Deep tunnel rockburst disaster blasting circulation level early warning method and device and processing equipment

Abstract

The invention provides a deep tunnel rockburst disaster blasting circulation level early warning method and device and processing equipment, which are used for analyzing a micro-seismic characteristic evolution rule of a deep tunnel rockburst disaster on the basis of four stages of a specially-designed rockburst inoculation process, so that a deep tunnel rockburst disaster blasting circulation level early warning model obtained by training has a good early warning effect on the deep tunnel rockburst disaster. The rock burst disaster occurrence stage and the rock burst level can be accurately early warned at the blasting circulation level, and good data support is provided for deep tunnel engineering.

Description

Technical Field

[0001] This application relates to the field of geology, specifically to a method, device, and processing equipment for early warning of rockburst disasters in deep-buried tunnels during blasting cycles. Background Technology

[0002] With the continuous development of underground engineering, the application of microseismic monitoring technology in underground space development is becoming increasingly important, especially in fields such as mining, tunnel construction, and nuclear waste disposal. Researchers have proposed a variety of rockburst early warning methods by analyzing the characteristic parameters in microseismic data, which has greatly improved the safety of deep-buried tunnel engineering.

[0003] However, the inventors of this application have found that, despite the significant potential of microseismic monitoring technology in disaster prediction and early warning, its current application still faces many challenges and shortcomings: 1) Existing microseismic data analysis methods are mostly focused on the characteristic analysis of short bursts, while the microseismic activity patterns during long quiet periods have not been fully studied, resulting in low accuracy of early warning systems when dealing with microseismic activity during long quiet periods. 2) Current early warning methods mainly rely on real-time monitoring data to predict rockburst events, which makes it impossible for the system to achieve true early warning under some key conditions.

[0004] Therefore, it is urgent to improve existing microseismic monitoring and early warning methods, especially in understanding the patterns of microseismic activity during long periods of calm, and to further enhance the ability to predict and warn of rockburst disasters. Summary of the Invention

[0005] This application provides a method, device, and processing equipment for early warning of rockburst disasters in deep-buried tunnels based on a specially designed four-stage rockburst incubation process. This method analyzes the evolution of microseismic characteristics of rockburst disasters in deep-buried tunnels. The early warning model for rockburst disasters in deep-buried tunnels trained in this way can accurately predict the stage and level of rockburst disasters at the blasting cycle level, providing good data support for deep-buried tunnel engineering.

[0006] Firstly, this application provides a method for early warning of rockburst disasters in deep-buried tunnels using a blasting cycle approach, the method comprising: For deeply buried tunnels, the first microseismic monitoring data within a historical time period is obtained as the initial sample; The first microseismic monitoring data is divided into multiple sub-microseismic monitoring data by using three data intervals formed by four consecutive rockburst occurrence times as a data division unit. The stage in which the three data intervals are located is regarded as three rockburst disaster incubation stages, and the stage in which the last rockburst occurrence time in the four consecutive rockburst occurrence times is regarded as the rockburst disaster occurrence stage. The four consecutive rockburst occurrence times are marked with corresponding rockburst occurrence levels. Based on the training samples configured using the second microseismic monitoring data, a blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels was trained. This model is a deep learning model, which is used to predict the stage of rockburst disaster occurrence and the corresponding rockburst occurrence level based on the microseismic monitoring data within the continuous time period input to the model, combined with the microseismic monitoring data from three data intervals.

[0007] Secondly, this application provides a blasting cycle-based early warning device for rockburst disasters in deep-buried tunnels, the device comprising: The acquisition unit is used to acquire the first microseismic monitoring data within a historical time period for deep-buried tunnels, as an initial sample; The division unit is used to divide the first microseismic monitoring data into multiple sub-microseismic monitoring data by using three data intervals formed by four consecutive rockburst occurrence times as a data division unit, forming the second microseismic monitoring data. The stage in which the three data intervals are located is regarded as three rockburst disaster incubation stages, and the stage in which the last rockburst occurrence time in the four consecutive rockburst occurrence times is regarded as the rockburst disaster occurrence stage. The four consecutive rockburst occurrence times are marked with corresponding rockburst occurrence levels. The training unit is used to train a blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels based on training samples configured from the second microseismic monitoring data. The blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels is a deep learning model. Based on the microseismic monitoring data within the continuous time period input to the model, the model combines microseismic monitoring data from three data intervals to predict the stage of rockburst disaster occurrence and the corresponding rockburst occurrence level.

