Mine time-delay rockburst dynamic short-term and temporary intelligent forecasting system

Abstract

The invention discloses a dynamic short-term and temporary intelligent forecasting system for mine time-delay rockburst, and belongs to the field of short-term and temporary forecasting of rockburst disasters. The system comprises a deep well three-dimensional micro-seismic field global sensing module, a micro-seismic characteristic parameter extraction module, a rockburst catastrophe evolution data set construction module, a data preprocessing module, a deep learning prediction module, an interpretability analysis module and a dynamic short-term and temporary intelligent forecasting module. A dynamic tracking type micro-seismic sensing network is constructed through a three-level annular nested layout strategy, micro-seismic event data are collected in real time, spatial-temporal characteristic parameters are extracted, a rockburst time sequence data set is constructed in combination with a sliding time window method, and the occurrence probability and intensity level of rockburst are predicted. The system adopts a grey wolf-simulated annealing hybrid algorithm to optimize model parameters, and improves the credibility of a prediction result through SHAP value analysis. The system can realize dynamic short-term and temporary early warning of rockburst disasters, and provides scientific decision support for mine safety management.

Description

Technical Field

[0001] The present invention belongs to the field of short-term and impending rockburst disaster prediction, and more particularly relates to a dynamic short-term and impending intelligent prediction system for time-lag type rockburst in mines. Background Art

[0002] Rockbursts in deep metal mines are a common and significant safety hazard in underground engineering projects. Their sudden and destructive nature poses a serious threat to personnel safety and engineering facilities. Short-term prediction of rockbursts in mines remains a challenge in the international mining industry. This is particularly true in deep mine environments, where complex geological conditions and high stress environments make rockburst prediction research both more challenging and more urgent.

[0003] Rockburst disasters are often accompanied by microseismic events. Microseismic monitoring technology can acquire massive amounts of microseismic time-series data, including parameters such as the source location, released energy, moment magnitude, and stress drop. The temporal evolution of microseismic data contains a wealth of information about rockburst precursors. Deeply exploring its spatiotemporal evolutionary characteristics can aid in rockburst prediction and forecasting. However, when acquiring microseismic monitoring data in real time, manual analysis is difficult to rapidly extract the patterns of rockbursts from this massive amount of data and accurately predict rockbursts. Therefore, a deep learning algorithm is used to automatically extract rockburst characteristics from real-time microseismic data, enabling rapid and accurate prediction of rockbursts.

[0004] Existing intelligent prediction methods for metal mine rockbursts based on microseismic data can only perform post-disaster statistical analysis on microseismic data obtained after a rockburst event occurs, ignoring the evolution of microseismic data during the rockburst disaster process. This makes it difficult to provide dynamic forecasts and early warnings for the disaster during real-time monitoring of the entire rockburst process at deep well engineering sites. Therefore, how to rationally construct a real-time microseismic monitoring dataset with indicative warning labels such as deep well rockburst intensity classification or occurrence probability, fully utilize real-time microseismic data to build and train an intelligent model that can dynamically predict and accurately predict rockburst disasters in real-time monitoring scenarios; after obtaining the model and evaluating good predictions, combine statistical machine learning methods to perform feature analysis and explain the black box structure of the deep learning model; and finally, construct a more efficient, transparent, and more generalizable real-time dynamic, short-term intelligent prediction method for deep metal mine rockbursts based on real-time microseismic monitoring time series data. This method, which meets the urgent needs for real-time monitoring, dynamic prediction, and short-term prediction of deep metal mine rockbursts, has become a technical challenge that needs to be solved urgently. Summary of the Invention

[0005] The present invention provides a time-delay type rockburst dynamic short-term and imminent intelligent prediction system, device and equipment based on microseismic real-time monitoring data, which relates to the technical field of mine rockburst monitoring and early warning. The present invention includes the following steps: arranging microseismic real-time monitoring at the site of underground engineering in metal mines, obtaining microseismic real-time data, and constructing a dynamic time series dataset of rockburst disasters through a time dimension data preprocessing method, and dividing the dataset into a training set and a test set; verifying and optimizing an intelligent prediction model on the training dataset to form a short-term and imminent intelligent prediction method for rockbursts in deep metal mines that can dynamically achieve accurate prediction.

