positioning and tracking features, the disaster-affecting thermal variables of each target positioning and tracking feature are determined based on the microseismic event discrimination weights corresponding to each target positioning and tracking feature; the risk prediction coefficient of the target microseismic waveform monitoring data is determined based on the microseismic event discrimination weights corresponding to the target positioning and tracking feature with the highest disaster-affecting thermal variables; and on the basis of the priority features of each target microseismic waveform monitoring data of the geological monitoring task in the spatiotemporal geological monitoring data set and the risk prediction coefficient, the microseismic trend deduction vector of the geological monitoring task in the spatiotemporal geological monitoring data set is generated.

[0174] During the excavation of underground caverns of large hydropower stations, the spatiotemporal geological monitoring data set refers to the data set obtained by continuous monitoring of geological activities within a certain time and space range, which covers various information of geological activities. The geological monitoring task refers to specific monitoring activities, such as regular microseismic waveform monitoring of a certain area of ​​the underground cavern to obtain geological activity information in the area. The target microseismic waveform monitoring data is the microseismic waveform data actually monitored in the geological monitoring task, which contains the key information of the microseismic event. The target positioning and tracking features are the features related to the microseismic event extracted from the target microseismic waveform monitoring data, such as the occurrence time, location, and scale of the microseismic event. The microseismic event discrimination weight is the weight given when discriminating the microseismic event based on these features, which reflects the importance of the feature in the discrimination process.

[0175] In the preferred technical solution, the microseismic event location and tracking system will first process and analyze the target microseismic waveform monitoring data to extract the target location and tracking features. These features may include the location, scale, occurrence time, waveform characteristics, etc. of the microseismic event, which can fully describe the nature and state of the microseismic event. At the same time, the system will also distinguish the microseismic event based on these features and assign corresponding microseismic event discrimination weights. This weight represents the importance or reliability of the feature in distinguishing the type and nature of the microseismic event.

[0176] Next, the microseismic event location and tracking system will determine the risk prediction coefficient of the target microseismic waveform monitoring data based on the microseismic event discrimination weight corresponding to the target location and tracking feature. The purpose of this step is to evaluate the risk level of the geological activity represented by the target microseismic waveform monitoring data to construction safety. The system will weight the target location and tracking features according to the size of the microseismic event discrimination weight to obtain a comprehensive risk prediction coefficient. The higher this coefficient is, the greater the risk of the geological activity represented by the target microseismic waveform monitoring data to construction safety.

[0177] When the target microseismic waveform monitoring data has at least two target positioning and tracking features, the microseismic event positioning and tracking system will further analyze the correlation and impact between these features. The system will determine the disaster impact thermal variables of each feature based on the microseismic event discrimination weights corresponding to each target positioning and tracking feature. This variable reflects the degree of influence of the feature on the occurrence and development of geological disasters. The system will sort the disaster impact thermal variables, find the microseismic event discrimination weight corresponding to the highest feature, and determine the risk prediction coefficient of the target microseismic waveform monitoring data accordingly. The purpose of this step is to further refine the assessment of geological activity risks and ensure that the geological activities that pose the greatest threat to construction safety can be accurately identified.

[0178] With the risk prediction coefficient, the microseismic event location tracking system will generate the microseismic trend deduction vector of the geological monitoring task in the spatiotemporal geological monitoring data set according to the priority characteristics and risk prediction coefficient of each target microseismic waveform monitoring data of the geological monitoring task in the spatiotemporal geological monitoring data set. The purpose of this step is to represent the trend and change law of microseismic activity in the geological monitoring task in the form of vectors, so as to facilitate the subsequent monitoring status identification and geological disaster early warning. The system will construct a multidimensional vector space based on the time sequence, spatial position and risk prediction coefficient of the target microseismic waveform monitoring data in the spatiotemporal geological monitoring data set. In this vector space, each target microseismic waveform monitoring data will correspond to a vector, and the direction and length of the vector represent the trend and intensity of microseismic activity. By synthesizing and calculating these vectors, the system can obtain the microseismic trend deduction vector of the geological monitoring task in the spatiotemporal geological monitoring data set.

