Coal mine earthquake prediction method and system based on Bayesian reasoning

Abstract

The invention provides a coal mine earthquake prediction method and system based on Bayesian reasoning, and the method comprises the steps: obtaining observed micro-earthquake event data in a coal mine region; a microseism prediction model is constructed, an optimal solution of model parameters of the microseism prediction model is obtained based on the microseism event data in the historical time period, and the model parameters comprise event triggering parameters and event occurrence parameters; the optimal solution of the model parameters is used as prior information, and posteriori distribution is obtained based on Bayesian reasoning for observed micro-seismic event data; and extracting posterior probabilities in a set probability range from the posterior distribution, obtaining the maximum magnitude corresponding to the prediction time period based on the model parameters corresponding to each posterior probability, and calculating the occurrence probability of the maximum magnitude event based on the maximum magnitudes corresponding to all the posterior probabilities in the set probability range and a given magnitude threshold.

Description

Technical Field

[0001] The present invention relates to the field of geophysics and mining engineering technology, and in particular to a method and system for predicting coal mine earthquakes based on Bayesian reasoning. Background Art

[0002] During underground coal mining or fracturing operations, the redistribution of stratum stress can induce various seismic events of varying sizes, namely mine tremors or microseismic events. For mining areas, how to timely and accurately predict and evaluate the probability of these seismic events, and then plan safe and efficient mining processes, is an important research direction for today's safe mine operations.

[0003] Existing mine earthquake prediction technologies are usually based on statistical or physical models, such as the Modified Omori Law (MOL). Although MOL is simple and efficient in describing the attenuation law of aftershocks, it is difficult to handle sequences with significant triggering effects or complex induced characteristics. Summary of the invention

[0004] The present invention aims to solve one of the technical problems in the related art at least to a certain extent.

[0005] To this end, the first purpose of the present invention is to propose a coal mine earthquake prediction method based on Bayesian reasoning to improve the accuracy of coal mine earthquake prediction.

[0006] The second object of the present invention is to propose a coal mine earthquake prediction system based on Bayesian reasoning.

[0007] A third objective of the present invention is to provide an electronic device.

[0008] A fourth object of the present invention is to provide a computer-readable storage medium.

[0009] To achieve the above-mentioned object, the first aspect of the present invention proposes a method for predicting coal mine earthquakes based on Bayesian reasoning, comprising:

[0010] Obtain data on microseismic events observed in the coal mine area;

[0011] Constructing a microseismic prediction model, and obtaining an optimal solution of model parameters of the microseismic prediction model based on the microseismic event data in a historical period, wherein the model parameters include event triggering parameters and event occurrence parameters;

[0012] Taking the optimal solution of the model parameters as prior information, the posterior distribution is obtained based on Bayesian reasoning for the observed microseismic event data;

[0013] The posterior probability within the set probability range is extracted from the posterior distribution, and the maximum magnitude corresponding to the prediction period is obtained based on the model parameters corresponding to each posterior probability. The probability of occurrence of the maximum magnitude event is calculated based on the maximum magnitude corresponding to all posterior probabilities within the set probability range and a given magnitude threshold.

[0014] In the method of the first aspect of the present invention, the optimal solution of the model parameters of the microseismic prediction model is obtained based on the microseismic event data in the historical period, including: obtaining the optimal solution of the model parameters of the microseismic prediction model based on the microseismic event data in the historical period through the maximum likelihood method or the Bayesian inference method.

[0015] In the method of the first aspect of the present invention, the microseismic prediction model satisfies:

[0016]

[0017] In the formula, λ ω (t|H t ) is the conditional strength, ω={μ,A,c,p,α} is the model parameter, H t is the historical record of all events that have occurred before time t in the microseismic event data, μ is the background event occurrence rate, A is the event trigger rate coefficient, t i is the time when the ith event occurs, α is the magnitude trigger intensity coefficient, m i is the magnitude of the ith event, m 0 is the cutoff magnitude for analysis, c is the time shift constant, and p is the time decay exponent. The event triggering parameters include the event triggering rate coefficient, the magnitude triggering intensity coefficient, the time shift constant, and the time decay exponent. The event occurrence parameters include the background event occurrence rate.

[0018] In the method of the first aspect of the present invention, the posterior distribution is obtained by Markov Chain Monte Carlo sampling based on Bayesian inference.

