An intelligent early warning method and system for underground engineering dynamic disasters

Abstract

The present invention provides an intelligent early warning method and system for underground engineering power disasters, which relates to the technical field of underground engineering power disaster early warning. The method includes: obtaining microseismic historical monitoring data of underground engineering; preprocessing the microseismic historical monitoring data, obtaining relevant time series parameters, and obtaining a reconstructed signal; obtaining multiple rolling statistical features of the reconstructed signal; selecting a preset number of rolling statistical features according to importance to form a rolling statistical feature matrix; constructing an underground engineering power disaster prediction model; inputting the rolling statistical feature matrix data set into the underground engineering power disaster prediction model for training; obtaining real-time microseismic monitoring data; extracting the real-time rolling statistical feature matrix data set; inputting the real-time rolling statistical feature matrix data set into the trained underground engineering power disaster prediction model, and outputting the disaster warning result. The present invention realizes efficient, accurate, and intelligent early warning of underground engineering power disasters.

Description

Technical Field

[0001] The present invention relates to the technical field of early warning of underground engineering power disasters, and in particular to an intelligent early warning method and system for underground engineering power disasters. Background Art

[0002] The intelligent early warning method for dynamic disasters in underground projects is a method that uses modern sensing technology, data analysis and machine learning algorithms to monitor various potential dynamic disasters in underground projects (such as earthquakes, gas explosions, rock bursts, etc.) in real time, and predict the possibility of disasters through intelligent models. When the early warning model detects potential risks, the system will automatically issue an early warning and inform relevant personnel in advance to take emergency measures, thereby reducing the threat of disasters to the safety of projects and personnel.

[0003] As underground space becomes deeper, projects become larger, and the surrounding stress environment becomes more complex, the frequency and severity of dynamic disasters are also on the rise. Accurate monitoring and reliable early warning of dynamic disasters in underground projects are the key to solving the above problems.

[0004] However, due to the complex underground engineering environment, the microseismic monitoring data obtained by the existing intelligent early warning methods for underground engineering dynamic disasters may contain noise and inaccuracies, and the disaster warning effect is not ideal. When selecting features, it is impossible to well screen out features that are highly relevant to disaster prediction, resulting in some irrelevant features affecting the performance of the model and reducing the accuracy of the prediction. Summary of the invention

[0005] In order to solve the technical problems that due to the complex underground engineering environment, the microseismic monitoring data obtained by the existing intelligent early warning method for underground engineering dynamic disasters may contain noise and inaccuracy, the disaster warning effect is not ideal, and during feature selection, it is impossible to well screen out features that are highly relevant to disaster prediction, resulting in some irrelevant features affecting the performance of the model and reducing the accuracy of the prediction, the present invention provides an intelligent early warning method and system for underground engineering dynamic disasters.

[0006] The technical solution provided by the embodiment of the present invention is as follows:

[0007] First aspect:

[0008] An embodiment of the present invention provides an intelligent early warning method for underground engineering power disasters, comprising:

[0009] S1: Obtain historical microseismic monitoring data of underground projects;

[0010] S2: Preprocess the historical microseismic monitoring data to obtain relevant time series parameters;

[0011] S3: performing wavelet decomposition and signal reconstruction on relevant time series parameters to obtain a reconstructed signal;

[0012] S4: Using the ARIMA model, multiple rolling statistical features of the reconstructed signal are obtained;

[0013] S5: Calculate the importance of each rolling statistical feature;

[0014] S6: Arrange each rolling statistical feature in descending order of importance, select a preset number of rolling statistical features with the highest order to form a rolling statistical feature matrix, and form a rolling statistical feature matrix data set from the rolling statistical feature matrices corresponding to each microseismic historical monitoring data;

[0015] S7: Constructing a prediction model for underground engineering dynamic disasters based on gradient boosting decision tree;

[0016] S8: input the rolling statistical feature matrix data set into the underground engineering dynamic disaster prediction model for training;

[0017] S9: Obtain real-time microseismic monitoring data;

[0018] S10: extracting a real-time rolling statistical feature matrix data set of real-time microseismic monitoring data;

[0019] S11: Input the real-time rolling statistical feature matrix data set into the trained underground engineering dynamic disaster prediction model and output the disaster warning result.