[0179] Finally, the microseismic event location and tracking system will apply the generated microseismic trend deduction vector to the monitoring status identification of the spatiotemporal geological monitoring log. The system will analyze the changing trend and intensity characteristics of the microseismic trend deduction vector and compare it with the preset threshold or model. If the trend of microseismic activity is found to be abnormal or exceeds the preset range, the system will issue an alarm to remind the construction personnel that there may be potential geological disaster risks. At the same time, the system will also generate the status identification results of the spatiotemporal geological monitoring log monitoring, including the current geological monitoring status, potential geological disaster risks, and recommended countermeasures. This result is of great reference value to the construction personnel, which can help them better understand the changes in geological conditions, take corresponding safety measures in time, and ensure the safety and smooth progress of the construction.

[0180] In this way, the microseismic event location and tracking system realizes the precise location and tracking of the target microseismic waveform monitoring data of each geological monitoring task in the spatiotemporal geological monitoring log in the underground cavern excavation scenario of a large hydropower station, and generates a microseismic trend deduction vector based on the obtained target location and tracking features and microseismic event discrimination weights. By extracting the target location and tracking features in the target microseismic waveform monitoring data and calculating the risk prediction coefficient in combination with the microseismic event discrimination weights, the microseismic event location and tracking system can more accurately assess the risk level of geological activities to construction safety and issue alarms in a timely manner. This helps to improve the accuracy and real-time performance of geological monitoring and ensure that construction personnel can take corresponding safety measures in a timely manner. Through in-depth analysis and processing of the target microseismic waveform monitoring data, the microseismic event location and tracking system can discover potential geological disaster risks and generate microseismic trend deduction vectors for representation. This enables construction personnel to more intuitively understand the changing trends and intensity characteristics of geological activities, so that they can take timely countermeasures to reduce the impact of geological disasters on construction safety and progress. By generating microseismic trend deduction vectors and identifying monitoring status, the microseismic event location and tracking system can help construction personnel better understand the changes in geological conditions and timely optimize and adjust construction plans. This helps reduce unnecessary construction waste and duplication of labor and improve construction efficiency and quality. At the same time, through effective early warning of geological disasters and suggestions for response measures, the system can also help construction personnel reduce construction risks and ensure the smooth progress of the project.

[0181] In the following technical scheme, the disaster-affecting thermal variables of each target positioning and tracking feature are determined based on the microseismic event discrimination weights corresponding to each target positioning and tracking feature, including: feature space mapping of the microseismic event discrimination weights corresponding to each target positioning and tracking feature to obtain the disaster-affecting thermal variables of each target positioning and tracking feature, and the disaster-affecting thermal variables have a second quantitative relationship with the microseismic event discrimination weights; or, based on the microseismic event discrimination weights and weight threshold values ​​corresponding to each target positioning and tracking feature, the grid node association weights corresponding to each target positioning and tracking feature are determined; based on the grid node association weights corresponding to each target positioning and tracking feature, the disaster-affecting thermal variables of each target positioning and tracking feature are determined, and the disaster-affecting thermal variables have a second quantitative relationship with the grid node association weights.

[0182] During the excavation of underground caverns of large hydropower stations, target positioning and tracking features are features related to microseismic events extracted from microseismic waveform monitoring data, such as the time, location, and scale of microseismic events, which can fully describe the nature and state of microseismic events. The microseismic event discrimination weight is the weight given when discriminating microseismic events based on these features, which reflects the importance or reliability of the features in the discrimination process. The disaster impact thermal variable is a variable used to describe the degree of influence of target positioning and tracking features on geological disasters, and its size has a second quantitative relationship with the microseismic event discrimination weight or the grid node association weight.

[0183] In detail, the microseismic event location and tracking system will first perform feature space mapping on the microseismic event discrimination weights corresponding to each target location and tracking feature. The purpose of this step is to convert the microseismic event discrimination weights from the original feature space to a new feature space in order to better describe and compare the degree of influence of different target location and tracking features on geological disasters. Through feature space mapping, the system can obtain the disaster-influencing thermal variables of each target location and tracking feature. This variable has a second quantitative relationship with the microseismic event discrimination weight, that is, there is a certain mathematical relationship between them, and they can be converted to each other through a certain calculation formula.