[0008] Thirdly, this application provides a processing device, including a processor and a memory, wherein a computer program is stored in the memory, and the processor executes the method provided in the first aspect of this application when it invokes the computer program in the memory.

[0009] Fourthly, this application provides a computer-readable storage medium storing a plurality of instructions adapted for loading by a processor to execute the method provided in the first aspect of this application.

[0010] From the above, it can be concluded that this application has the following beneficial effects: For the purpose of early warning of rockburst disasters in deep-buried tunnels at the blast cycle level, this application is based on a specially conceived four-stage rockburst incubation process to analyze the evolution law of microseismic characteristics of rockburst disasters in deep-buried tunnels. The early warning model of rockburst disasters at the blast cycle level trained in this way can accurately predict the occurrence stage and rockburst level of rockburst disasters at the blast cycle level, providing good data support for deep-buried tunnel engineering. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a flowchart illustrating a blasting cycle-based early warning method for rockburst disasters in deep-buried tunnels, as described in this application. Figure 2 This is a schematic diagram of a blasting cycle-level early warning device for rockburst disasters in deep-buried tunnels, as per this application. Figure 3 This is a schematic diagram of one type of processing equipment used in this application. Detailed Implementation

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0014] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules is not necessarily limited to those explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices. The naming or numbering of steps appearing in this application does not imply that the steps in the method flow must be performed in the chronological / logical order indicated by the naming or numbering. The execution order of named or numbered process steps can be changed according to the desired technical purpose, as long as the same or similar technical effect is achieved.

[0015] The module division described in this application is a logical division. In practical applications, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the coupling or direct coupling or communication connection between modules shown or discussed may be through some interfaces, and the indirect coupling or communication connection between modules may be electrical or other similar forms, none of which are limited in this application. Furthermore, the modules or sub-modules described as separate components may or may not be physically separated, may or may not be physical modules, or may be distributed in multiple circuit modules. Some or all of the modules may be selected to achieve the purpose of the solution in this application according to actual needs.

[0016] Before introducing the blasting cycle-level early warning method for rockburst disasters in deep-buried tunnels provided in this application, the background content involved in this application will be introduced first.

[0017] The method, device, and computer-readable storage medium for early warning of rockburst disasters in deep-buried tunnels provided in this application can be applied to processing equipment to analyze the evolution of microseismic characteristics of rockburst disasters in deep-buried tunnels based on a specially conceived four-stage rockburst incubation process. The early warning model for rockburst disasters in deep-buried tunnels trained in this way can accurately predict the stage of rockburst disaster occurrence and its rockburst level at the blasting cycle level, providing good data support for deep-buried tunnel engineering.

[0018] The method for early warning of rockburst disasters in deep-buried tunnels using a cyclical early warning system can be implemented by a cyclical early warning device for rockburst disasters in deep-buried tunnels, or by different types of processing devices such as servers, physical hosts, or user equipment (UE) that integrate the cyclical early warning device. The cyclical early warning device can be implemented in hardware or software. The UE can be a smartphone, tablet, laptop, desktop computer, or personal digital assistant (PDA) or other terminal device. The processing devices can be configured in a cluster.

[0019] It is understandable that in practical applications, considering that the data processing in this application is mainly based on existing data, the processing equipment that implements the deep-buried tunnel rockburst disaster blasting cycle early warning method of this application or that is equipped with the corresponding application service of the deep-buried tunnel rockburst disaster blasting cycle early warning method of this application usually only needs to meet the required data processing capabilities, and its specific equipment type and equipment deployment form are quite flexible.

[0020] If it involves the direct acquisition capability of existing data, namely relevant microseismic monitoring data, as mentioned above, it is understandable that the existing microseismic monitoring data acquisition system can be integrated into the equipment cluster of the processing equipment, or the existing microseismic monitoring data acquisition system can be integrated into the processing equipment through hardware and software modifications. Of course, it is also possible to call the existing microseismic monitoring data acquisition system by calling a service.

[0021] Similarly, if there is a need to display the dynamics of the solution processing (such as model prediction results), the processing device can also promote the corresponding content display through its own configured display screen (including touch screen), external display devices, or other devices configured with display screens.

[0022] Alternatively, corresponding to different processing stages such as model training and model application, the processing equipment can also be divided into several equipment parts to form a more flexible equipment deployment architecture that can meet diverse application needs. The following section introduces the blasting cycle-based early warning method for rockburst disasters in deep-buried tunnels provided in this application.