[0019] The method of the first aspect of the present invention also includes: obtaining the earthquake frequency corresponding to the prediction period based on the model parameters corresponding to each posterior probability.

[0020] In the method of the first aspect of the present invention, after the microseismic event data in the coal mine area is acquired, the microseismic event data needs to be preprocessed.

[0021] To achieve the above-mentioned purpose, the second aspect of the present invention proposes a coal mine earthquake prediction system based on Bayesian reasoning, comprising:

[0022] An acquisition module is used to acquire the microseismic event data observed in the coal mine area;

[0023] A parameter estimation module, used to construct a microseismic prediction model, and obtain an optimal solution of model parameters of the microseismic prediction model based on the microseismic event data in a historical period, wherein the model parameters include event triggering parameters and event occurrence parameters;

[0024] The prediction module is used to use the optimal solution of the model parameters as prior information, and obtain the posterior distribution based on Bayesian reasoning for the observed microseismic event data; extract the posterior probability within a set probability range from the posterior distribution, obtain the maximum magnitude corresponding to the prediction period based on the model parameters corresponding to each posterior probability, and calculate the probability of occurrence of the maximum magnitude event based on the maximum magnitude corresponding to all posterior probabilities within the set probability range and a given magnitude threshold.

[0025] In the system of the second aspect of the present invention, in the parameter estimation module, the microseismic prediction model satisfies:

[0026]

[0027] In the formula, λ ω (t|H t ) is the conditional strength, ω={μ,A,c,p,α} is the model parameter, H t is the historical record of all events that have occurred before time t in the microseismic event data, μ is the background event occurrence rate, A is the event trigger rate coefficient, t i is the time when the ith event occurs, α is the magnitude trigger intensity coefficient, m i is the magnitude of the ith event, m 0 is the cutoff magnitude for analysis, c is the time shift constant, and p is the time decay exponent. The event triggering parameters include the event triggering rate coefficient, the magnitude triggering intensity coefficient, the time shift constant, and the time decay exponent. The event occurrence parameters include the background event occurrence rate.

[0028] To achieve the above-mentioned purpose, the third aspect of the present invention proposes an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method proposed in the first aspect of the present invention.

[0029] To achieve the above-mentioned purpose, the fourth aspect of the present invention proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the method proposed in the first aspect of the present invention.

[0030] The method, system, electronic device and storage medium for predicting coal mine earthquakes based on Bayesian reasoning provided by the present invention obtain the observed microseismic event data in the coal mine area; construct a microseismic prediction model, obtain the optimal solution of the model parameters of the microseismic prediction model based on the microseismic event data in the historical period, and the model parameters include event triggering parameters and event occurrence parameters; use the optimal solution of the model parameters as prior information, and obtain the posterior distribution based on Bayesian reasoning for the observed microseismic event data; extract the posterior probability within the set probability range from the posterior distribution, obtain the maximum magnitude corresponding to the prediction period based on the model parameters corresponding to each posterior probability, and calculate the maximum magnitude event occurrence probability based on the maximum magnitude corresponding to all posterior probabilities within the set probability range and the given magnitude threshold. In this case, considering the triggering effect and occurrence of microseisms, a microseismic prediction model is constructed, wherein the model parameters include event triggering parameters and event occurrence parameters, and the optimal solution of the model parameters of the obtained microseismic prediction model is used as prior information, combined with Bayesian reasoning to obtain the posterior distribution, and then the maximum magnitude corresponding to multiple posterior probabilities within the set probability range is obtained, thereby calculating the maximum magnitude event occurrence probability. Compared with the prior art, the present invention can more accurately predict earthquakes in the prediction period, thereby improving the accuracy of coal mine earthquake prediction.

[0031] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0033] Figure 1 A schematic diagram of a flow chart of a method for predicting coal mine earthquakes based on Bayesian reasoning provided by an embodiment of the present invention;

[0034] Figure 2 A block diagram of a coal mine earthquake prediction system based on Bayesian reasoning provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0035] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0036] The following describes a method and system for predicting coal mine earthquakes based on Bayesian reasoning according to an embodiment of the present invention with reference to the accompanying drawings.

[0037] The embodiment of the present invention provides a method for predicting coal mine earthquakes based on Bayesian reasoning to improve the accuracy of coal mine earthquake prediction. Mine earthquakes are also called microearthquakes. Mine earthquakes refer to earthquake events induced during coal mining.