[0184] Specifically, the feature space mapping process can include a series of mathematical transformations and operations, such as linear transformation, nonlinear transformation, normalization, etc. The purpose of these transformations and operations is to convert the original microseismic event discrimination weights into more comparable and interpretable disaster impact thermal variables. For example, the system can normalize the microseismic event discrimination weights, convert them into values ​​between 0 and 1, and then determine the value range of the disaster impact thermal variables based on the size of these values.

[0185] In addition, an embodiment of the present invention also provides another method for determining thermal variables affected by disasters. The microseismic event location and tracking system can determine the grid node association weight corresponding to each target location and tracking feature based on the microseismic event discrimination weight corresponding to each target location and tracking feature and the weight threshold value. The purpose of this step is to associate the microseismic event discrimination weight with a specific grid node in order to better describe and compare the degree of influence of different target location and tracking features in the occurrence and development of geological disasters.

[0186] The weight threshold is a preset threshold used to determine whether the microseismic event discrimination weight is large enough to have a significant impact on geological hazards. If the microseismic event discrimination weight of a target positioning tracking feature is greater than the weight threshold, the system will consider this feature to have a significant impact on geological hazards and assign it a higher grid node association weight. Otherwise, the system will consider this feature to have a small impact on geological hazards and assign it a lower grid node association weight.

[0187] After determining the grid node association weights corresponding to each target positioning and tracking feature, the microseismic event positioning and tracking system will determine the disaster impact thermal variables of each target positioning and tracking feature based on these weights. This variable has a second quantitative relationship with the grid node association weight, that is, there is also a certain mathematical relationship between them. Through a certain calculation formula, the system can convert the grid node association weights into disaster impact thermal variables, thereby more intuitively describing and comparing the degree of influence of different target positioning and tracking features on geological disasters.

[0188] Through the above two methods, the microseismic event location tracking system can accurately determine the disaster-affecting thermal variables of each target location tracking feature. This variable not only reflects the degree of influence of the target location tracking feature on geological disasters, but can also be used for the subsequent generation of microseismic trend deduction vectors and early warning of geological disasters. Specifically, the system will judge the occurrence and development trend of geological disasters according to the size and change trend of the disaster-affecting thermal variables, and issue an alarm in time to remind construction personnel to take corresponding safety measures. By implementing the above technical scheme, the microseismic event location tracking system can more accurately determine the disaster-affecting thermal variables of each target location tracking feature, thereby improving the accuracy and reliability of geological disaster early warning. Specifically, the beneficial effects are mainly reflected in the following aspects: through the method of feature space mapping or grid node association weight, the microseismic event location tracking system can more accurately describe and compare the degree of influence of different target location tracking features on geological disasters. This enables the system to more accurately judge the occurrence and development trend of geological disasters, thereby improving the accuracy of early warning. Since the disaster-affecting thermal variables are calculated based on the microseismic event discrimination weight or the grid node association weight, they have stronger objectivity and reliability. Compared with traditional early warning methods based on experience or statistical data, this method can better reflect the actual risk level of geological disasters, thereby enhancing the reliability of early warning.

[0189] In an independent embodiment, the monitoring status of the spatiotemporal geological monitoring log is identified based on the microseismic trend deduction vector corresponding to each geological monitoring task to obtain the status identification result of the spatiotemporal geological monitoring log monitoring, including: determining the residual disturbance characteristics of each geological monitoring task in the spatiotemporal geological monitoring data set based on the microseismic trend deduction vector corresponding to each geological monitoring task; performing self-attention feature mining on each geological monitoring task based on the residual disturbance characteristics to obtain the self-attention monitoring characteristics of each geological monitoring task; determining the monitoring correlation between geological monitoring tasks based on the microseismic trend deduction vector corresponding to each geological monitoring task, the monitoring correlation including at least one of trend deduction association characteristics and trend deduction mutually exclusive characteristics; performing cross-attention feature mining based on the monitoring correlation between the geological monitoring tasks to obtain the cross-attention monitoring characteristics between the geological monitoring tasks; generating the status identification result of the spatiotemporal geological monitoring log monitoring based on the self-attention monitoring characteristics of each geological monitoring task and the cross-attention monitoring characteristics between geological monitoring tasks.