[0023] First, refer to Figure 1 , Figure 1 This paper presents a flowchart illustrating a method for early warning of rockburst hazards in deep-buried tunnels using a blasting cycle approach, which may include the following steps S101 to S103: Step S101: For deeply buried tunnels, acquire the first microseismic monitoring data within a historical time period as the initial sample; Understandably, to meet the training requirements of the blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels, it is necessary to obtain microseismic monitoring data from historical time periods / past time periods. This microseismic monitoring data is either real data directly collected by the corresponding microseismic sensors deployed in deep-buried tunnels, or data processed based on the real data directly collected by the corresponding microseismic sensors deployed in deep-buried tunnels, such as some specific indicators.

[0024] For ease of explanation, this application distinguishes the microseismic monitoring data at different stages using the terms "first", "second", and "target".

[0025] In practice, the acquisition and processing of the first microseismic monitoring data is usually done by extracting existing data, such as by manual input or retrieval from the system. Of course, it is also possible that in some cases, as the scheme of this application is deployed, microseismic monitoring data will be continuously / dynamically collected during the excavation of the deep-buried tunnel to meet the data usage requirements of this application scheme. This can be included in the scope of processing dynamic / real-time acquisition of microseismic monitoring data.

[0026] As for the microseismic sensors that specifically collect the first microseismic monitoring data, they are relatively mature and common sensors in the engineering field. Therefore, not much is said about the microseismic sensors here. The deployment location of the sensors is quite flexible and can be adapted to actual needs and the progress of the deep-buried tunnel excavation project. For example, the sensors can be deployed at a certain distance in front of or behind the tunnel face to better capture the microseismic signals at the tunnel face.

[0027] Furthermore, it is understandable that the first microseismic monitoring data can be either data collected from the current deep-buried tunnel, thus having a high degree of specificity, or data collected from different deep-buried tunnels, thus having greater generalizability. This can be adjusted according to actual needs.

[0028] Step S102: Using the three data intervals formed by four consecutive rockburst occurrence times as a data division unit, the first microseismic monitoring data is divided into multiple sub-microseismic monitoring data to form the second microseismic monitoring data. The stage in which the three data intervals are located is regarded as three rockburst disaster incubation stages, and the stage in which the last rockburst occurrence time in the four consecutive rockburst occurrence times is regarded as the rockburst disaster occurrence stage. The four consecutive rockburst occurrence times are marked with corresponding rockburst occurrence levels. As a key aspect of this application, it is understandable that the inventors of this application have specifically conceived / designed a four-stage mechanism for rockburst incubation, which serves as a blasting cycle-level early warning concept / theory to assist in deep-level, high-precision blasting cycle-level early warning of rockburst disasters in deeply buried tunnels.

[0029] Specifically, for continuous rockbursts, this application divides the rockburst incubation interval by the blasting time node of the currently selected rockburst occurrence time and the time nodes of the rockburst occurrence times of the previous three excavation cycles. The time period in which the microseismic monitoring data of the previous three cycles are located is defined as the rockburst disaster incubation stage, specifically stage I, stage II, and stage III. The time point in which the microseismic monitoring data of the currently selected rockburst occurrence time is defined as the rockburst disaster occurrence stage, i.e., stage IV. Stage IV is the prediction target.

[0030] In this case, the first microseismic monitoring data obtained above can be understood to involve a large number of rockburst occurrence times. At this time, based on the four-stage mechanism of the rockburst gestation process, and corresponding to the configuration requirements of the model prediction logic / subsequent training samples, the data can be divided into multiple groups of sub-microseismic monitoring data, using the three data intervals formed by four consecutive rockburst occurrence times as a data division unit, thus forming the second microseismic monitoring data at the overall level.

[0031] Furthermore, it is understandable that the rockburst occurrence time is marked with a corresponding rockburst occurrence level. This corresponds to the situation where the subsequent deep-buried tunnel rockburst disaster blasting cycle level early warning model not only predicts the occurrence time of the rockburst disaster stage (stage IV) but also predicts the corresponding rockburst occurrence level.

[0032] If the original / initial first microseismic monitoring data does not include the corresponding rockburst occurrence level at the time of rockburst occurrence, then after obtaining the first microseismic monitoring data, it is also necessary to extract / identify the corresponding rockburst occurrence level at the time of rockburst occurrence.

[0033] Step S103: Based on the training samples configured from the second microseismic monitoring data, train a blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels. The blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels is a deep learning model. The blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels is used to predict the stage of rockburst disaster occurrence and the corresponding rockburst occurrence level based on the microseismic monitoring data within the continuous time period input to the model, combined with the microseismic monitoring data from three data intervals.