[0006] In order to achieve the above object, the present invention is implemented by adopting the following technical solutions:

[0007] The system described above includes the following modules:

[0008] A deep well three-dimensional microseismic field global perception module, based on the distribution characteristics of mine roadways and deep rock mass mechanical parameters, constructs a dynamic tracking microseismic sensor network topology structure by adopting a three-level circular nested layout strategy, in which high-density three-component intelligent microseismic sensors are deployed in the key monitoring area, an adaptive azimuth calibration array is configured in the transition area, and broadband microseismic event capture nodes are arranged in the peripheral area. Each node integrates an edge computing module for collecting and filtering the original microseismic waveform signal;

[0009] A microseismic characteristic parameter extraction module, by collecting microseismic signal waveform data in real time, extracts time dimension characteristic parameters and space dimension characteristic parameters, including the number of microseismic events, microseismic event energy, apparent stress, stress drop, microseismic event density, source location, etc.;

[0010] A rockburst disaster evolution dataset construction module, constructs a microseismic spatio-temporal characteristic parameter time series dataset with rockburst intensity grading or occurrence probability labels based on the sliding time window method;

[0011] A data preprocessing module, uses an adaptive logarithmic function to standardize the data and optimize the characteristic space of time series data;

[0012] A deep learning prediction module, uses a deep learning model that couples 3D-TCN and BiGRU with an Attention mechanism, and combines a grey wolf-simulated annealing hybrid algorithm to optimize the model structure parameters to predict the rockburst occurrence probability and intensity level within the next 12 hours;

[0013] An interpretability analysis module, through the transparent analysis of the relationship between the model input and output, improves the credibility of the system prediction logic;

[0014] A dynamic short-term and imminent intelligent prediction module, based on the above modules, realizes real-time dynamic prediction and early warning of rockburst disasters.

[0015] In one solution, the three-level circular nested layout strategy in the deep well three-dimensional microseismic field global perception module includes a key monitoring area, a transition area, and a peripheral area. Among them, the sensor sampling rate in the key monitoring area is ≥10 kHz, the sensor array in the transition area includes 6-degree-of-freedom attitude sensors, and the sampling frequency range of the broadband microseismic event capture nodes in the peripheral area is 1 - 1 kHz.

[0016] In one solution, the time-dimensional characteristic parameters extracted in the microseismic characteristic parameter extraction module include the number of microseismic events, the energy of microseismic events, apparent stress, apparent volume, bulk strain potential, stress drop, and moment magnitude mutation index. The spatial-dimensional characteristic parameters include microseismic event density, event density spatio-temporal fractal dimension, and focal location.

[0017] In one solution, the rockburst disaster evolution dataset construction module constructs a time series dataset through a sliding time window method. The input is 720 timestamp data within the past 12 hours, and the output is the rockburst occurrence probability label and intensity level label within the next 12 hours. The intensity levels are divided into no rockburst, minor rockburst, moderate rockburst, and severe rockburst.

[0018] In one solution, the data preprocessing module uses an adaptive logarithmic function to standardize the data. By adjusting the scaling ratio constant and parameters, it avoids the distortion effect caused by excessive numerical differences and optimizes the data distribution to adapt to the input of the deep learning model.

[0019] In one solution, the deep learning prediction module uses 3D-TCN to extract the spatio-temporal local features of microseismic parameters, and captures the long sequence features in the spatio-temporal dimension through BiGRU combined with the Attention mechanism, and finally outputs the rockburst occurrence probability and intensity level.

[0020] In one solution, the deep learning prediction module optimizes the model structure parameters through a hybrid algorithm of grey wolf - simulated annealing. Among them, the grey wolf algorithm is used for global search of structure parameters, and the simulated annealing mechanism is used for local refinement optimization. The objective function is based on prediction accuracy and model generalization ability.

[0021] In one solution, the interpretability analysis module uses the SHAP value analysis method to reveal the mapping relationship between microseismic characteristic parameters and rockburst prediction results, and improves the transparency and credibility of the deep learning model.