[0020] Second aspect:

[0021] An embodiment of the present invention provides an intelligent early warning system for underground engineering power disasters, comprising:

[0022] processor;

[0023] A memory having computer-readable instructions stored therein, which, when executed by a processor, implements the intelligent early warning method for dynamic disasters in underground engineering as described in the first aspect.

[0024] The third aspect:

[0025] An embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the intelligent early warning method for underground engineering power disasters as described in the first aspect is implemented.

[0026] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0027] In the present invention, through wavelet decomposition and signal reconstruction, noise can be effectively removed from microseismic monitoring data, and abnormal fluctuations in the signal can be captured more clearly, which is helpful to improve the quality of data, so that the early warning model can be trained and predicted based on more accurate data, and the rolling statistical characteristics of the signal are obtained by using the ARIMA model, and the characteristics with the greatest influence on disaster prediction are selected by calculating the importance of the characteristics, thereby optimizing the feature selection process, reducing the interference of irrelevant characteristics, and improving the prediction accuracy. Based on the gradient boosting decision tree, an underground engineering dynamic disaster prediction model is constructed, which can perform weighted training on multiple characteristics, adjust model parameters, and improve the fitting ability and accuracy of the model. The real-time rolling statistical feature matrix data set is input into the trained underground engineering dynamic disaster prediction model, and the thresholds of different rolling statistical characteristics are flexibly set, so that the early warning standard can be dynamically adjusted according to the environmental conditions of different underground projects, thereby improving the adaptability and accuracy of the early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0029] Figure 1 A schematic diagram of a flow chart of an intelligent early warning method for underground engineering power disasters provided by an embodiment of the present invention;

[0030] Figure 2 A schematic structural diagram of an intelligent early warning system for underground engineering dynamic disasters provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0031] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0032] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0033] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0034] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0035] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0036] Reference Manual Attached Figure 1 , showing a flow chart of an intelligent early warning method for underground engineering dynamic disasters provided by an embodiment of the present invention.

[0037] The embodiment of the present invention provides an intelligent early warning method for underground engineering power disasters, the method comprising:

[0038] S1: Obtain historical microseismic monitoring data of underground projects.

[0039] Among them, microseismic historical monitoring data refers to the historical records of microseismic events collected by sensors deployed in underground projects (such as seismometers, accelerometers, etc.). Microseismic refers to small-scale earthquakes or vibrations caused by rock rupture, geological movement, mining, etc. in the underground environment. These vibrations are usually invisible to the human eye, but can be monitored through high-precision sensors.

[0040] It should be noted that obtaining historical microseismic monitoring data of underground projects can provide valuable information about the long-term stability of underground projects and help identify potential safety hazards. By monitoring historical microseismic data, patterns and trends in microseismic activity can be discovered, providing a basis for predicting possible dynamic disasters.

[0041] S2: Preprocess the historical microseismic monitoring data to obtain relevant time series parameters.

[0042] Among them, the relevant time series parameters refer to the time series characteristics extracted from the processed microseismic data, which are used to describe the regularity and intensity of microseismic activities and provide a basis for subsequent disaster prediction.

[0043] In a possible implementation, the relevant time series parameters include microseismic cumulative energy, microseismic frequency, microseismic average energy and microseismic maximum energy.

[0044] The microseismic cumulative energy refers to the sum of the energy released by all microseismic events within a period of time.

[0045] Microseismic frequency refers to the number of microseismic events that occur within a certain period of time. Microseismic frequency is an important indicator for evaluating the dynamic changes of underground projects. The average microseismic energy refers to the average energy of all microseismic events within a certain period of time. It is obtained by calculating the energy of each microseismic event and taking its average value. The maximum microseismic energy refers to the maximum energy released by a single microseismic event recorded within a certain period of time.

[0046] It should be noted that preprocessing can remove noise and outliers, ensure data accuracy, and reduce interference in model training. The extracted time series parameters can help the model better capture the changing trends of microseismic activities and reveal the potential risks of underground projects. By accurately analyzing parameters such as the frequency and energy of microseismic activities, the precursors of disasters can be identified early, providing a more accurate basis for disaster warning, taking preventive measures in advance, and ensuring the safety of personnel and facilities.

[0047] S3: Perform wavelet decomposition and signal reconstruction on relevant time series parameters to obtain a reconstructed signal.

