Rockburst Intensity Research Articles

Rockburst intensity evaluation by a novel systematic and evolved approach: machine learning booster and application

Rockburst intensity evaluation by a novel systematic and evolved approach: machine learning booster and application

Rockburst Prediction From the Perspective of Energy Release: A Case Study of a Diversion Tunnel at Jinping II Hydropower Station

As a man-made engineering hazard, it is widely accepted that the rockbursts are the result of energy release. Previous studies have examined the unloading of in-situ stress resulting from deep tunnel excavation as a quasi-static process but the transient stress variation during excavation has received less attention. This research discusses rockbursts that happened during the construction of a diversion tunnel at Jinping II hydropower station. The brittle-ductile-plastic (BDP) transition property of Jinping marble was numerically described by the Hoek-Brown strength criterion, and the dynamic energy release process derived from the transient unloading of in-situ stress was studied using an index, local energy release rate. Studies have shown that, due to transient unloading, the strain energy of the surrounding rock mass goes through a dynamic process of decreasing at first, increasing second, then reducing before finally stabilizing. The first decrease of strain energy results from elastic unloading waves and does not cause brittle failure in rock masses, which is consistent with the elastic condition but the secondary reduction of strain energy is because the accumulated strain energy in rock masses exceeds the storage limit, which will inevitably trigger the brittle failure in the rock mass. Thus, the shorter the distance to the tunnel wall the bigger and more intense the energy release. Finally, a relationship between the average value of the local energy release rate and the rockburst intensity was established to assess the risk of rockburst induced by the blasting excavation of a deep tunnel.

Real-time prediction of rockburst intensity using an integrated CNN-Adam-BO algorithm based on microseismic data and its engineering application

Rockburst is a dynamic geological disaster common during underground excavation, which significantly threatens the safety of personnel, equipment, and property. This paper proposes an integrated CNN-Adam-BO algorithm based on microseismic monitoring data to achieve real-time prediction of rockburst intensity. A total of 400 rockburst cases were collected from a water conveyance tunnel in northwestern China, including 100 no rockbursts, 252 low rockbursts, and 48 moderate rockbursts. During modeling, adaptive moment estimation optimization and Bayes optimization were separately used to tune the CNN network hyper-parameters (i.e., convolution kernels and bias) as well as dropout probability, and a parametric analysis was carried out to optimize the CNN structure and training parameters (i.e., the CNN architecture, the number of kernels in the convolutional layer, the number of neurons in the full connection layer, and the number of training epochs). To assess model performance, 2 global metrics (i.e., accuracy and Kappa) and 3 within-class metrics (i.e., precision, recall, and F1-score) were introduced. Additionally, a comparison analysis with other machine learning models, such as a back-propagation neural network, a support vector machine, and two CNN-based benchmark models, was conducted in regard to prediction ability and running speed. Results show that the CNN-Adam-BO model is the most superior, which achieved an accuracy of 91.67%, Kappa of 0.8392, F1-score of 0.8750, 0.9474, and 0.8333 (on no/low/moderate rockburst), and running time of 190.12 s. Taking 2 extra rockburst cases as examples, a complete workflow for how to apply the constructed model to construct a field rockburst risk warning system was illustrated. Further, the effect of data quality, especially data imbalance, on model performance was discussed, indicating that the synthetic minority oversampling technique is an effective way to eliminate data imbalance. Finally, superiority of the dimension enhancement method of the microseismic sequence described in this paper was verified via comparison with the continuous wavelet transform and cross wavelet transform.

A novel tree-based algorithm for real-time prediction of rockburst risk using field microseismic monitoring