[0034] After processing the second set of microseismic monitoring data, it is understandable that the specific model training phase can then proceed.

[0035] In model training, the general approach is as follows: based on a large number of training samples with labeled true prediction values, in each training round, a training sample is input into the model, allowing the model to perform corresponding prediction processing to achieve forward propagation. Then, based on the prediction results input into the model, the loss function is calculated in combination with the true prediction values, and the model parameters are optimized based on the loss function calculation results to achieve backpropagation. After a large number of iterations of training, when the training requirements such as training time, number of training sessions, or prediction accuracy are met, the model training can be completed, and a prediction model that can be put into practical use can be obtained.

[0036] For this application, it is understood that the sub-microseismic monitoring data of each group in the second microseismic monitoring data can be used as a sample of labeled prediction true values, that is, the model input corresponding to the first three rockburst disaster incubation stages, and the model output corresponding to the rockburst disaster occurrence stage / the fourth rockburst occurrence time.

[0037] Thus, under a specific model training scheme, the corresponding training samples can be obtained by configuring multiple sets of sub-microseismic monitoring data in the second microseismic monitoring data. Specifically, in the model training scheme, the training samples can also be divided into training group / data, test group / data, or they can also be divided into training group / data, test group / data, and validation group / data. This can also involve extraction processing under extraction strategies such as random extraction strategies.

[0038] Understandably, the specific model structure, training scheme, and loss function used during model training can be flexibly configured according to actual needs. Existing schemes can be used, further optimizations can be made based on existing schemes, or novel schemes can be developed in-house.

[0039] As an example, a traditional model training scheme can be adopted, in which the original samples are extracted into training samples consisting of 80% training data and 20% test data.

[0040] Thus, under the four-stage mechanism of rockburst incubation process specifically designed in this application, combined with the powerful artificial intelligence (AI) learning capabilities of the deep learning model itself, the trained deep-buried tunnel rockburst disaster blasting cycle early warning model can have a deep and high-precision deep-buried tunnel rockburst disaster blasting cycle early warning capability.

[0041] It should be noted that, unlike the monitoring data characteristics of short burst periods that traditional technologies focus on, the blasting cycle-level early warning involved in this application refers to the prediction of rockburst events based on microseismic monitoring data with a large time span and including microseismic activity during long quiet periods.

[0042] In reality, the time span between different rockburst disaster events is very large. Traditional technologies focus on predicting short-term or impending rockburst disasters based on relevant monitoring data characteristics before a rockburst disaster occurs.

[0043] Clearly, there is a fundamental difference between the prediction logic involved in traditional technology and the solution proposed in this application. Even if the original training samples in traditional technology involve a large number of rockburst disasters, each rockburst disaster is still separated in the specific model training work. This is because the prediction logic only focuses on the characteristics of short-term monitoring data, which directly ignores the complex microseismic activity characteristics of long quiet periods that are not easy to identify without realizing it.

[0044] Thus, for the relevant decision-making and specific implementation of deep-buried tunnel projects, this application can analyze the microseismic activity characteristics of long quiet periods from a longer-term and more forward-looking perspective, reveal potential rockburst disaster signals, and provide data support for the prediction results of rockburst disasters. This can ensure that deep-buried tunnel projects can be promoted more safely and stably, deployed in a timely manner, and avoid the loss of human and material resources.

[0045] Furthermore, as an exemplary embodiment, the model architecture of the early warning model for rockburst disaster blasting cycle in deep-buried tunnels specifically adopts the Transformer model.

[0046] Understandably, the attention-based Transformer model, with its advantage in capturing long-term dependencies in time-series data, can provide more reliable results even with complex and uncertain data by systematically learning the evolution of characteristic parameters in different rockburst cases at each stage. It can accurately predict the values ​​of characteristic parameters in the rockburst disaster occurrence stage (stage IV) of the test set and further determine the rockburst level, thereby achieving blasting cycle-level early warning for rockburst disasters.

[0047] In terms of specific model structure, in short, the Transformer model mainly consists of two parts: an encoder and a decoder. Each part is composed of several identical layers stacked together. Each layer of the encoder contains two sub-layers: a multi-head self-attention mechanism and a feedforward neural network. Each layer of the decoder adds a multi-head attention sub-layer to process the encoder output.

[0048] At the same time, in practical applications, this application also considers the limited number of rockburst cases and the significant imbalance of sample categories in traditional operations, which further limits the generalization ability and prediction accuracy of existing early warning systems.