[0190] During the excavation of underground caverns of large hydropower stations, geological monitoring tasks refer to the specific tasks of monitoring the geological state during the excavation of underground caverns, such as monitoring the stability of rock mass and the flow of groundwater. The microseismic trend deduction vector is a vector of the geological state change trend deduced by certain algorithms and models based on the microseismic waveform data monitored by the microseismic event positioning and tracking system. The spatiotemporal geological monitoring dataset refers to the dataset containing the spatiotemporal change information of the geological state during the excavation of underground caverns. The residual disturbance feature refers to the difference between the actual monitoring data and the deduced data of the geological monitoring task in the spatiotemporal geological monitoring dataset. Self-attention feature mining refers to the feature extraction of geological monitoring tasks to highlight their own important features and change laws. Monitoring correlation refers to the mutual influence and correlation degree between different geological monitoring tasks in the monitoring process. Cross-attention feature mining refers to the extraction of mutual influence and correlation features between geological monitoring tasks to reveal the complex relationship between them.

[0191] In an embodiment of the present invention, the microseismic event location and tracking system first determines the residual disturbance characteristics of each geological monitoring task in the spatiotemporal geological monitoring data set based on the microseismic trend deduction vector corresponding to each geological monitoring task. The purpose of this step is to find out the abnormalities or disturbances in the geological monitoring task by comparing the actual monitoring data with the deduced data, so as to provide a basis for subsequent monitoring status identification. The system analyzes each geological monitoring task one by one, calculates the difference between its actual monitoring data and the deduced data, and extracts and records these differences as residual disturbance characteristics.

[0192] Next, the microseismic event location and tracking system will mine the self-attention features of each geological monitoring task based on the residual disturbance characteristics to obtain the self-attention monitoring features of each geological monitoring task. The purpose of this step is to highlight the important characteristics and change laws of the geological monitoring task itself through feature extraction and mining, and provide more accurate and comprehensive information for subsequent state identification. The system will further analyze and process the residual disturbance features, extract features that can reflect the characteristics of the geological monitoring task itself, such as change frequency, amplitude, trend, etc., and record and store these features as self-attention monitoring features.

[0193] After completing the self-attention feature mining, the microseismic event location and tracking system will further determine the monitoring correlation between geological monitoring tasks based on the microseismic trend deduction vectors corresponding to each geological monitoring task. The purpose of this step is to reveal the mutual influence and correlation between different geological monitoring tasks, and provide more comprehensive and in-depth information for subsequent state identification. The system will analyze and compare the microseismic trend deduction vectors between different geological monitoring tasks to find out the trend deduction correlation features and trend deduction mutually exclusive features between them. Trend deduction correlation features refer to the similar or consistent change trends presented by different geological monitoring tasks during the monitoring process, while trend deduction mutually exclusive features refer to the opposite or contradictory change trends presented between them.

[0194] After determining the monitoring correlation between geological monitoring tasks, the microseismic event location and tracking system will perform cross-attention feature mining based on this correlation to obtain cross-attention monitoring features between geological monitoring tasks. The purpose of this step is to reveal the complex relationship between geological monitoring tasks by extracting and mining the mutual influence and correlation features between them, and provide more comprehensive and accurate information for subsequent state identification. The system will conduct in-depth analysis and processing of the monitoring correlation between different geological monitoring tasks, extract features that can reflect the mutual influence and correlation between them, such as change synchronization, degree of mutual influence, etc., and record and store these features as cross-attention monitoring features.

[0195] Finally, the microseismic event location and tracking system will generate the state recognition results of the spatiotemporal geological monitoring log monitoring based on the self-attention monitoring features of each geological monitoring task and the cross-attention monitoring features between geological monitoring tasks. The purpose of this step is to comprehensively analyze and judge the various features extracted and mined previously, and obtain the monitoring state recognition results of the geological monitoring task in the spatiotemporal geological monitoring log. The system will conduct a comprehensive analysis of the self-attention monitoring features and the cross-attention monitoring features to determine whether the state of the geological monitoring task is normal, whether there is anomaly or risk, and output and record these judgment results as state recognition results.