[0038] Figure 1 A schematic flow chart of a method for predicting coal mine earthquakes based on Bayesian reasoning provided in an embodiment of the present invention.

[0039] like Figure 1 As shown, the coal mine earthquake prediction method based on Bayesian reasoning includes the following steps:

[0040] Step S101, obtaining microseismic event data observed in the coal mine area.

[0041] In step S101, the observed microseismic event data includes historical microseismic event data and microseismic event data collected in real time.

[0042] In step S101, the observed microseismic event data can be continuously collected by a collection device, which can be, for example, a node-type three-component detector.

[0043] In step S101, the microseismic event data includes waveform data such as the velocity of transverse and longitudinal waves, displacement, frequency, magnitude, analyzed cutoff magnitude, time of event occurrence, etc.

[0044] In step S101, the microseismic event data also includes event trigger rate coefficient, magnitude trigger intensity coefficient, time translation constant, time decay index, background event occurrence rate and other event-related data.

[0045] In step S101, the microseismic event data may be divided into a plurality of event records, each of which includes corresponding waveform data, frequency, magnitude, analyzed cutoff magnitude, time of event occurrence, and event-related data.

[0046] In step S101, in order to improve the accuracy of subsequent predictions, after obtaining the microseismic event data in the coal mine area, it is necessary to pre-process the microseismic event data.

[0047] Specifically, the collected raw data (i.e., the microseismic event data output by the acquisition device) is first subjected to bandpass filtering, and then a seismic event database is established. The seismic event database includes multiple event records. The duration of each event record is a time step. Finally, all event records in the seismic event database are preliminarily screened to remove irrelevant event records without obvious take-off signals, such as event records with obvious manual acquisition errors or irrelevant to underground blasting and ground testing; wherein, no obvious take-off signal means that the signal of the event in the record is below the integrity magnitude threshold. Among them, the integrity magnitude threshold refers to the minimum magnitude value to ensure the reliability of the seismic event record. The integrity magnitude threshold can be obtained based on previous experience or event amplitude-completeness analysis. Thus, the seismic event records in the seismic event database after the preliminary screening are reliable and complete.

[0048] Step S102, constructing a microseismic prediction model, and obtaining an optimal solution of model parameters of the microseismic prediction model based on microseismic event data in a historical period, wherein the model parameters include event triggering parameters and event occurrence parameters.

[0049] In step S102, the model parameters include event triggering parameters and event occurrence parameters. The constructed microseismic prediction model (MPM) can describe the triggering and occurrence rate of microseismic or mining earthquake events.

[0050] In step S102, a microseismic prediction model is constructed based on a conditional intensity function, where the conditional intensity function is a mathematical function that describes the event occurrence rate. The constructed microseismic prediction model satisfies:

[0051]

[0052] In the formula, λ ω (t|H t ) is the conditional strength. ω={μ,A,c,p,α} is the model parameter, and the model parameter ω is the parameter set to be estimated. t is the historical record of all events that have occurred before time t in the microseismic event data. μ is the background event rate (i.e., the unconditional independent event rate). A is the event trigger rate coefficient, which characterizes the intrinsic ability of the main shock to trigger aftershocks. i is the time when the ith event occurs. α is the magnitude trigger intensity coefficient, which is used to describe the impact of the event magnitude on the potential for triggering subsequent seismic activity. i is the magnitude of the ith event, m 0is the cutoff magnitude for analysis. c is the time shift constant, which characterizes the triggering characteristics in a short period of time after the main shock. p is the time decay exponent, which determines the rate at which the aftershock trigger decays over time. The event triggering parameters include the event triggering rate coefficient, the magnitude triggering intensity coefficient, the time shift constant, and the time decay exponent. The event occurrence parameters include the background event occurrence rate.

[0053] In step S102, the optimal solution of the model parameters of the microseismic prediction model is obtained based on the microseismic event data in the historical period, including: obtaining the optimal solution of the model parameters of the microseismic prediction model based on the microseismic event data in the historical period by the maximum likelihood method or the Bayesian inference method.

[0054] Specifically, microseismic event data within a historical period of the microseismic event data observed in step S101 are selected, and all event records included in the microseismic event data within the historical period are used to obtain the optimal solution for the model parameters of the microseismic prediction model through the maximum likelihood method or the Bayesian inference method.