[0049] In this regard, as an exemplary embodiment, before training the blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels based on the training samples configured using the second microseismic monitoring data in step S103, the method of this application may further include: The second microseismic monitoring data were sample augmented to balance the number of samples from different rockburst case categories.

[0050] Understandably, the sample expansion / processing here effectively increases the data volume / sample size based on the original data volume of the second microseismic monitoring data. On the other hand, it also pays attention to class balance during the sample expansion process, considering the different rockburst case sample categories originally involved in the second microseismic monitoring data.

[0051] In this way, by increasing the sample size and balancing the categories, the high-quality model training effect can be further guaranteed, thereby ensuring the generalization ability and prediction accuracy of the blasting cycle early warning model for rockburst disasters in deep-buried tunnels.

[0052] Furthermore, as an exemplary embodiment, the sample augmentation of the second microseismic monitoring data mentioned above may specifically include: The second microseismic monitoring data were augmented using a pre-configured GAN model.

[0053] GAN stands for Generative Adversarial Networks.

[0054] As is understandable, a GAN model comprises two main components: a generator and a discriminator. The generator's task is to generate realistic data, while the discriminator's task is to distinguish between generated and real data. The core idea of ​​the GAN model is to use a game-theoretic adversarial process. Through their interaction, the generator continuously improves the quality of its generated samples, while the discriminator continuously enhances its ability to distinguish between real and fake samples. Ultimately, when the GAN model is fully trained, the generator can generate samples that closely resemble real data, effectively fulfilling the sample augmentation task required by this application.

[0055] Meanwhile, in some designs of this application, the model prediction of the rockburst occurrence level can also be based on the key reference part of the characteristic parameters. In this regard, this application can also perform additional calculations and annotations on these characteristic parameters involved in the second microseismic monitoring data, which corresponds to the corresponding characteristic parameter calculation work involved in the subsequent actual application of the model.

[0056] Correspondingly, as an exemplary embodiment, the method of this application may further include: For the second microseismic monitoring data, the characteristic parameters corresponding to each stage are calculated. These characteristic parameters include the cumulative logarithm of microseismic energy release, the cumulative apparent volume, the cumulative number of events, and the logarithm of energy density.

[0057] In this context, the early warning model for rockburst disasters in deep-buried tunnels can specifically involve predicting the corresponding rockburst occurrence level based on the prediction results of characteristic parameters at each stage of the rockburst disaster.

[0058] Among them, the logarithm of microseismic energy release is the logarithm of the seismic energy released by a microseismic event, base 10; the cumulative apparent volume refers to the sum of the spatial volumes affected by all microseismic events within a certain period of time, reflecting the spatial distribution range of microseismic activity; the cumulative number of events refers to the total number of valid microseismic events within the monitoring period, reflecting the frequency of microseismic activity; and the logarithm of energy density is the logarithm of the ratio of cumulative total energy to cumulative apparent volume, base 10, reflecting the energy release intensity per unit volume.

[0059] Understandably, these four characteristic parameters are existing / mature indicators in earthquake / microseismic research. Therefore, considering that the meaning of the indicators themselves is not the focus of this application, they are not explained in detail here.

[0060] Thus, after completing the training of the blasting cycle early warning model for rockburst disasters in deep-buried tunnels, it can be put into practical use. According to the specific application requirements of the model, it can be specifically configured in the corresponding application scenarios to promote the corresponding blasting cycle early warning work for rockburst disasters in deep-buried tunnels based on the real-time microseismic monitoring requirements.

[0061] It is easy to understand that this application can also involve specific model application stages. Correspondingly, as an exemplary embodiment, after training the deep-buried tunnel rockburst disaster blasting cycle early warning model based on the training samples configured according to the second microseismic monitoring data in step S103, the method of this application may further include: Obtain the target microseismic monitoring data that is currently needed for prediction; The target microseismic monitoring data is input into the blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels for predictive processing; Extract the prediction results from the early warning model of rockburst disaster blasting cycle in deep-buried tunnels.

[0062] The specific data source for the microseismic monitoring data of this target is quite flexible. It can be manually input, real-time acquisition, or retrieved from local storage locations or online systems.

[0063] Furthermore, in practical operation, the model can also be triggered by a corresponding early warning mission for rockburst disaster blasting cycle in deep-buried tunnels. In this case, the target microseismic monitoring data can be either directly carried in the mission information or obtained according to the instructions in the mission information.

[0064] Among them, the blasting cycle-level early warning task for rockburst disasters in deep-buried tunnels can be initiated manually, autonomously according to the corresponding task initiation strategy, or by receiving tasks forwarded from other devices. These can all be flexibly configured according to the actual situation.