[0196] Through the above steps, the microseismic event location tracking system can accurately identify the status of geological monitoring tasks in the spatiotemporal geological monitoring log, providing a strong geological safety guarantee for the excavation process of underground caverns of large hydropower stations. Through the analysis and processing of microseismic trend deduction vectors and residual disturbance characteristics, the system can more accurately identify abnormalities or disturbances in geological monitoring tasks, improving the accuracy and reliability of geological monitoring. Through self-attention feature mining and cross-attention feature mining, the system can more comprehensively reveal the important characteristics and change laws of geological monitoring tasks themselves, as well as the mutual influence and correlation between different geological monitoring tasks, enhancing the comprehensiveness and depth of geological monitoring. By accurately identifying the status of geological monitoring tasks in the spatiotemporal geological monitoring log, the system can timely discover and warn of potential geological risks, providing a stronger geological safety guarantee for the excavation process of underground caverns of large hydropower stations.

[0197] Based on the same or similar technical concept as above, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is run, the method for locating and tracking microseismic events is executed.

[0198] Based on the same or similar technical concepts mentioned above, an embodiment of the present invention further provides a computer program product, including a computer program or computer executable instructions, which, when executed by a processor, implements a microseismic event positioning and tracking method.

[0199] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0200] The above disclosure is only a preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the relevant contents of the present invention are still within the scope of the present invention.

Claims

1. A method for locating and tracking microseismic events, characterized in that: Applied to a microseismic event positioning and tracking system, the microseismic event positioning and tracking method comprises: Performing microseismic event location tracking on past microseismic waveform data in an AI algorithm debugging sample set by using a second-order differential grid wave velocity algorithm to obtain past microseismic event location tracking labels corresponding to each past microseismic waveform data, wherein the AI ​​algorithm debugging sample set includes the past microseismic waveform data and a priori microseismic event location tracking annotations corresponding to the past microseismic waveform data; Based on the past microseismic waveform state vector of the past microseismic waveform data and the past microseismic event positioning and tracking label, determining the debugging error update coefficient corresponding to the past microseismic waveform data, the past microseismic waveform state vector includes at least one of the microseismic event type and the prior annotation number of the prior microseismic event positioning and tracking annotation corresponding to the past microseismic waveform data, the microseismic event type includes a non-noise disturbance type and a noise disturbance type, and the prior annotation number represents the number of past microseismic waveform data with the prior microseismic event positioning and tracking annotation in the AI ​​algorithm debugging sample set; Determining the algorithm positioning and tracking error of the past microseismic waveform data according to the debugging error update coefficient, the past microseismic event positioning and tracking label, and the priori microseismic event positioning and tracking annotation; Optimizing the second-order difference grid wave velocity algorithm according to the positioning tracking error of the algorithm; The second-order difference grid wave velocity algorithm is obtained through double-cycle debugging; The method of optimizing the second-order difference grid wave velocity algorithm based on the positioning and tracking error of the algorithm includes: In a first debugging cycle, the second-order differential grid wave velocity algorithm is optimized according to a first algorithm positioning tracking error, wherein the first algorithm positioning tracking error is an algorithm positioning tracking error corresponding to past microseismic waveform data in a first AI algorithm debugging sample set, and the past microseismic waveform data in the first AI algorithm debugging sample set has non-correlated prior microseismic event positioning tracking annotations; In a second debugging cycle, based on the second-order differential grid wave velocity algorithm debugged in the first debugging cycle, the second-order differential grid wave velocity algorithm is optimized according to a second algorithm positioning and tracking error, wherein the second algorithm positioning and tracking error is an algorithm positioning and tracking error corresponding to past microseismic waveform data in a second AI algorithm debugging sample set, and the past microseismic waveform data in the second AI algorithm debugging sample set has at least one prior microseismic event positioning and tracking annotation; In the first debugging cycle, the determining of the debugging error update coefficient corresponding to the past microseismic waveform data based on the past microseismic waveform state vector of the past microseismic waveform data and the past microseismic event positioning tracking tag includes: Based on the microseismic event type of the past microseismic waveform data and the positioning and tracking label of the past microseismic event, determining a first debugging error update coefficient corresponding to the past microseismic waveform data, wherein the first debugging error update coefficient is used to adjust the algorithm positioning and tracking error of the noise disturbance type; Determine a second debugging error update coefficient corresponding to the past microseismic waveform data based on the number of prior annotations of the prior microseismic event positioning and tracking annotations and the number of first classification keywords in the first AI algorithm debugging sample set, the number of the first classification keywords represents the total number of prior annotation keywords included in the first AI algorithm debugging sample set, and the second debugging error update coefficient has a first quantitative relationship with the number of prior annotations; Determining the algorithm positioning and tracking error of the past microseismic waveform data according to the debugging error update coefficient, the past microseismic event positioning and tracking label, and the priori microseismic event positioning and tracking annotation includes: The first algorithm positioning and tracking error of the past microseismic waveform data is determined based on the first debugging error update coefficient, the second debugging error update coefficient, the past microseismic event positioning and tracking label, and the priori microseismic event positioning and tracking annotation.