[0048] Among them, wavelet decomposition is a signal processing technology that can effectively reveal information of different scales or frequency bands of the signal by decomposing the original signal into multiple frequency components. Signal reconstruction refers to reconstructing the original signal or the improved signal by inverse transforming the decomposed signal components (low-frequency and high-frequency components). Signal reconstruction can restore important signal features, remove noise, and retain effective information. It is suitable for improving signal quality and extracting features.

[0049] In a possible implementation manner, the wavelet decomposition of the relevant time series parameters is specifically performed as follows:

[0050] The relevant time series parameters are decomposed into three layers using Haar wavelet.

[0051] Among them, Haar wavelet is a simple discrete wavelet function, which is often used for signal decomposition and compression. Three-layer decomposition refers to performing three wavelet decompositions on the signal, gradually decomposing the signal into multiple frequency levels.

[0052] It should be noted that wavelet decomposition can effectively extract local features from microseismic signals and can clearly distinguish signal components at different levels. Signal reconstruction can remove noise and interference, improve signal quality, and retain key dynamic features.

[0053] S4: Use the ARIMA model to obtain multiple rolling statistical features of the reconstructed signal.

[0054] Among them, the ARIMA model (autoregressive integrated moving average model) is a statistical model commonly used in time series data analysis and prediction. The ARIMA model captures the regularity of time series data by combining autoregression (AR), difference (I) and moving average (MA). The rolling statistical feature refers to a set of features obtained by sliding calculations on the data with a fixed window size and extracting statistics (such as mean, standard deviation, skewness, kurtosis, etc.) in the time series data.

[0055] In a possible implementation manner, the rolling statistical characteristic data specifically include: a rolling mean, a rolling standard deviation, a rolling skewness, a rolling kurtosis, a rolling upper quartile, and a rolling lower quartile.

[0056] Among them, the rolling mean refers to the mean of the data calculated according to a certain time step (or sample size) in a sliding window, and is continuously updated as the window moves. The rolling standard deviation refers to the standard deviation of the data in a sliding window, and is updated as the window slides. It reflects the volatility or discreteness of the data within a certain time range and is often used to measure the stability or abnormal fluctuations of time series. The rolling skewness refers to the skewness of the data distribution calculated in the sliding window. The rolling kurtosis refers to the sharpness of the data distribution calculated in the sliding window. The kurtosis measures the thickness of the tail of the data distribution. The rolling upper quartile refers to the 75th percentile of the data calculated in the sliding window, indicating the distribution position of larger values ​​in the data. The rolling lower quartile refers to the 25th percentile of the data calculated in the sliding window, indicating the distribution position of smaller values ​​in the data.

[0057] It should be noted that using the ARIMA model to obtain multiple rolling statistical features of the reconstructed signal can effectively capture the trend, seasonality and volatility in the time series. Combined with the rolling statistical features, it is possible to monitor the dynamic changes of the signal in different time windows, identify potential anomalies or disaster risks in a timely manner, and provide more accurate early warnings.

[0058] S5: Calculate the importance of each rolling statistical feature.

[0059] In a possible implementation, S5 is specifically:

[0060]

[0061] Among them, importance represents the importance of rolling statistical features, t represents the t-th decision tree, Gain (t, i ) represents the information gain when the rolling statistical features are split in decision tree t, and f j represents rolling statistical features, T represents the set of decision trees, T i Represents a split node in a decision tree.

[0062] It should be noted that calculating the importance of each rolling statistical feature can effectively identify the most influential features and remove irrelevant or redundant features, thereby improving the efficiency and accuracy of the model.

[0063] S6: Arrange each rolling statistical feature in descending order of importance, select a preset number of rolling statistical features with the highest ranking to form a rolling statistical feature matrix, and form a rolling statistical feature matrix data set with the rolling statistical feature matrices corresponding to each microseismic historical monitoring data.

[0064] It should be noted that sorting the rolling statistical features by importance and selecting the top few important features to form a feature matrix can improve the prediction accuracy and efficiency of the model. By giving priority to the features that have the greatest impact on disaster warning, the interference of irrelevant features can be avoided, reducing the complexity and computational burden of the model.

[0065] S7: Construct a prediction model for underground engineering dynamic disasters based on gradient boosting decision tree.