Rockburst is a kind of complex and catastrophic dynamic geological disaster in the development and utilization of underground space, which seriously threatens the safety of personnel and environment. Due to the suddenness in time and randomness in space, the prediction of rockburst becomes a great challenge. Microseismic monitoring is capable to continuously capture rock microfracture signals in real time, which offers an effective means for rockburst prediction. With the explosive growth of monitoring data, the conventional manual forecasting methods are laborious and time-consuming. Therefore, artificial intelligence was introduced to improve prediction efficiency. A novel tree-based algorithm was proposed. Its basic idea was to automatically recognize precursory microseismic sequences for the real-time prediction of rockburst intensity. The database consisting of 1500 microseismic events was analyzed. To establish precursory microseismic sequences, dimensionality reduction of the database was first implemented by t-SNE algorithm. Then, k-means clustering algorithm was employed for labelling 1500 microseismic events. Before that, canopy algorithm was adopted to determine the number of clusters. Finally, 300 precursory microseismic sequences were formed by the grouping rule. They were further partitioned into two parts through stratified sampling: 70% for training and 30% for validation. The validation results indicated that the precursor tree with pruning achieved a higher prediction accuracy of 98.9% than one without pruning on the validation set. And the increase was separately 12.2%, 9.2% and 28.6% on the whole validation set and each classes (low/moderate rockburst). In comparison with low rockburst, moderate rockburst was minority class. The improved accuracy on moderate rockburst suggested that pruning can enhance the recognition ability of precursor tree for the minority class. Additionally, two extra rockburst cases were collected from a diversion tunnel in northwestern China, which provided a complete workflow about how to apply the built precursor tree model to achieve field rockburst warning in engineering practice. The tree-based algorithm served as a new and promising way for the real-time rockburst prediction, which successfully integrated field microseismic monitoring and artificial intelligence.

Study on Evolution Mechanism of Structure-Type Rockburst: Insights from Discrete Element Modeling

Taking the “11.28” rockburst occurred in the Jinping II Hydropower Station as the engineering background, the evolution mechanism of structure-type rockburst was studied in detail based on the particle flow code. The results indicate that the failure mechanism of structure-type rockburst includes a tensile fracture induced by tangential compressive stress and a shear fracture caused by shear stress due to overburdened loadings and shear slip on the structural plane. In addition, it is found that the differences between structure-type rockburst and strainburst mainly include (a) the distribution of the local concentrated stress zone after excavation, (b) the evolution mechanism, and (c) the failure locations. Finally, the influence of four factors on the structure-type rockburst are explored. The results show that (1) when the friction coefficient is greater than 0.5, the effect of structural plane is weakened, and the rock near excavation tends to be intact, the structural-type rockburst intensity decreases; (2) the dissipated and radiated energy in structural-type rockburst reduces with rockmass heterogeneity m; (3) the lateral pressure coefficient has a significant effect on the intensity of deep rock failure, specifically in the form of the rapid growth in dissipative energy; (4) and the structural-type rockburst is more pronounced at a structural plane length near 90 mm.

Rockburst Prediction Based on the KPCA-APSO-SVM Model and Its Engineering Application

The progress of construction and safe production in mining, water conservancy, tunnels, and other types of deep underground engineering is seriously affected by rockburst disasters. This makes it essential to accurately predict rockburst intensity. In this paper, the ratio of maximum tangential stress of surrounding rock to rock uniaxial compressive strength (σθ/σc), the ratio of rock uniaxial compressive strength to rock uniaxial tensile strength (σc/σt), and the elastic energy index of rock (Wet) were chosen as input indices, and rockbursts were graded as level I (none rockburst), level II (light rockburst), level III (medium rockburst), and level IV (strong rockburst). A total of 104 groups of rockburst engineering samples, collected widely from around the world, were divided into a training set (84 groups of samples) and a test set (20 groups of samples). Based on the kernel principal component analysis (KPCA), the adaptive particle swarm optimization (APSO) algorithm, and the support vector machine (SVM), the KPCA-APSO-SVM model was established. The proposed model showed satisfactory classification performance: the prediction accuracies of the training set and test set were 98.81% and 95%, respectively. In addition, the trained prediction model was applied to five rockburst engineering cases and compared with the BP neural network model, SVM model, and APSO-SVM model. The comparative results show that the KPCA-APSO-SVM model has a higher prediction accuracy; as such, it provides a new reliable method for rockburst prediction.

Numerical Simulation on Energy Concentration and Release Process of Strain Rockburst

Rockburst mechanism has been a hot topic in the stability analysis of underground carven excavation, and the accurate description of energy evolution process is very critical to rockburst prediction. To study the evolution process of rockburst, such as V-shaped rockburst pit, theoretical formula derivation and numerical simulation are adopted to research the dynamic response characteristics during the formation process of rockburst pits quantitatively. The results show that rockburst intensity distribution varies with failure depth. It can be divided into three zone: slow-increase, rapid-increase and slow-decrease. For a circular tunnel with radius R, the strain energy release rate and vibration response of surrounding rock increases gradually within (0–0.06) R; reaches the peak value around (0.06–0.1) R and drops to a balance beyond 0.1R. Due to the same law of them, the rockburst risk can be conveniently predicted by monitoring vibration of surrounding rock with a certain depth. This work is beneficial to provide a good reference for rockburst prediction.