[0055] Taking the Maximum Likelihood (ML) method as an example, the log-likelihood function can be maximized to estimate the model parameters. Based on formula (1), the log-likelihood function of formula (2) can be obtained, and formula (2) is as follows:

[0056]

[0057] Where lnL(ω) is the log-likelihood value. The value set of time t is {t 1 ,t 2 ,…,t N}, any element in the value set can be represented as t j . N is the number of elements in the value set at time t. [T 0 ,T e ] is the historical period. [T s ,T e ] is the target fitting time interval. All event records in the historical period are used in combination with formula (2) to obtain multiple log-likelihood values. A numerical optimization algorithm (such as Newton method or quasi-Newton method) is used to search among all log-likelihood values ​​to obtain the model parameters corresponding to the maximum log-likelihood value. The model parameters corresponding to the maximum log-likelihood value are the optimal solution ω* of the model parameters. Where T s Can be selected based on needs.

[0058] Step S103, taking the optimal solution of the model parameters as prior information, and obtaining the posterior distribution based on Bayesian reasoning for the observed microseismic event data.

[0059] In step S103, for the observed microseismic event data, under Bayesian reasoning, the Gamma prior distribution is first set, and then the posterior distribution p(ω|S) is obtained by Markov Chain Monte Carlo (MCMC) sampling. Where S represents the observed microseismic data in step 101. The prior information is used as the initial value of the model parameter in the initialization step of MCMC sampling. This can accelerate the convergence of the algorithm during the sampling process.

[0060] Step S104, extracting the posterior probability within the set probability range from the posterior distribution, obtaining the maximum magnitude corresponding to the prediction period based on the model parameters corresponding to each posterior probability, and calculating the probability of occurrence of the maximum magnitude event based on the maximum magnitude corresponding to all posterior probabilities within the set probability range and a given magnitude threshold.

[0061] In step S104, the probability range is set based on the maximum posterior probability in the posterior distribution curve. For example, the corresponding range can be selected on both sides of the maximum posterior probability in the posterior distribution curve to obtain the set probability range. By extracting the posterior probability within the set probability range from the posterior distribution, a plurality of large sample posterior probability values ​​(i.e., large sample parameter values) can be obtained.

[0062] In step S104, it also includes: obtaining the earthquake frequency corresponding to the prediction period based on the model parameters corresponding to each posterior probability.

[0063] Specifically, the process of obtaining the probability of occurrence of the maximum magnitude event in this step includes:

[0064] 1) Extract a large sample of parameter values ​​from the posterior distribution p(ω|S);

[0065] 2) For each large sample parameter value, based on the model parameter ω corresponding to the large sample parameter value, the prediction time interval [T e ,T e +ΔT] and record the earthquake frequency (also called earthquake frequency, frequency, number of earthquake events), maximum magnitude m ex ;

[0066] 3) Count the proportion of all simulations that exceed a given magnitude threshold to obtain the maximum event exceedance probability (i.e., the probability of the maximum magnitude event occurring). ex The probability of >m is calculated by formula (3):

[0067]

[0068] Among them, Pr B (m ex>m|S,ΔT) is the maximum event exceedance probability, m is the given magnitude threshold, ΔT is the duration of the prediction time interval, θ represents the relationship parameter between frequency-magnitude distribution, and Θ and Ω are both set parameter spaces. EV (m ex >m|θ,ω,ΔT) is calculated by extreme value distribution (EVD). Thus, the posterior distribution is combined with extreme value probability analysis to evaluate the maximum magnitude and the probability of occurrence of the maximum magnitude event within the prediction interval.

[0069] In order to verify the effect of the method of the present invention, prediction capability verification was performed.

[0070] Specifically, after obtaining the probability of occurrence of the maximum magnitude event, the accuracy and consistency of the prediction in terms of number and magnitude distribution can be evaluated by comparing it with the microseismic data actually observed in the prediction period (i.e., the prediction time interval) through statistical methods. Among them, the N test (event number test) and the M test (magnitude distribution test) are two commonly used statistical methods. In the present invention, the N test is used to evaluate the prediction accuracy in terms of the number of microseismic events, and the M test is used to evaluate the prediction consistency in terms of magnitude distribution.

[0071] For the N test: The N test is used to evaluate the accuracy of the predicted number of earthquake events (i.e. the frequency of earthquakes recorded after simulated seismic activity) and the number of earthquake events actually observed during the predicted period.