[0065] Thus, after inputting the target microseismic monitoring data that needs to be processed into the early warning task for rockburst disaster in deep-buried tunnels, the corresponding prediction results of the target microseismic monitoring data can be extracted from the model output.

[0066] Understandably, once the prediction results from the model are extracted, further data processing can be involved, such as local storage, off-site storage, result display, output completion notification, result push, or further data analysis. The specific processing can be flexibly configured according to the pre-configured and real-time data processing strategies.

[0067] In accordance with the specific application requirements of deep-buried tunnel engineering, as a preferred embodiment, after extracting the prediction results output by the deep-buried tunnel rockburst disaster blasting cycle early warning model, the method of this application may further include: The response is driven by the rockburst disaster monitoring system based on the prediction results.

[0068] Understandably, in practical applications, organizations / teams can configure an online rockburst disaster monitoring system to meet the work needs of deep-buried tunnel projects. The proposed solution can be directly integrated into this system, or the proposed solution can be provided as an external data service for the system to call or push corresponding processing results to the system.

[0069] When the rockburst disaster monitoring system is specifically designed to monitor and manage rockburst disaster-related work in deep-buried tunnels, it can trigger further response processing based on the current prediction results. For example, it can issue personnel warnings in the form of broadcasts or voice calls within the system, or control relevant equipment in deep-buried tunnels to carry out evacuation / relocation work under the established Internet of Things mechanism.

[0070] Thus, with the blasting cycle-level early warning already in place, further specific response measures significantly improve the safety control capabilities in engineering construction. Especially in complex engineering environments such as deep tunnels, timely measures can be taken to reduce the risk of sudden rockburst disasters and ensure the safety of construction personnel and equipment.

[0071] In conclusion, regarding the above-mentioned scheme, this application, based on a specially conceived four-stage rockburst incubation process, analyzes the evolution of microseismic characteristics of rockburst disasters in deep-buried tunnels to provide early warning of rockburst disasters in blasting cycles for deep-buried tunnels. The resulting early warning model for rockburst disasters in blasting cycles can accurately predict the stage and level of rockburst disasters, providing excellent data support for deep-buried tunnel engineering.

[0072] The above is an introduction to the blasting cycle-level early warning method for rockburst disasters in deep-buried tunnels provided in this application. In order to facilitate better implementation of the blasting cycle-level early warning method for rockburst disasters in deep-buried tunnels provided in this application, this application also provides a blasting cycle-level early warning device for rockburst disasters in deep-buried tunnels from the perspective of functional modules.

[0073] See Figure 2 , Figure 2 This is a schematic diagram of a blasting cycle-level early warning device for rockburst disasters in deep-buried tunnels, as described in this application. Specifically, the blasting cycle-level early warning device 200 for rockburst disasters in deep-buried tunnels may include the following structure: The acquisition unit 201 is used to acquire the first microseismic monitoring data within a historical time period for deep-buried tunnels as an initial sample. The division unit 202 is used to divide the first microseismic monitoring data into multiple sub-microseismic monitoring data by taking three data intervals formed by four consecutive rockburst occurrence times as a data division unit, forming the second microseismic monitoring data. The stage in which the three data intervals are located is regarded as three rockburst disaster incubation stages, and the stage in which the last rockburst occurrence time in the four consecutive rockburst occurrence times is regarded as the rockburst disaster occurrence stage. The four consecutive rockburst occurrence times are marked with corresponding rockburst occurrence levels. Training unit 203 is used to train a blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels based on training samples configured from the second microseismic monitoring data. The blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels is a deep learning model. Based on the microseismic monitoring data within the continuous time period input to the model, the model combines microseismic monitoring data from three data intervals to predict the stage of rockburst disaster occurrence and the corresponding rockburst occurrence level.

[0074] As an exemplary embodiment, the model architecture of the early warning model for rockburst disaster blasting cycle in deep-buried tunnels specifically adopts the Transformer model.

[0075] As another exemplary embodiment, the device further includes an expansion unit 204 for... The second microseismic monitoring data were sample augmented to balance the number of samples from different rockburst case categories.

[0076] As another exemplary embodiment, sample augmentation of the second microseismic monitoring data includes: The second microseismic monitoring data were augmented using a pre-configured GAN model.

[0077] As another exemplary embodiment, the device further includes a computing unit 205, for: For the second microseismic monitoring data, the characteristic parameters corresponding to each stage are calculated. The characteristic parameters include the cumulative logarithm of microseismic energy release, the cumulative apparent volume, the cumulative number of events, and the logarithm of energy density. The early warning model for rockburst disasters in deep-buried tunnels is based on the prediction results of characteristic parameters at each stage of the rockburst disaster to predict the corresponding rockburst occurrence level.