Mechanism of Structural–Slip Rockbursts in Civil Tunnels: An Experimental Investigation

Structural–slip rockbursts are frequently faced in civil tunnels, which pose huge threats to the safety of personnel and equipment in tunnels. In this study, the core features of structural–slip rockbursts were summarized qualitatively based on the many structural–slip rockburst cases encountered in civil tunnels. Direct shear tests on splitting granite joints were performed to reproduce structural–slip rockbursts in the laboratory, and stress–deformation relationship, sound waves, acoustic emissions (AEs), ejected fragments and failure characteristics of the structural planes were investigated to comprehensively make clear the mechanisms of structural–slip rockbursts. Two generation modes of structural–slip rockbursts were revealed in the experiments. One occurs in the early stage of structural plane slip and shows stress drops resulted from the brittle rupture of asperities on structural surfaces, and the other is that a part of ruptured asperities are gradually pulverized into powder shapes and the powdery debris impels the movement mode of structural planes suddenly from sliding friction to rolling friction, which lead to sudden stress drops and cause the rockbursts. Further, the normal stress effect on the rockburst intensity was discussed. It was found that the larger the normal stress, the greater the intensity of rockbursts. Moreover, the characteristics of rockburst intensities in the tests were analyzed by means of the sound waveforms, and it was induced that the rockburst intensity presents alternating circulation laws. The findings from this study may be valuable for understanding the structural–slip mechanism in civil tunnels.

Fractal Analysis of Fragmentation Distribution of Rockbursts Induced by Low-Frequency Seismic Disturbances

Low-frequency seismic disturbances frequently induce violent rockburst hazards, seriously threatening the safety of deep excavation and mining engineering. To investigate the characteristics and mechanisms of rockbursts induced by seismic disturbances, in this study a series of true triaxial experiments, including the moderate seismically induced, the weak seismically induced, and the self-initiated rockburst experiments under different conditions were conducted. The fractal geometry theory was applied to study rockbursts and the fractal dimensions of fragmentation distribution of different types of rockbursts were calculated. The results show that the fragmentation distributions of both the seismically induced and self-initiated rockbursts exhibit fractal behaviors. For the moderate seismically induced rockbursts, as the static stresses (i.e., the maximum and minimum static stresses) and disturbance amplitude increase, the fractal dimension increases, whereas, as the disturbance frequency increases, the fractal dimension decreases first and then increases. Under similar static loading conditions, the moderate seismically induced rockbursts have the largest fractal dimension, followed by the self-initiated rockbursts, and the weak seismically induced rockbursts have the smallest fractal dimension. There is a linear relationship between the average fractal dimension and kinetic energy of these rockbursts, implying that the fractal dimension can serve as an indicator for estimating rockburst intensity. Furthermore, from a fractal point of view, the energy input, dissipation, and release of these rockbursts are all linear processes.

A Connection Cloud Model Coupled With Improved Conflict Evidence Fusion Method for Prediction of Rockburst Intensity

As a common geological hazard in underground engineering, rockburst has many uncertain factors and interactions, and is of a complicated mechanism. Based on the information fusion theory, a methodology using a novel connection cloud model, which can overcome the defect of the traditional cloud model requiring factors to be normal distribution when determining the basic probability assignment function of the evidence is proposed for predicting the rockburst intensity. And concepts of Lance-distance and Deng-entropy are also introduced here to improve the traditional evidence fusion algorithm. The practical application shows that the result obtained from the proposed model is consistent with that in the actual site testing, so this model is feasible and effective for the prediction of the rockburst intensity. Furthermore, compared with results from other six improved evidence fusion rules, the improved evidence fusion method proposed here has higher fusion convergence precision, faster convergence rate, and more accurate and reliable results, this may provide a new reference for other similar uncertain problems.