[0022] In one solution, the dynamic short-term intelligent forecasting module combines the real-time monitoring data of the microseismic field with the deep learning prediction results to provide dynamic early warning information for rockburst disasters, including the rockburst occurrence probability and intensity level, and provides decision-making support for mine safety management.

[0023] Advantages of the present invention:

[0024] Online analysis of real-time microseismic monitoring time series data through a deep learning model to achieve dynamic short-term and imminent forecasting of rockburst disasters.

[0025] The model can automatically and efficiently capture the subtle changes in the evolution process of microseismic data, improving the accuracy and reliability of rockburst disaster prediction.

[0026] Combining statistical machine learning methods to conduct a deconstruction analysis of the deep learning model, making the prediction model more transparent, credible, and interpretable.

[0027] The systematic method for obtaining real-time microseismic monitoring data, constructing a dynamic data set for rockburst disasters, and validating, optimizing, and implementing the function of the dynamic short-term and imminent forecasting model for rockbursts is applicable to deep metal mines under different geological conditions and has good generalization ability. Description of the Drawings

[0028] Figure 1 Is the system construction flow chart of the present invention;

[0029] Figure 2 Is the probability data set diagram of the rockburst phenomenon of the present invention;

[0030] Figure 3 Is the classification data set diagram of the rockburst intensity level of the present invention;

[0031] Figure 4 Is the dynamic intelligent short-term and imminent forecasting system diagram of the time-delay type rockburst in deep mines of the present invention. Detailed Implementation Modes

[0032] To facilitate the understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. The typical embodiments of the present invention are shown in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0033] Unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as understood by those skilled in the technical field to which the present invention belongs. The terms used in the description of the present invention in this specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. To facilitate the understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. The typical embodiments of the present invention are shown in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0034] As Figure 1 shown, a dynamic short-term and imminent intelligent forecasting system for time-delay type rockbursts in mines is constructed by the following steps:

[0035] Step 1: Construction of a global perception system for deep well 3D microseismic fields

[0036] Based on the distribution characteristics of mine tunnels and the mechanical parameters of deep rock masses, a reconfigurable sensor array system with spatiotemporal self-correction capabilities was established, forming a dynamic tracking microseismic sensing network topology. A three-level, ring-shaped, nested layout strategy was adopted: high-density, three-component intelligent microseismic sensors (sampling rate ≥ 10kHz) were deployed in key monitoring areas, an adaptive orientation calibration array (including a 6-degree-of-freedom attitude sensor) was configured in transition areas, and broadband microseismic event capture nodes (1-1kHz) were deployed in peripheral areas. Each node integrated an edge computing module to enable raw microseismic waveform signal acquisition and filtering.

[0037] Step 2: Extraction of microseismic characteristic parameters

[0038] Through the tracking arrangement of microseismic sensor arrays during deep well mining, microseismic signal waveform data is collected, transmitted and stored in real time.

[0039] Extract the time series data of characteristic parameters of time dimension from waveform and microseismic event data (including the number of microseismic events, microseismic event energy E, apparent stress σ, apparent volume V, volume potential ε, stress drop Δσ and other parameter accumulation and change rate and moment magnitude M). w Mutation index I).

[0040]

[0041] The time series data of spatial dimension time series characteristic parameters (microseismic event density ρ, event density spatiotemporal fractal dimension D, microseismic event source location (x, y, z)) are extracted from waveform and microseismic event data.

[0042]

[0043] r is the relative size of the microseismic event space at time t.

[0044] Step 3: Construction of rockburst disaster evolution dataset

[0045] According to the evolution process of rock burst disaster, a time series dataset of microseismic spatiotemporal characteristic parameters with rock burst intensity classification or occurrence probability labels is constructed.

[0046] ① Through the sliding time window, we construct the input 720 time stamps (12 hours) time series data and output the label of the rock burst phenomenon that occurs within the next 12 hours after the last input time stamp (if it occurs, the label is 1; if it does not occur, the label is 0), as shown in the following example: Figure 2 ;

[0047] ② When the output indicates that a rockburst will occur within 12 hours of the last input timestamp, the label is 1, that is, a rockburst will occur, and the intensity of the rockburst that may occur is marked (0 for no rockburst, 1 for mild rockburst, 2 for moderate rockburst, and 3 for severe rockburst). Figure 3 .