[0066] Among them, the gradient boosted decision tree (GBDT) is an integrated learning method that improves prediction performance by combining multiple decision tree models. GBDT adopts the idea of ​​an additive model. Each new tree is modified based on the previous tree to gradually reduce the model error.

[0067] It should be noted that GBDT can effectively handle complex nonlinear relationships and high-dimensional features, and is suitable for dealing with the variable data characteristics in underground engineering. GBDT has strong generalization ability and can achieve better results with less training data to avoid overfitting.

[0068] S8: Input the rolling statistical feature matrix data set into the underground engineering dynamic disaster prediction model for training.

[0069] In a possible implementation, S8 specifically includes:

[0070] S801: Divide the rolling statistical feature matrix data set into a training set and a test set, wherein the data duration ratio of the training set and the test set is not less than 7:3.

[0071] S802: Input the training set into the underground engineering dynamic disaster prediction model for training, and output the prediction test set.

[0072] S803: According to the test set and the prediction test set, the model parameters of the underground engineering dynamic disaster prediction model are adjusted until the performance index of the underground engineering dynamic disaster prediction model reaches the preset performance index standard.

[0073] Model parameters include: number of random seeds, number of trees, tree depth, learning rate, minimum sample split node, and minimum sample leaf node.

[0074] Among them, the random seed is a value used to initialize the random number generator, which is used when random selection is involved in the model training process (for example, data partitioning, model initialization, etc.). In the gradient boosted decision tree (GBDT), the number of trees refers to the total number of decision trees in the integrated model. The depth of the tree represents the maximum number of layers from the root node to the leaf node of the decision tree. The learning rate is a parameter that controls the degree of influence of each new tree on the final prediction result of the model. The minimum sample split node is a parameter in the decision tree algorithm, which is used to control the minimum number of samples required when splitting a node. The minimum sample leaf node is a parameter that controls the minimum number of samples required for each leaf node.

[0075] The performance indicators specifically include mean square error and coefficient of determination.

[0076] It should be noted that adjusting the parameters of the underground engineering dynamic disaster prediction model according to the test set and the prediction test set can optimize the prediction performance of the model, identify and correct model deviations, and improve the accuracy and robustness of the model.

[0077] In a possible implementation manner, the mean square error is specifically:

[0078]

[0079] Among them, MSE represents the mean square error, y i represents the i-th value of the rolling statistical feature test set, represents the i-th value of the rolling statistical feature prediction test set, , n represents the total number of data of rolling statistical features;

[0080] The coefficient of determination is:

[0081]

[0082] in, represents the coefficient of determination, Represents the average value of the rolling statistical feature test set;

[0083] The preset performance indicator standards are as follows:

[0084] < 0.01

[0085] > 0.6

[0086] Among them, MSE represents mean square error, represents the coefficient of determination.

[0087] It should be noted that inputting the rolling statistical feature matrix data set into the underground engineering dynamic disaster prediction model for training can enable the model to learn the complex relationship between various features in the time series and the occurrence of disasters. Through training, the model can identify potential risk signals based on historical data, thereby issuing early warnings before disasters occur.

[0088] S9: Obtain real-time microseismic monitoring data.

[0089] It should be noted that obtaining real-time microseismic monitoring data enables underground projects to instantly grasp the dynamic changes of the surrounding environment and promptly detect possible microseismic events, providing timely and accurate basic data for disaster warning, and being able to respond in the early stages of potential risks, thereby effectively avoiding or reducing the occurrence of disasters and ensuring project safety and the safety of personnel lives.

[0090] S10: Extracting a real-time rolling statistical feature matrix data set of real-time microseismic monitoring data.

[0091] S11: Input the real-time rolling statistical feature matrix data set into the trained underground engineering dynamic disaster prediction model and output the disaster warning result.

[0092] In a possible implementation, S11 specifically includes:

[0093] S1101: Input the real-time rolling statistical feature matrix data set into the trained underground engineering dynamic disaster prediction model.

[0094] S1102: By analyzing the rolling statistical characteristics of each relevant time series parameter in the time period t1 in real time, the rolling statistical characteristics of each relevant time series parameter in the time period t2 are predicted, wherein the ratio of t1 to t2 is not less than 7:3.