Fuzzy Mathematics Comprehensive Forecasting Analysis of Metal Mine Rockburst Based on Multiple Criteriahere

The underground environment in deep mines is complicated and the risk of rockburst is high during excavation. Therefore, it is of great significance to study the tendency of rockburst during excavation. At present, scholars from various countries have analyzed rockburst phenomena from the aspects of strength, stiffness, energy, etc., and have proposed various assumptions and criteria. However, these assumptions and criteria often only consider individual factors, which will lead to one-sidedness and limitations. This article takes the deep mining process of a metal mine as the research background. Based on the results of each independent criterion, fuzzy mathematical theory is introduced to evaluate the rockburst intensity and fuzzy comprehensiveness of various types of rocks in the project. The research results show that the multi-criteria comprehensive evaluation of fuzzy mathematics can well integrate the analysis results of each criterion, and has higher consistency with the actual situation in engineering.

Mechanism of Vibration Energy Action on Dynamic Instability of Shock-Type Rockburst Carrier System

In coal mining, the rockburst intensity triggered by the mine earthquake is often greater in the mining area, and the precursor information is more difficult to capture. In this paper, this kind of rockburst is called shock-type rockburst and studied as the research object. Based on the evolution process and the theoretical model of the carrier system, the energy criterion of the occurrence of shock-type rockburst was deduced based on the theories of mathematics and statistical physics. The results showed that the process of shock-type rockburst can go through four stages: inoculation stage, occurrence stage, development stage, and termination stage, and the secondary triggering phenomenon may occur in the occurrence stage. The theoretical model of the shock-type rockburst carrier system was constructed, and the key characteristic parameter of shock-type rockburst was put forward. Then, the energy of mine earthquake was quantitatively characterized in the form of kinetic energy of rock block. Through the combination of the two parameters, the energy criterion of shock-type rockburst was obtained as follows: once the energy of mine earthquake acting on the rockburst source area was 2%–9% of the static energy, shock-type rockburst can occur. Then, this criterion was verified by three typical cases. Finally, based on the distribution conditions of “red seam” overburden in the Yanzhou mining area of China, the secondary triggering principle of rockburst was expounded, that is, the secondary triggering principle of rockburst caused by the shear fracture and sliding settlement of “red seam” overburden. To sum up, a theoretical system for judging the instability of the shock-type rockburst carrier system was formed. This study provides a theoretical basis for earthquake prevention and disaster reduction in coal mining.

Experimental investigation on the effects of thermal treatment on the physical and mechanical properties of shale

Shale is affected by high temperature and loads in deep mining, shale gas extraction, underground disposal of nuclear waste, and other shale projects. This paper aims to investigate the effects of thermal treatment on the physical and mechanical properties of shale. The uniaxial compression test and acoustic emission (AE) monitoring were conducted on shale specimens with different bedding plane inclinations (0°, 30°, 45°, 60°, and 90°) after thermal treatment of different temperatures. The results show that the P-wave velocity decreases with the increase in bedding plane inclination at each thermal treatment temperature showing anisotropy. A thermal explosion occurs when the specimen reaches the macroscopic thermal damage threshold temperature, and the threshold temperature and intensity of the thermal explosion are linked to the bedding plane inclination. The peak stress of specimens with the bedding plane inclinations of 0°, 45°, and 90° increases with the rise in thermal treatment temperature. After thermal treatment, the relationship curve between the cumulative AE counts of specimens with different bedding plane inclinations and time presents a step-jump trend. The thermal treatment has no significant effect on the failure modes of specimens with bedding plane inclinations of 30°, 60°, and 90°, while specimens with bedding plane inclinations of 0° and 45° begin to splitting failure with a rise in the thermal treatment temperature. The rockburst failure occurs in the shale specimens, and the main phenomena are spalling, particle ejection, and block collapse. The thermal treatment has a significant influence on the rockburst intensity of specimens with different bedding plane inclinations.