[0078] As another exemplary embodiment, the device further includes an application unit 206 for: Obtain the target microseismic monitoring data that is currently needed for prediction; The target microseismic monitoring data is input into the blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels for predictive processing; Extract the prediction results from the early warning model of rockburst disaster blasting cycle in deep-buried tunnels.

[0079] As another exemplary embodiment, the device further includes a response unit 207, configured to: The response is driven by the rockburst disaster monitoring system based on the prediction results.

[0080] This application also provides a processing device from a hardware architecture perspective, see [link / reference]. Figure 3 , Figure 3 This diagram illustrates a structural schematic of the processing device of this application. Specifically, the processing device may include a processor 301, a memory 302, and an input / output device 303. The processor 301 executes the computer program stored in the memory 302 to implement, for example... Figure 1 The corresponding embodiments describe the steps of the blasting cycle-based early warning method for rockburst disasters in deep-buried tunnels; or, when the processor 301 executes the computer program stored in the memory 302, it implements the following: Figure 2 Corresponding to the functions of each unit in the embodiment, the memory 302 is used to store the functions executed by the processor 301 as described above. Figure 1 The computer program required for the blasting cycle-level early warning method for rockburst disasters in deep-buried tunnels in the corresponding embodiment.

[0081] For example, a computer program may be divided into one or more modules / units, one or more of which are stored in memory 302 and executed by processor 301 to complete this application. One or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in a computer device.

[0082] The processing device may include, but is not limited to, processor 301, memory 302, and input / output device 303. Those skilled in the art will understand that the illustrations are merely examples of the processing device and do not constitute a limitation on the processing device. It may include more or fewer components than illustrated, or combine certain components, or different components. For example, the processing device may also include network access devices, buses, etc., and processor 301, memory 302, input / output device 303, etc., are connected via a bus.

[0083] Processor 301 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the processing device, connecting various parts of the device through various interfaces and lines.

[0084] The memory 302 can be used to store computer programs and / or modules. The processor 301 implements various functions of the computer device by running or executing the computer programs and / or modules stored in the memory 302 and by calling data stored in the memory 302. The memory 302 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function, etc.; the data storage area may store data created according to the use of the processing device, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0085] When processor 301 executes a computer program stored in memory 302, it can specifically perform the following functions: For deeply buried tunnels, the first microseismic monitoring data within a historical time period is obtained as the initial sample; The first microseismic monitoring data is divided into multiple sub-microseismic monitoring data by using three data intervals formed by four consecutive rockburst occurrence times as a data division unit. The stage in which the three data intervals are located is regarded as three rockburst disaster incubation stages, and the stage in which the last rockburst occurrence time in the four consecutive rockburst occurrence times is regarded as the rockburst disaster occurrence stage. The four consecutive rockburst occurrence times are marked with corresponding rockburst occurrence levels. Based on the training samples configured using the second microseismic monitoring data, a blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels was trained. This model is a deep learning model, which is used to predict the stage of rockburst disaster occurrence and the corresponding rockburst occurrence level based on the microseismic monitoring data within the continuous time period input to the model, combined with the microseismic monitoring data from three data intervals.

[0086] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the deep-buried tunnel rockburst disaster blasting cycle-level early warning device, processing equipment, and its corresponding units described above can be found by referring to, for example... Figure 1 The description of the blasting cycle-level early warning method for rockburst disasters in deep-buried tunnels in the corresponding embodiment will not be repeated here.

[0087] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0088] Therefore, this application provides a computer-readable storage medium storing a plurality of instructions that can be loaded by a processor to execute the present application. Figure 1 The steps of the blasting cycle-based early warning method for rockburst disasters in deep-buried tunnels in the corresponding embodiment can be found in the following example. Figure 1 The description of the blasting cycle-level early warning method for rockburst disasters in deep-buried tunnels in the corresponding embodiments will not be repeated here.

[0089] The computer-readable storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0090] Because of the instructions stored in the computer-readable storage medium, the present application can be executed as described above. Figure 1 The steps of the blasting cycle-based early warning method for rockburst disasters in deep-buried tunnels in the corresponding embodiments can be implemented as described in this application. Figure 1 The beneficial effects that the blasting cycle-level early warning method for rockburst disasters in deep-buried tunnels can achieve in the corresponding embodiments are detailed in the preceding description and will not be repeated here.