[0048] Step 4: Data preprocessing

[0049] Preprocessing operations (improved Log-Gaussian hybrid mapping normalization) are performed on the microseismic spatiotemporal parameter time series data set;

[0050] x'=log[β(x+ε)],

[0051]

[0052] Among them, x and x' represent the time series data of each parameter before and after preprocessing, β represents the adaptive adjustment parameter of the parameter time series data, α represents the scaling constant, and ε represents the constant to prevent the value from being too small after logarithmic mapping.

[0053] Adaptive logarithmic function is used to process the time series data of each parameter to control the distortion effect on the feature space caused by the excessive difference in the process of numerical increase or decrease, and at the same time, adaptive weight addition is given to the numerical values that are relatively large but controlled.

[0054] Step 5: Deep learning algorithm modeling and training

[0055] A deep learning algorithm model is adopted, which combines a multi-physics spatial-temporal attention network (MSTA-Net) with a multi-modal spatiotemporal feature deep mining module. The deep learning network model uses 3D-TCN as the algorithm base to extract local features of microseismic parameter time series data in the spatiotemporal dimension. The hybrid network of BiGRU and Attention mechanism modules is coupled to perform feature analysis on long sequences in the spatiotemporal dimension. Finally, linear and nonlinear connections are used to output the probability of rockburst occurring within the next 12 hours. When the probability of rockburst occurring within the next 12 hours from the last input timestamp is greater than 50%, the intensity level of the possible rockburst is collaboratively output. Based on this, two dynamic time series data sets of microseismic spatiotemporal dimension feature parameters are modeled and trained.

[0056] Data acquisition layer: Deploy microseismic sensor arrays based on on-site mining and geological conditions to collect real-time microseismic source location (x, y, z), energy E, apparent stress σ, apparent volume V, volume potential ε, moment magnitude M w and parameters such as stress drop Δσ;

[0057] Spatiotemporal feature fusion:

[0058] Time dimension: The time scale sliding method (the minimum time series scale Δt = 10 min) is used to extract the energy release rate ΔE / Δt and the event frequency gradient dN / dt. The time window sliding method (window length T = 6h, 12h, 18h and 24h) is used to obtain cyclic time series data that can be input into the intelligent model.

[0059] Spatial dimension: Divide the monitoring area into three-dimensional grids (side length ≤ 15m) and calculate the microseismic event density ρ per unit volume;

[0060] Warning label generation: Based on the on-site rockburst damage degree, geological conditions, microseismic energy and other parameters, a four-level rockburst intensity warning label is constructed:

[0061] Label Y∈{0: no rockburst, 1: slight rockburst, 2: moderate rockburst, 3: severe rockburst} 2. Coupled prediction model driven by spatiotemporal attention mechanism

[0062] Model architecture: 1D CNN-BiLSTM-Attention coupled network

[0063] Input layer: standardized multidimensional time series data matrix X(t)∈R^(N×M), where N is the time step and M is the feature dimension;

[0064] Feature extraction layer:

[0065] 1D CNN: extract local temporal features (convolution kernel size = 3, number of channels = 64);

[0066] BiLSTM: captures long-sequence spatiotemporal features (hidden layer units = 128);

[0067] Attention layer: Calculate feature weight α i =softmax(q^T tanh(W·h i + b)), focusing on the evolution of key disasters, integrating spatiotemporal characteristics and regressing the extent of rockburst damage;

[0068] Output layer: The Softmax classifier outputs the probability distribution P(Y|X) of rockburst level in the future Δt period.

[0069] Step 6: Model structure parameter optimization

[0070] By analyzing the characteristics of the microseismic spatiotemporal parameter time series data set and comparing it with the biological meta-heuristic algorithm to verify the optimization effect, the gray wolf-simulated annealing hybrid algorithm was selected to optimize the network structure parameters and model hyperparameters. The improved adaptive gray wolf algorithm was used for global search, and the simulated annealing mechanism with memory effect was coupled for local fine tuning. Finally, a deep learning network model with optimal structural parameters was obtained.