Strength of ensemble learning in multiclass classification of rockburst intensity

SummaryRockbust is a violent expulsion of rock due to the extreme release of strain energy stored in surrounding rock mass, leading to considerable damages to underground strucures and equipment, and threatening workers' safety. As the operational depth of engineering projects increases, a larger number of factors influence the mechanism of rockburst. Therefore, accurate classification of rockburst intensity cannot be achieved based on conventional criteria. It is urgent to develop new models with high accuracy and ease to implement in practice. This study proposed an ensemble machine learning method by aggregating seven individual classifiers including back propagation neural network, support vector machine, decision tree, k‐nearest neighbours, logistic regression, multiple linear regression and Naïve Bayes. In addition, we proposed nine data imputation methods to replace the missing values in the compiled database including 188 rockburst instances. Five‐fold cross validation and the beetle antennae search algorithm are used to tune hyperparameters and voting weights of the individual classifiers. The results show that the rockburst classification accuracy obtained by the classifier ensemble has increased by 15.4% compared with the best individual classifier on the test set. The predictor importance obtained by the classifier ensemble shows that the elastic energy index is the most sensitive input variable for rockburst intensity classification. This robust ensemble method can be extended to solve other classification problems in underground engineering projects.

Rockburst prediction model based on comprehensive weight and extension methods and its engineering application

In view of the dynamic instability of rock mass in high geostress areas during underground engineering excavation, the comprehensive weight and extension methods are adopted to research the rockburst prediction. Firstly, five main influencing factors including uniaxial compressive strength, stress coefficient, brittleness coefficient, elastic energy index, and integrity of rock mass are used as the evaluation indexes of rockburst prediction according to the conditions required for rockburst occurrence. The assessment index system of rockburst intensity is constituted. Secondly, the analytic hierarchy process (AHP) and variation coefficient methods are used to determine the comprehensive weight of evaluation index, and the rockburst prediction model is established based on the extension evaluation method. Thirdly, the parameter programming and numerical calculation of the proposed prediction model are carried out in the MATLAB software. The user visualization execution window and software system of rockburst prediction model are realized. Finally, the software system is applied to the rockburst prediction in the water diversion tunnels of Jiangbian hydropower station and Jinping secondary hydropower station. The prediction results are compared with the actual situation and other evaluation methods. The results show that (i) the establishment of the user visualization window realizes the visualization and systematization of rockburst prediction model, which improves the data import rate and calculation efficiency. (ii) The prediction results of the proposed software system agree well with the actual situation, and they are more accurate than other evaluation methods. (iii) The proposed software system of rockburst prediction can also be used in coal mine, metro, and other underground projects, which has good engineering application values.

A Microseismicity-Based Method of Rockburst Intensity Warning in Deep Tunnels in the Initial Period of Microseismic Monitoring

Rockburst disasters in deep tunnels cause serious casualties and economic losses. It is a great challenge to make a warning for rockbursts in geotechnical engineering. In this work, a microseismicity-based rockburst intensity warning method is proposed that is suitable for use in deep tunnels in the initial period of microseismic (MS) monitoring. The method first involves collecting information on a sample of no more than five cases. Then, the event to be analyzed is combined with the sample events and subjected to cluster analysis. Finally, a rockburst intensity warning is generated according to the results of the cluster analysis or after a second cluster analysis. It is a comprehensive, multi-parameter rockburst intensity warning method that only needs a few rockburst cases for input which makes it suitable in the initial period of MS monitoring. The method also incorporates the novel idea of a second cluster analysis. An engineering application based on deep tunnels in the Jinping II hydropower station in Sichuan Province, China, shows that the rockburst intensity warning results based on the proposed method agree well with the actual situations in four tests carried out. The method will enrich the techniques used to warn of rockbursts based on microseismicity.

Strength-weakening effect and shear-tension failure mode transformation mechanism of rockburst for fine-grained granite under triaxial unloading compression