[0091] The above provides a detailed description of the method, device, processing equipment, and computer-readable storage medium for early warning of rockburst disasters in deep-buried tunnels. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the core ideas of this application. At the same time, those skilled in the art will recognize that there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for early warning of rockburst disasters in deep-buried tunnels through blasting cycles, characterized in that, The method includes: For deeply buried tunnels, the first microseismic monitoring data within a historical time period is obtained as the initial sample; The first microseismic monitoring data is divided into multiple sub-microseismic monitoring data by using three data intervals formed by four consecutive rockburst occurrence times as a data division unit, forming the second microseismic monitoring data. The stage in which the three data intervals are located is regarded as three rockburst disaster incubation stages, and the stage in which the last rockburst occurrence time in the four consecutive rockburst occurrence times is regarded as the rockburst disaster occurrence stage. The four consecutive rockburst occurrence times are marked with corresponding rockburst occurrence levels. Based on the training samples configured from the second microseismic monitoring data, a blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels is trained. The blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels is a deep learning model. The blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels is used to predict the stage of rockburst disaster occurrence and the corresponding rockburst occurrence level based on the microseismic monitoring data within the continuous time period input to the model, combined with the microseismic monitoring data of the three data intervals.

2. The method according to claim 1, characterized in that, The model architecture of the deep-buried tunnel rockburst disaster blasting cycle early warning model specifically adopts the Transformer model.

3. The method according to claim 1, characterized in that, Before training the blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels using the training samples configured based on the second microseismic monitoring data, the method further includes: The second microseismic monitoring data was sample augmentation to balance the number of samples from different rockburst case categories.

4. The method according to claim 1, characterized in that, The sample augmentation of the second microseismic monitoring data includes: The second microseismic monitoring data is augmented using a pre-configured GAN model.

5. The method according to claim 1, characterized in that, The method further includes: For the second microseismic monitoring data, the characteristic parameters corresponding to each stage are calculated, wherein the characteristic parameters include the cumulative logarithm of microseismic energy release, the cumulative apparent volume, the cumulative number of events, and the logarithm of energy density; The deep-buried tunnel rockburst disaster blasting cycle level early warning model is specifically based on the prediction results of the characteristic parameters of the rockburst disaster occurrence stage to predict the corresponding rockburst occurrence level.

6. The method according to claim 1, characterized in that, After training the deep-buried tunnel rockburst disaster blasting cycle early warning model using the training samples configured based on the second microseismic monitoring data, the method further includes: Obtain the target microseismic monitoring data that is currently needed for prediction; The target microseismic monitoring data is input into the deep-buried tunnel rockburst disaster blasting cycle early warning model for prediction processing; Extract the prediction results output by the blasting cycle-level early warning model for rockburst disasters in deep-buried tunnels.

7. The method according to claim 6, characterized in that, After extracting the prediction results output by the deep-buried tunnel rockburst disaster blasting cycle early warning model, the method further includes: The rockburst disaster monitoring system will proceed with response processing based on the predicted results.

8. A blasting cycle-level early warning device for rockburst disasters in deep-buried tunnels, characterized in that, The device includes: The acquisition unit is used to acquire the first microseismic monitoring data within a historical time period for deep-buried tunnels, as an initial sample; The division unit is used to divide the first microseismic monitoring data into multiple groups of sub-microseismic monitoring data by using three data intervals formed by four consecutive rockburst occurrence times as a data division unit, forming the second microseismic monitoring data. The stage in which the three data intervals are located is regarded as three rockburst disaster incubation stages, and the stage in which the last rockburst occurrence time in the four consecutive rockburst occurrence times is regarded as the rockburst disaster occurrence stage. The four consecutive rockburst occurrence times are identified with corresponding rockburst occurrence levels. The training unit is used to train a deep-buried tunnel rockburst disaster blasting cycle level early warning model based on training samples configured from the second microseismic monitoring data. The deep-buried tunnel rockburst disaster blasting cycle level early warning model is a deep learning model. The deep-buried tunnel rockburst disaster blasting cycle level early warning model is used to predict the rockburst disaster occurrence stage and the corresponding rockburst occurrence level based on the microseismic monitoring data within the continuous time period input to the model, combined with the microseismic monitoring data of the three data intervals.

9. A processing device, characterized in that, The method includes a processor and a memory, wherein the memory stores a computer program, and the processor executes the method as described in any one of claims 1 to 7 when it invokes the computer program in the memory.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a plurality of instructions adapted for loading by a processor to perform the method of any one of claims 1 to 7.

Images