2. The microseismic event location and tracking method according to claim 1, characterized in that: In the second debugging cycle, the determining of the debugging error update coefficient corresponding to the past microseismic waveform data based on the past microseismic waveform state vector of the past microseismic waveform data and the past microseismic event positioning tracking tag includes: Based on the microseismic event type of the past microseismic waveform data and the positioning and tracking label of the past microseismic event, determining a first debugging error update coefficient corresponding to the past microseismic waveform data, wherein the first debugging error update coefficient is used to adjust the algorithm positioning and tracking error of the noise disturbance type; Determine a second debugging error update coefficient corresponding to the past microseismic waveform data based on the number of past classification keywords of the past microseismic waveform data, the number of prior annotations of the prior microseismic event positioning and tracking annotations, and the number of second classification keywords in the second AI algorithm debugging sample set; Determining the algorithm positioning and tracking error of the past microseismic waveform data according to the debugging error update coefficient, the past microseismic event positioning and tracking label, and the priori microseismic event positioning and tracking annotation includes: A second algorithm positioning and tracking error of the past microseismic waveform data is determined based on the first debugging error update coefficient, the second debugging error update coefficient, the past microseismic event positioning and tracking label, and the priori microseismic event positioning and tracking annotation.

3. The microseismic event location and tracking method according to claim 2, characterized in that: The determining of the second debugging error update coefficient corresponding to the past microseismic waveform data based on the number of past classification keywords of the past microseismic waveform data, the number of prior annotations of the prior microseismic event positioning and tracking annotations, and the number of second classification keywords in the second AI algorithm debugging sample set includes: On the basis that the number of past classification keywords of the past microseismic waveform data is a set number, the second debugging error update coefficient corresponding to the past microseismic waveform data is determined according to the number of prior annotations of the prior microseismic event positioning and tracking annotations and the number of the second classification keywords in the second AI algorithm debugging sample set; On the basis that the number of past classification keywords of the past microseismic waveform data is greater than the set number, the prior learning confidence of each prior microseismic event positioning and tracking annotation is determined according to the prior annotation number of each prior microseismic event positioning and tracking annotation, and the prior microseismic event positioning and tracking annotation with the least number of prior annotations is determined as the target prior microseismic event positioning and tracking annotation; based on the second number of classification keywords and the confidence comparison result between the prior learning confidence of the target prior microseismic event positioning and tracking annotation and the prior learning confidence of other prior microseismic event positioning and tracking annotations, the second debugging error update coefficient corresponding to the past microseismic waveform data is determined.

4. The microseismic event location and tracking method according to claim 1, characterized in that: The microseismic event location tracking is performed on the past microseismic waveform data in the AI ​​algorithm debugging sample set by using the second-order difference grid wave velocity algorithm to obtain the past microseismic event location tracking label corresponding to each past microseismic waveform data, including: The past microseismic waveform data in the AI ​​algorithm debugging sample set are transmitted to the event element mining branch in the second-order difference grid wave velocity algorithm to obtain the past event element information corresponding to the past microseismic waveform data generated by the event element mining branch; The past event element information is transmitted to the event location tracking branch in the second-order difference grid wave velocity algorithm to obtain the past microseismic event location tracking label corresponding to each past microseismic waveform data generated by the event location tracking branch.