The occurrence of rockburst is closely related to initial high stress state and excavation unloading effects. To investigate the effects of three-dimensional (3D) stress state and unloading rate on the rockburst mechanism, triaxial unloading compression tests were performed on cubic fine-grained granite specimens (50 mm × 50 mm × 50 mm). Two unloading rates (0.08 and 40 MPa/s) were employed in the tests with seven initial 3D confining pressures (5, 10, 20, 30, 40, 50, and 60 MPa). For comparison, triaxial and biaxial compression tests were conducted at the same initial confining pressures. The results show that, under the same confining pressures, the variations in peak strength of granite specimens are in the following order: triaxial compression test > biaxial compression test > triaxial unloading compression test under low unloading rate > triaxial unloading compression test under high unloading rate, which indicating that unloading induces an obvious strength-weakening effect on fine-grained granite. In the triaxial unloading compression test with the same unloading rate, a higher confining pressure results in a greater rock bearing strength and stress drop. More energy will be stored in the rock, and the rockburst intensity will be more serious when the rock is destroyed. When the confining pressure is constant, the lower unloading rate is beneficial for improving the rock bearing strength, and more elastic energy will be accumulated and stored in the rock. Once rockburst occurs, the energy released by rock is larger and the degree of rockburst is stronger. The failure mode of fine-grained granite specimens in triaxial unloading compression tests transforms from shear-tension failure to tension failure with increasing confining pressure, which is the main reason for the weakening of rock strength. Establishing a strength criterion for 3D rock considering the unloading effect is a key for understanding the mechanism of rockburst.

A rockburst intensity criterion based on the Geological Strength Index, experiences learned from a deep tunnel

The occurrence of rockburst is related to the surrounding rock stress and geological rock mass strength. Rockburst occurs frequently during the construction at a deep tunnel in southwest China (a tunnel with a depth of approximately 2080 m, and a length of 13 km), leading to casualties, damage to construction equipment, and schedule delays. The application of the rockburst criteria based on the stress-strength/strength-stress ratio shows that they are not suitable for the construction stage. This paper suggests a simple and feasible rockburst intensity criterion based on the Geological Strength Index (GSI) for the construction stage. In this criterion, the strength of the rock mass exposed during excavation is estimated by the Hoek-Brown criterion and is used to reflect the geological conditions of the surrounding rock, considering the influences of the joint conditions and engineering disturbance. The maximum principal stress along the tunnel is obtained by the in situ stress inversion. The ratio of the rock mass strength to the maximum principal stress (σcm/σ1) is taken as the evaluation index. The criterion is as follows: σcm/σ1 > 1.2, none rockburst; 0.75 < σcm/σ1 ≤ 1.2, slight rockburst; 0.4 < σcm/σ1 ≤ 0.75, moderate rockburst; and σcm/σ1 ≤ 0.4, intense rockburst. The engineering applications show the reliability of this rockburst intensity criterion. The results of this paper provide a reference for rockburst intensity evaluation during the construction stage of deep tunnels.

Rock burst criterion based on clay mineral content

This study aims to establish a criterion for rock burst based on clay mineral content (CMC). A classification system for rock burst intensity is proposed based on the results of 124 laboratory tests. A CMC-based criterion for the probability and intensity of a rock burst is subsequently developed through a statistical analysis of the experimental results. Rock burst in laboratory tests can be classified as not occurring, weak, moderate, or strong according to sound, associated phenomena, fragment shape, and failure modes. When the CMC ≤ 10%, the probability of a rock burst is high, and the rock burst intensity is from moderate to strong. If 10% 30%, the probability of a rock burst is low. This rock burst criterion is used to evaluate the probability of rock burst at the Yaoqiao and Kongzhuang coal mines, China to demonstrate the feasibility of this criterion.

Predicting rockburst with database using particle swarm optimization and extreme learning machine

Rockburst is a major type of geological hazard that has a very adverse impact on underground engineering in deeply buried areas under high geo-stress. In this study, extreme learning machine (ELM) was used to predict and classify rockburst intensity, and particle swarm optimization (PSO) was used to optimize the input weight matrix and the hidden layer bias in ELM. Six quantitative rockburst parameters were used as input for the PSO-ELM network, including the maximum tangential stress of the surrounding rock σθ, the uniaxial compressive strength of rock σc, the tensile strength of rock σt, the stress ratio σθ/σc, the rock brittleness ratio σc/σt and the elastic energy index Wet. The network was used to learn from a database of 344 collected worldwide rockburst cases, on which the PSO-ELM rockburst prediction model was established and verified using 8-fold cross-validation and independent test set validation. The model was then tested on a new set of fifteen typical rockburst cases from Jiangbian hydropower station in China. The results showed that the PSO-ELM model performed well in rockburst level prediction. In addition, the model showed superior performance compared with previously proposed machine learning models for rockburst prediction and the rockburst empirical criteria, which underscores its utility in future rockburst prediction.