[0072] Assume that the predicted number of earthquake events expected to occur during the prediction period is λ, and the actual number of earthquake events actually observed during the prediction period is k. If the events are independent and follow a Poisson distribution, their probability mass function (PMF) is as follows:

[0073]

[0074] Where P(k;λ) is the probability. If P(k;λ) is less than the preset significance level (such as 0.05), the predicted number of earthquake events is rejected, and it is considered that the prediction is significantly different from the observation.

[0075] For the M test: The M test is used to compare the consistency between the predicted earthquake magnitude distribution and the actually observed magnitude distribution. It is usually used to evaluate whether the predicted magnitude distribution conforms to the statistical characteristics of the actual observed data, such as the Gutenberg-Richter relationship. This is an empirical relationship that describes the distribution of earthquake magnitudes, indicating that within a certain time and region, the relationship between the frequency X of earthquakes and the magnitude M is as shown in formula (5):

[0076] log 10X=a-bM(5)

[0077] Where a and b are empirical parameters, the b value is usually close to 1, and X is the frequency greater than or equal to the magnitude M.

[0078] The Gutenberg-Richter relationship of formula (5) is used to fit the predicted magnitude distribution and determine the parameter a p and b p Similarly, the Gutenberg-Richter relationship is used to fit the observed magnitude distribution and determine the parameter a o and b o ; Use statistical tests (such as t-tests or nonparametric tests) to compare a p with a o 、b p With b o Is there a significant difference? If the difference between the two groups is within the set difference range, it is considered that there is no significant difference between the prediction and the observation;

[0079] The difference between the predicted magnitude cumulative distribution function (CDF) and the observed magnitude CDF is calculated using the KS test (Kolmogorov-Smirnov test), and the maximum difference between the two cumulative distribution functions is calculated, as shown in formula (6):

[0080]

[0081] In the formula, F 预测 (M) is the CDF of the predicted magnitude, F 观测 (M) is the observed magnitude CDF. D is the difference.

[0082] If the difference D is greater than the critical value, or P(k;λ) is less than the significance level, the magnitude distribution is considered to be significantly different.

[0083] By using the above N test and M test, the predicted number of earthquake events, the predicted magnitude and the predicted magnitude distribution are obtained by the method of the present invention. It is found that the difference D is less than or equal to the critical value, a p with a o 、b p With b o The differences between the two groups are within the set difference range, and P(k;λ) is greater than or equal to the significance level. Therefore, the prediction method of the present invention has better accuracy and consistency in number and magnitude distribution.

[0084] In order to implement the above embodiment, the present invention also proposes a coal mine earthquake prediction system based on Bayesian reasoning.

[0085] Figure 2A block diagram of a coal mine earthquake prediction system based on Bayesian reasoning provided in an embodiment of the present invention.

[0086] like Figure 2 As shown, the coal mine earthquake prediction system based on Bayesian reasoning includes an acquisition module 11, a parameter estimation module 12 and a prediction module 13, wherein:

[0087] An acquisition module 11 is used to acquire microseismic event data observed in the coal mine area;

[0088] The parameter estimation module 12 is used to construct a microseismic prediction model and obtain the optimal solution of the model parameters of the microseismic prediction model based on the microseismic event data in the historical period. The model parameters include event triggering parameters and event occurrence parameters.

[0089] Prediction module 13 is used to use the optimal solution of the model parameters as prior information, and obtain the posterior distribution based on Bayesian reasoning for the observed microseismic event data; extract the posterior probability within the set probability range from the posterior distribution, obtain the maximum magnitude corresponding to the prediction period based on the model parameters corresponding to each posterior probability, and calculate the probability of the maximum magnitude event based on the maximum magnitude corresponding to all posterior probabilities within the set probability range and a given magnitude threshold.

[0090] Furthermore, in a possible implementation manner of the embodiment of the present invention, a preprocessing module is also included, and the preprocessing module is used to preprocess the microseismic event data output by the acquisition module 11 and then send it to the parameter estimation module 12.

[0091] Furthermore, in a possible implementation of the embodiment of the present invention, in the parameter estimation module 12, the microseismic prediction model satisfies the formula (1) in the method embodiment:

[0092]

[0093] In the formula, λ ω (t|H t ) is the conditional strength, ω={μ,A,c,p,α} is the model parameter, H t is the historical record of all events that have occurred before time t in the microseismic event data, μ is the background event occurrence rate, A is the event trigger rate coefficient, t i is the time when the ith event occurs, α is the magnitude trigger intensity coefficient, m i is the magnitude of the ith event, m 0 is the cutoff magnitude for analysis, c is the time shift constant, and p is the time decay exponent. The event triggering parameters include the event triggering rate coefficient, the magnitude triggering intensity coefficient, the time shift constant, and the time decay exponent. The event occurrence parameters include the background event occurrence rate.