[0095] Specifically, by real-time analysis of the rolling statistical characteristics within the t1 time period to predict the characteristics of the t2 time period, it is possible to identify and predict the potential risks of underground projects in advance. The ratio of t1 to t2 is not less than 7:3, ensuring sufficient historical data as the basis for prediction and improving the accuracy and stability of the prediction.

[0096] S1103: When the rolling characteristics selected in time period t2 reach the warning critical index, a disaster warning is triggered.

[0097] S1104: Take the intersection of the time period sets that reach the warning critical index in the preset number of rolling statistical features selected for each time series parameter as the preliminary prediction results of each time series parameter, and take the union of the preliminary warning results obtained for each time series parameter to obtain the final warning result.

[0098] In a possible implementation, the critical warning indicator is specifically:

[0099] When the rolling statistical characteristics include rolling skewness or rolling kurtosis, the critical warning indicator is:

[0100] R f > a

[0101]

[0102] Among them, R f represents the pulse ratio, a represents the multiple pulse ratios R before the underground engineering power disaster occurs f The median of x indicates that the ratio of the amplitude of the rolling statistical characteristics within the time window to the initial amplitude is greater than the multiple amplitude ratios R before the underground engineering dynamic disaster occurs. a The median time period of the time window, X represents the total time period of the time window.

[0103] When the rolling statistical features include rolling upper and lower quartiles, rolling mean, or rolling standard deviation, the warning critical indicators are:

[0104] R a > b

[0105]

[0106] Where R a represents the amplitude ratio, and b represents the multiple amplitude ratios R before the underground engineering dynamic disaster occurs. a The median of A n It represents the amplitude value of the nth time period in the time window, and A1 represents the amplitude value of the initial time period.

[0107] It should be noted that inputting the real-time rolling statistical feature matrix data set into the trained underground engineering dynamic disaster prediction model can use the latest monitoring data for real-time analysis and prediction, and timely identify potential disaster risks and issue early warnings, thereby providing relevant personnel with sufficient time for emergency response, improving the safety of underground engineering and reducing losses.

[0108] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0109] In the present invention, through wavelet decomposition and signal reconstruction, noise can be effectively removed from microseismic monitoring data, and abnormal fluctuations in the signal can be captured more clearly, which is helpful to improve the quality of data, so that the early warning model can be trained and predicted based on more accurate data, and the rolling statistical characteristics of the signal are obtained by using the ARIMA model, and the characteristics with the greatest influence on disaster prediction are selected by calculating the importance of the characteristics, thereby optimizing the feature selection process, reducing the interference of irrelevant characteristics, and improving the prediction accuracy. Based on the gradient boosting decision tree, an underground engineering dynamic disaster prediction model is constructed, which can perform weighted training on multiple characteristics, adjust model parameters, and improve the fitting ability and accuracy of the model. The real-time rolling statistical feature matrix data set is input into the trained underground engineering dynamic disaster prediction model, and the thresholds of different rolling statistical characteristics are flexibly set, so that the early warning standard can be dynamically adjusted according to the environmental conditions of different underground projects, thereby improving the adaptability and accuracy of the early warning.

[0110] Reference Manual Attached Figure 2 , showing a structural schematic diagram of an intelligent early warning system for underground engineering dynamic disasters provided by the present invention.

[0111] The present invention further provides an intelligent early warning system 20 for underground engineering power disasters, which is applied to the above-mentioned intelligent early warning method for underground engineering power disasters, and comprises:

[0112] Processor 201.

[0113] The memory 202 stores computer-readable instructions. When the computer-readable instructions are executed by the processor 201, the intelligent early warning method for dynamic disasters in underground engineering as described in the method embodiment is implemented.

[0114] The intelligent early warning system 20 for underground engineering power disasters provided by the present invention can execute the above-mentioned intelligent early warning method for underground engineering power disasters and achieve the same or similar technical effects. To avoid repetition, the present invention will not go into details.