[0071] Global search phase for optimizing structural parameters:

[0072]

[0073] W(k) is the adaptive weight, X(k) is the current parameter marker of the algorithm's iterative optimization. α, β, and δ represent three different marker positions. K represents the iterative update convergence factor.

[0074] Local refinement stage of optimizing structural parameters:

[0075]

[0076] P Accepted represents the probability of accepting the updated optimal solution, E(A) and E(B) represent the objective functions of the current solution and the new solution, respectively. T represents the current temperature of the simulated annealing algorithm. λ represents the decay constant.

[0077] At the same time, different from the current research on rockburst intensity prediction methods, which mainly focuses on post-disaster summary analysis and static model prediction and analysis, this step can obtain an intelligent model that can realize real-time dynamic prediction and short-term forecast of rockburst disasters;

[0078] Step 7: Model function deconstruction analysis

[0079] By combining statistical machine learning methods such as SHAP values with deep-well geological information, the dynamic, short-term, and intelligent prediction model for rockburst disasters was functionally deconstructed. Through interpretable analysis, the black-box attribute mapping relationships of the deep learning model were made transparent, transforming it into a gray-box or even white-box model. Ultimately, under real-time microseismic monitoring, interpretable deep learning network models were trained and optimized. This end-to-end model, with input data and dynamic output, allowed the hazard level of rockburst disasters to be assessed, accurately predicting the probability of metal mine rockbursts and the intensity of potential rockburst disasters.

[0080] Step 8: Construction of dynamic short-term rockburst prediction system

[0081] Finally, by linking the global tracking perception system, the rockburst disaster dynamic data set construction and scaling mapping preprocessing technology based on microseismic spatiotemporal feature time series data, and the intelligent prediction model of multimodal spatiotemporal deep feature extraction, a microseismic tracking real-time monitoring-mine time-lag rockburst dynamic intelligent short-term prediction system was obtained, which realized the tracking perception of the spatiotemporal evolution characteristics of the microseismic field in the entire mine, the extraction and analysis of the spatiotemporal features of rockburst disasters, and the dynamic intelligent prediction of time-lag rockbursts, such as Figure 4 .

[0082] The dynamic intelligent short-term and imminent prediction system for rockburst has high-precision dynamic tracking, effectively improving the acquisition efficiency and positioning accuracy compared with the traditional layout method; compared with the traditional intelligent prediction method for rockburst intensity levels, it emphasizes more on the real-time analysis before the disaster of rockburst prediction and forecasting, the process-based prediction of disaster evolution, and the dynamic mode prediction; at the same time, the system can better adapt to the monitoring and early warning work of rockburst dynamic disasters during deep well mining, and can provide a scientific basis and technical paradigm for the perception-warning-prevention system of time-delay rockburst during subsequent deep mining of mines, including steps such as data base, data processing, feature extraction and analysis, and prediction and forecasting.

[0083] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The described program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above various methods. Among them, the storage medium can be a magnetic disk, an optical disc, a read-only memory (ROM), or a random access memory (RAM), etc.

[0084] It should be understood that the detailed description of the technical solutions of the present invention with the aid of the preferred embodiments above is illustrative rather than restrictive. Those of ordinary skill in the art can modify the technical solutions recorded in each embodiment on the basis of reading the specification of the present invention, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present invention.