[0094] Furthermore, in a possible implementation of an embodiment of the present invention, in the parameter estimation module 12, the optimal solution of the model parameters of the microseismic prediction model is obtained based on the microseismic event data within the historical period, including: obtaining the optimal solution of the model parameters of the microseismic prediction model based on the microseismic event data within the historical period through the maximum likelihood method or the Bayesian inference method.

[0095] Furthermore, in a possible implementation of the embodiment of the present invention, in the prediction module 13, based on Bayesian reasoning, the posterior distribution is obtained through Markov chain Monte Carlo sampling.

[0096] Furthermore, in a possible implementation of the embodiment of the present invention, the prediction module 13 further includes: obtaining the earthquake frequency corresponding to the prediction period based on the model parameters corresponding to each posterior probability.

[0097] It should be noted that the aforementioned explanation of the embodiment of the coal mine earthquake prediction method based on Bayesian reasoning is also applicable to the coal mine earthquake prediction system based on Bayesian reasoning in this embodiment, and will not be repeated here.

[0098] In an embodiment of the present invention, by obtaining the observed microseismic event data in the coal mine area; constructing a microseismic prediction model, obtaining the optimal solution of the model parameters of the microseismic prediction model based on the microseismic event data in the historical period, the model parameters include event triggering parameters and event occurrence parameters; using the optimal solution of the model parameters as prior information, for the observed microseismic event data, obtaining the posterior distribution based on Bayesian reasoning; extracting the posterior probability within the set probability range from the posterior distribution, obtaining the maximum magnitude corresponding to the prediction period based on the model parameters corresponding to each posterior probability, and calculating the maximum magnitude event occurrence probability based on the maximum magnitude corresponding to all posterior probabilities within the set probability range and the given magnitude threshold. In this case, considering the triggering effect and occurrence of microseisms, a microseismic prediction model is constructed, wherein the model parameters include event triggering parameters and event occurrence parameters, using the optimal solution of the model parameters of the obtained microseismic prediction model as prior information, combining Bayesian reasoning to obtain the posterior distribution, and then obtaining the maximum magnitude corresponding to multiple posterior probabilities within the set probability range, thereby calculating the maximum magnitude event occurrence probability. Compared with the prior art, the present invention can more accurately predict earthquakes in the prediction period, thereby improving the accuracy of coal mine earthquake prediction.

[0099] In the method and system of the present invention, the proposed microseismic prediction model (MPM) can describe the triggering and occurrence rate of microseismic or mine earthquake events, and introduce the Bayesian prediction framework at the prediction level. Among them, the MPM model parameters are estimated in the historical period, and the probability of occurrence of the maximum magnitude event and the potential maximum magnitude are evaluated by using Bayesian prediction within the prediction time interval, which can fully consider the triggering effect and parameter uncertainty of microseismic events, accurately analyze the spatiotemporal characteristics of coal mining-induced earthquakes, and make short-term or medium-term microseismic predictions. Through this method, scientific safety assessment and real-time risk warning can be provided for coal mine operations, effectively reducing the risk of miners and equipment being harmed by sudden earthquakes. Compared with only using the modified Omori law or the simple Poisson hypothesis, the present invention depicts complex triggering characteristics through the MPM model, and incorporates parameter uncertainty through the Bayesian framework, and the prediction accuracy and robustness are improved. On the basis of the MOL prediction model, the present invention proposes a coal mine earthquake prediction method based on Bayesian reasoning, which can calculate the maximum expected probability of an event exceeding a certain magnitude threshold within a given prediction time range. It can more accurately and comprehensively predict micro-earthquake / mine earthquake events that may occur during coal mining.

[0100] In order to implement the above embodiments, the present invention also proposes an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method provided by the above embodiments.

[0101] In order to implement the above embodiments, the present invention further proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the methods provided by the above embodiments.

[0102] In order to implement the above embodiments, the present invention further provides a computer program product, including a computer program, which implements the methods provided by the above embodiments when executed by a processor.

[0103] In the description of the aforementioned embodiments, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0104] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0105] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present invention belong.