[0115] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0116] In the present invention, through wavelet decomposition and signal reconstruction, noise can be effectively removed from microseismic monitoring data, and abnormal fluctuations in the signal can be captured more clearly, which is helpful to improve the quality of data, so that the early warning model can be trained and predicted based on more accurate data, and the rolling statistical characteristics of the signal are obtained by using the ARIMA model, and the characteristics with the greatest influence on disaster prediction are selected by calculating the importance of the characteristics, thereby optimizing the feature selection process, reducing the interference of irrelevant characteristics, and improving the prediction accuracy. Based on the gradient boosting decision tree, an underground engineering dynamic disaster prediction model is constructed, which can perform weighted training on multiple characteristics, adjust model parameters, and improve the fitting ability and accuracy of the model. The real-time rolling statistical feature matrix data set is input into the trained underground engineering dynamic disaster prediction model, and the thresholds of different rolling statistical characteristics are flexibly set, so that the early warning standard can be dynamically adjusted according to the environmental conditions of different underground projects, thereby improving the adaptability and accuracy of the early warning.

[0117] It should be understood that the processor in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0118] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0119] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When a computer instruction or computer program is loaded or executed on a computer, a process or function according to an embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0120] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0121] In the present invention, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0122] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0123] Those skilled 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, or a combination of computer software and electronic hardware. 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.

[0124] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0125] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0126] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0127] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0128] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0129] An embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, an intelligent early warning method for dynamic disasters in underground engineering is implemented as in the method embodiment.

[0130] A computer-readable storage medium provided by the present invention can implement the steps and effects of the intelligent early warning method for underground engineering dynamic disasters in the above-mentioned method embodiment. To avoid repetition, the present invention will not go into details.

[0131] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0132] In the present invention, through wavelet decomposition and signal reconstruction, noise can be effectively removed from microseismic monitoring data, and abnormal fluctuations in the signal can be captured more clearly, which is helpful to improve the quality of data, so that the early warning model can be trained and predicted based on more accurate data, and the rolling statistical characteristics of the signal are obtained by using the ARIMA model, and the characteristics with the greatest influence on disaster prediction are selected by calculating the importance of the characteristics, thereby optimizing the feature selection process, reducing the interference of irrelevant characteristics, and improving the prediction accuracy. Based on the gradient boosting decision tree, an underground engineering dynamic disaster prediction model is constructed, which can perform weighted training on multiple characteristics, adjust model parameters, and improve the fitting ability and accuracy of the model. The real-time rolling statistical feature matrix data set is input into the trained underground engineering dynamic disaster prediction model, and the thresholds of different rolling statistical characteristics are flexibly set, so that the early warning standard can be dynamically adjusted according to the environmental conditions of different underground projects, thereby improving the adaptability and accuracy of the early warning.

[0133] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0134] There are a few points to note:

[0135] (1) The drawings of the embodiments of the present invention only involve structures related to the embodiments of the present invention. Other structures may refer to conventional designs.

[0136] (2) For the sake of clarity, in the drawings used to describe the embodiments of the present invention, the thickness of layers or regions is exaggerated or reduced, that is, these drawings are not drawn according to the actual scale. It is understood that when an element such as a layer, film, region or substrate is referred to as being "on" or "under" another element, the element may be "directly" "on" or "under" the other element or there may be intermediate elements.

[0137] (3) In the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other to obtain new embodiments.

[0138] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. An intelligent early warning method for underground engineering power disasters, characterized in that: include: S1: Obtain historical microseismic monitoring data of underground projects; S2: preprocessing the historical microseismic monitoring data to obtain relevant time series parameters; S3: performing wavelet decomposition and signal reconstruction on the relevant time series parameters to obtain a reconstructed signal; S4: using the ARIMA model to obtain multiple rolling statistical features of the reconstructed signal; S5: Calculating the importance of each of the rolling statistical features; S6: Arrange the rolling statistical features in descending order of importance, select a preset number of rolling statistical features with the highest order to form a rolling statistical feature matrix, and form a rolling statistical feature matrix data set from the rolling statistical feature matrices corresponding to the microseismic historical monitoring data; S7: Constructing a prediction model for underground engineering dynamic disasters based on gradient boosting decision tree; S8: inputting the rolling statistical feature matrix data set into the underground engineering dynamic disaster prediction model for training; S9: Obtain real-time microseismic monitoring data; S10: extracting a real-time rolling statistical feature matrix data set of the real-time microseismic monitoring data; S11: Input the real-time rolling statistical feature matrix data set into the trained underground engineering dynamic disaster prediction model, and output the disaster warning result.