5. The microseismic event location and tracking method according to claim 1, characterized in that: The microseismic event positioning and tracking method further comprises: Calling the debugged second-order difference grid wave velocity algorithm to perform microseismic event positioning and tracking on the target microseismic waveform monitoring data of each geological monitoring task in the spatiotemporal geological monitoring log, and determining the target positioning and tracking features corresponding to the target microseismic waveform monitoring data of each geological monitoring task, wherein the spatiotemporal geological monitoring log includes at least one spatiotemporal geological monitoring data set and target microseismic waveform monitoring data of at least two geological monitoring tasks; The monitoring state of the spatiotemporal geological monitoring log is identified according to the target positioning tracking features corresponding to the target microseismic waveform monitoring data of each geological monitoring task, and the state identification result of the spatiotemporal geological monitoring log monitoring is obtained.

6. The microseismic event location and tracking method according to claim 5, characterized in that: The monitoring state identification of the spatiotemporal geological monitoring log is performed based on the target positioning tracking features corresponding to the target microseismic waveform monitoring data of each geological monitoring task, and the state identification result of the spatiotemporal geological monitoring log monitoring is obtained, including: According to the target positioning and tracking features corresponding to the target microseismic waveform monitoring data of each geological monitoring task in the same spatiotemporal geological monitoring data set, and the microseismic event discrimination weights corresponding to the target positioning and tracking features, a microseismic trend deduction vector of each geological monitoring task in the spatiotemporal geological monitoring data set is generated, wherein the microseismic event discrimination weights are event discrimination viewpoints obtained by performing microseismic event positioning and tracking on the target microseismic waveform monitoring data through the second-order difference grid wave velocity algorithm; Identify the monitoring status of the spatiotemporal geological monitoring log according to the microseismic trend deduction vector corresponding to each geological monitoring task, and obtain the monitoring status identification result of the spatiotemporal geological monitoring log; The method of generating a microseismic trend deduction vector of each geological monitoring task in the spatiotemporal geological monitoring data set based on the target positioning and tracking features corresponding to the target microseismic waveform monitoring data of each geological monitoring task in the spatiotemporal geological monitoring data set and the microseismic event discrimination weight corresponding to the target positioning and tracking features comprises: On the basis that the target microseismic waveform monitoring data has a non-linked target positioning and tracking feature, a risk prediction coefficient of the target microseismic waveform monitoring data is determined according to a microseismic event discrimination weight corresponding to the target positioning and tracking feature; On the basis that the target microseismic waveform monitoring data has at least two target positioning and tracking features, the disaster impact thermal variables of each target positioning and tracking feature are determined according to the microseismic event discrimination weights corresponding to each target positioning and tracking feature; the risk prediction coefficient of the target microseismic waveform monitoring data is determined according to the microseismic event discrimination weights corresponding to the target positioning and tracking feature with the highest disaster impact thermal variable; Generate the microseismic trend deduction vector of the geological monitoring task in the spatiotemporal geological monitoring data set according to the priority characteristics of each target microseismic waveform monitoring data of the geological monitoring task in the spatiotemporal geological monitoring data set and the risk prediction coefficient; Determining the disaster impact thermal variables of each target positioning and tracking feature according to the microseismic event discrimination weight corresponding to each target positioning and tracking feature includes: Perform feature space mapping on the microseismic event discrimination weights corresponding to each target positioning and tracking feature to obtain the disaster-affecting thermal variables of each target positioning and tracking feature, and the disaster-affecting thermal variables have a second quantitative relationship with the microseismic event discrimination weights; or, determine the grid node association weights corresponding to each target positioning and tracking feature based on the microseismic event discrimination weights and weight threshold values ​​corresponding to each target positioning and tracking feature; determine the disaster-affecting thermal variables of each target positioning and tracking feature based on the grid node association weights corresponding to each target positioning and tracking feature, and the disaster-affecting thermal variables have a second quantitative relationship with the grid node association weights.

7. A microseismic event location and tracking system, characterized in that: include: Processor, memory, and network interface; The processor is connected to the memory and the network interface; the network interface is used to provide data communication function, the memory is used to store program code, and the processor is used to call the program code to execute the microseismic event positioning and tracking method described in any one of claims 1-6.

8. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is run, the method for locating and tracking microseismic events according to any one of claims 1 to 6 is executed.

Images