[0106] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0107] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or a combination thereof: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0108] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0109] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0110] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present invention. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A method for predicting coal mine earthquakes based on Bayesian reasoning, characterized in that: include: Obtain data on microseismic events observed in the coal mine area; Constructing a microseismic prediction model, and obtaining an optimal solution of model parameters of the microseismic prediction model based on the microseismic event data in a historical period, wherein the model parameters include event triggering parameters and event occurrence parameters; Taking the optimal solution of the model parameters as prior information, the posterior distribution is obtained based on Bayesian reasoning for the observed microseismic event data; The posterior probability within the set probability range is extracted from the posterior distribution, and the maximum magnitude corresponding to the prediction period is obtained based on the model parameters corresponding to each posterior probability. The probability of occurrence of the maximum magnitude event is calculated based on the maximum magnitude corresponding to all posterior probabilities within the set probability range and a given magnitude threshold.

2. The method for predicting coal mine earthquakes based on Bayesian reasoning according to claim 1, characterized in that: Obtaining an optimal solution for the model parameters of the microseismic prediction model based on the microseismic event data in a historical period includes: The optimal solution of the model parameters of the microseismic prediction model is obtained based on the microseismic event data in the historical period through the maximum likelihood method or the Bayesian reasoning method.

3. The method for predicting coal mine earthquakes based on Bayesian reasoning according to claim 1, characterized in that: The microseismic prediction model satisfies: In the formula, λ ω (t|H t ) is the conditional strength, ω={μ,A,c,p,α} is the model parameter, H t is the historical record of all events that have occurred before time t in the microseismic event data, μ is the background event occurrence rate, A is the event trigger rate coefficient, t i is the time when the ith event occurs, α is the magnitude trigger intensity coefficient, m i is the magnitude of the ith event, m0 is the cutoff magnitude of the analysis, c is the time shift constant, and p is the time decay exponent. The event triggering parameters include the event triggering rate coefficient, the magnitude triggering intensity coefficient, the time shift constant, and the time decay exponent. The event occurrence parameters include the background event occurrence rate.

4. The method for predicting coal mine earthquakes based on Bayesian reasoning according to claim 1, characterized in that: Based on Bayesian inference, the posterior distribution is obtained through Markov Chain Monte Carlo sampling.

5. The method for predicting coal mine earthquakes based on Bayesian reasoning according to claim 1, characterized in that: Also includes: Based on the model parameters corresponding to each posterior probability, the earthquake frequency corresponding to the prediction period is also obtained.

6. The method for predicting coal mine earthquakes based on Bayesian reasoning according to claim 1, characterized in that: After obtaining the microseismic event data in the coal mine area, the microseismic event data also needs to be preprocessed.

7. A coal mine earthquake prediction system based on Bayesian reasoning, characterized in that: include: An acquisition module is used to acquire the microseismic event data observed in the coal mine area; A parameter estimation module, used to construct a microseismic prediction model, and obtain an optimal solution of model parameters of the microseismic prediction model based on the microseismic event data in a historical period, wherein the model parameters include event triggering parameters and event occurrence parameters; A prediction module, for using the optimal solution of the model parameters as prior information, and obtaining a posterior distribution based on Bayesian reasoning for observed microseismic event data; The posterior probability within the set probability range is extracted from the posterior distribution, and the maximum magnitude corresponding to the prediction period is obtained based on the model parameters corresponding to each posterior probability. The probability of occurrence of the maximum magnitude event is calculated based on the maximum magnitude corresponding to all posterior probabilities within the set probability range and a given magnitude threshold.

8. The coal mine earthquake prediction system based on Bayesian reasoning according to claim 7 is characterized in that: In the parameter estimation module, the microseismic prediction model satisfies: In the formula, λ ω (t|H t ) is the conditional strength, ω={μ,A,c,p,α} is the model parameter, H t is the historical record of all events that have occurred before time t in the microseismic event data, μ is the background event occurrence rate, A is the event trigger rate coefficient, t i is the time when the ith event occurs, α is the magnitude trigger intensity coefficient, m i is the magnitude of the ith event, m0 is the cutoff magnitude of the analysis, c is the time shift constant, and p is the time decay exponent. The event triggering parameters include the event triggering rate coefficient, the magnitude triggering intensity coefficient, the time shift constant, and the time decay exponent. The event occurrence parameters include the background event occurrence rate.

9. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 6 when executed by a processor.

Images