Claims

1. A dynamic short-term and imminent intelligent prediction system for time-delay rock bursts in mines, characterized in that: The described system includes the following modules: The deep well three-dimensional microseismic field global perception module, based on the mine roadway distribution characteristics and deep rock mass mechanical parameters, constructs a dynamic tracking microseismic sensor network topology using a three-level circular nested layout strategy. Among them, high-density three-component intelligent microseismic sensors are deployed in the key monitoring area, an adaptive azimuth calibration array is configured in the transition area, and broadband microseismic event capture nodes are arranged in the peripheral area. Each node integrates an edge computing module for the acquisition and filtering processing of raw microseismic waveform signals; The microseismic characteristic parameter extraction module extracts time dimension characteristic parameters and space dimension characteristic parameters by real-time collecting microseismic signal waveform data, including the number of microseismic events, microseismic event energy, apparent stress, stress drop, microseismic event density, and source location; The rockburst disaster evolution dataset construction module constructs a time series dataset of microseismic spatio-temporal characteristic parameters with rockburst intensity grading or occurrence probability labels based on the sliding time window method; The data preprocessing module standardizes the data using an adaptive logarithmic function to optimize the characteristic space of time series data; The deep learning prediction module uses a deep learning model that couples 3D-TCN and BiGRU with an Attention mechanism, and combines a grey wolf-simulated annealing hybrid algorithm to optimize the model structure parameters to predict the rockburst occurrence probability and intensity level within the next 12 hours; The interpretability analysis module improves the credibility of the system prediction logic through the transparent analysis of the relationship between model input and output; The dynamic short-term intelligent forecasting module realizes the real-time dynamic prediction and early warning of rockburst disasters based on the above modules.

2. The dynamic short-term and imminent intelligent prediction system for time-delay rock bursts in mines according to claim 1, characterized in that, The three-level circular nested layout strategy in the deep well three-dimensional microseismic field global perception module includes a key monitoring area, a transition area, and a peripheral area. Among them, the sensor sampling rate in the key monitoring area is ≥10kHz, the sensor array in the transition area includes 6-degree-of-freedom attitude sensors, and the sampling frequency range of the broadband microseismic event capture nodes in the peripheral area is 1-1kHz.

3. The dynamic short-term and imminent intelligent prediction system for time-delay rock bursts in mines according to claim 1, wherein, The time dimension characteristic parameters extracted in the microseismic characteristic parameter extraction module include the number of microseismic events, microseismic event energy, apparent stress, apparent volume, bulk strain potential, stress drop, and moment magnitude mutation index. The space dimension characteristic parameters include microseismic event density, event density spatio-temporal fractal dimension, and source location.

4. The dynamic short-term and imminent intelligent prediction system for time-delay rock bursts in mines according to claim 1, wherein, The rockburst disaster evolution dataset construction module constructs a time series dataset through the sliding time window method. The input is 720 timestamp data within the past 12 hours, and the output is the rockburst occurrence probability label and intensity level label within the next 12 hours. The intensity levels are divided into no rockburst, minor rockburst, moderate rockburst, and severe rockburst.

5. The dynamic short-term and imminent intelligent prediction system for mine time-delay rockburst according to claim 1, wherein The data preprocessing module standardizes the data using an adaptive logarithmic function, avoids the distortion effect caused by excessive numerical differences by adjusting the scaling ratio constant and parameters, and optimizes the data distribution to adapt to the input of the deep learning model.

6. The intelligent short-term and imminent dynamic prediction system for time-delay rock bursts in mines according to claim 1, wherein The deep learning prediction module uses 3D-TCN to extract the spatio-temporal local characteristics of microseismic parameters, and captures the long sequence characteristics in the spatio-temporal dimension through BiGRU combined with the Attention mechanism, and finally outputs the rockburst occurrence probability and intensity level.

7. The dynamic short-term and imminent intelligent prediction system for time-delay rock bursts in mines according to claim 1, wherein The deep learning prediction module optimizes the model structure parameters through the Grey Wolf-Simulated Annealing hybrid algorithm, where the Grey Wolf algorithm is used for global search of the structure parameters, and the simulated annealing mechanism is used for local refinement optimization. The objective function is based on prediction accuracy and model generalization ability.

8. The dynamic short-term and imminent intelligent prediction system for time-delay rock bursts in mines according to claim 1, characterized in that The interpretability analysis module adopts the SHAP value analysis method to reveal the mapping relationship between microseismic characteristic parameters and rockburst prediction results, improving the transparency and credibility of the deep learning model.

9. The dynamic short-term and imminent intelligent prediction system for time-delay rockburst in mines according to claim 1, wherein, The dynamic short-term intelligent forecasting module combines the real-time monitoring data of the microseismic field and the deep learning prediction results to provide dynamic early warning information for rockburst disasters, including the probability of rockburst occurrence and the intensity level.

Images