2. The intelligent early warning method for underground engineering power disasters according to claim 1 is characterized in that: The relevant time series parameters include microseismic cumulative energy, microseismic frequency, microseismic average energy and microseismic maximum energy.

3. The intelligent early warning method for underground engineering power disasters according to claim 1 is characterized in that: The wavelet decomposition of the relevant time series parameters is specifically as follows: The related time series parameters are decomposed into three layers by using Haar wavelet.

4. The intelligent early warning method for underground engineering power disasters according to claim 1 is characterized in that: The rolling statistical characteristic data specifically include: rolling mean, rolling standard deviation, rolling skewness, rolling kurtosis, rolling upper quartile and rolling lower quartile.

5. The intelligent early warning method for underground engineering power disasters according to claim 1 is characterized in that: The S5 is specifically: ； Among them, importance represents the importance of the rolling statistical feature, t represents the t-th decision tree, Gain (t, i ) represents the information gain when the rolling statistical feature is split in decision tree t, and f j represents rolling statistical features, T represents the set of decision trees, T i Represents a split node in a decision tree.

6. The intelligent early warning method for underground engineering power disasters according to claim 1 is characterized in that: The S8 specifically includes: S801: Divide the rolling statistical feature matrix data set into a training set and a test set, wherein the data duration ratio of the training set to the test set is not less than 7:3; S802: inputting the training set into the underground engineering dynamic disaster prediction model for training, and outputting a prediction test set; S803: adjusting the model parameters of the underground engineering dynamic disaster prediction model according to the test set and the prediction test set until the performance index of the underground engineering dynamic disaster prediction model reaches a preset performance index standard; The model parameters include: the number of random seeds, the number of trees, the depth of the tree, the learning rate, the minimum sample split node and the minimum sample leaf node; The performance indicators specifically include mean square error and determination coefficient.

7. The intelligent early warning method for underground engineering power disasters according to claim 6 is characterized in that: The mean square error is specifically: ； Among them, MSE represents the mean square error, y i represents the i-th value of the rolling statistical feature test set, represents the i-th value of the rolling statistical feature prediction test set, , n represents the total number of data in the rolling statistical feature test set; The determination coefficient is specifically: ； in, represents the coefficient of determination, Represents the average value of the rolling statistical feature test set; The preset performance index standards are specifically: < 0.01 > 0.6 Among them, MSE represents mean square error, represents the coefficient of determination.

8. The intelligent early warning method for underground engineering power disasters according to claim 1 is characterized in that: The S11 specifically includes: S1101: inputting the real-time rolling statistical feature matrix data set into the trained underground engineering dynamic disaster prediction model; S1102: by real-time analysis of rolling statistical features of various relevant time series parameters in time period t1, rolling statistical features of various relevant time series parameters in time period t2 are predicted, wherein the ratio of t1 to t2 is not less than 7:3; S1103: When the rolling feature selected in time period t2 reaches the warning critical index, a disaster warning is triggered; S1104: taking the intersection of the time period set that reaches the warning critical indicator in the preset number of rolling statistical features selected for each time series parameter as the preliminary prediction result of each time series parameter, and taking the union of the preliminary warning results obtained for each time series parameter to obtain the final warning result.

9. The intelligent early warning method for underground engineering power disasters according to claim 8 is characterized in that: The critical early warning indicators are specifically: When the rolling statistical feature includes rolling skewness or rolling kurtosis, the warning critical indicator is: R f > a ； Among them, R f represents the pulse ratio, a represents the multiple pulse ratios R before the underground engineering power disaster occurs f The median of x indicates that the ratio of the amplitude of the rolling statistical characteristics within the time window to the initial amplitude is greater than the multiple amplitude ratios R before the underground engineering dynamic disaster occurs. a The median time period of , X represents the total time period of the time window; When the rolling statistical features include rolling upper and lower quartiles, rolling mean or rolling standard deviation, the warning critical indicator is: R a > b ； Where R a represents the amplitude ratio, and b represents the multiple amplitude ratios R before the underground engineering dynamic disaster occurs. a The median of A n It represents the amplitude value of the nth time period in the time window, and A1 represents the amplitude value of the initial time period.

10. An intelligent early warning system for underground engineering power disasters, characterized in that: include: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, an intelligent early warning method for dynamic disasters in underground engineering as described in any one of claims 1 to 9 is implemented.

Images