Figure - available from: Neural Computing and Applications
This content is subject to copyright. Terms and conditions apply.
![Schematic representation of rockburst types and the effect of confinement [10]](publication/334687504/figure/fig1/AS:941477777637413@1601477202824/Schematic-representation-of-rockburst-types-and-the-effect-of-confinement-10.png)
Source publication
![A schematic model of self-organizing map network [52]](publication/334687504/figure/fig2/AS:941477777666092@1601477202894/A-schematic-model-of-self-organizing-map-network-52_Q320.jpg)
One of the main concerns associated with deep underground constructions is the violent expulsion of rock induced by unexpected release of strain energy from surrounding rock masses that is known as rockburst. Rockburst hazard causes substantial damages to the foundation of the structure and equipment and can be a menace to the safety of workers. Th...
Similar publications
Advances in measurement technology are producing increasingly time-resolved environmental exposure data. We aim to gain new insights into exposures and their potential health impacts by moving beyond simple summary statistics (e.g., means, maxima) to characterize more detailed features of high-frequency time series data. This study proposes a novel...
Wind generation has recently become an essential renewable power supply option. Wind generators are integrated with electrical machines that require correct functionality. However, the increasing use of non-linear loads introduces undesired disturbances that may compromise the integrity of the electrical machines inside the wind generator. Therefor...
In this study, some physico-chemical and mechanical properties and appropriate moisture levels for workability of soil series in Bilecik Şeyh Edebali University Campus area in Turkey were determined. Soil samples were taken from 0-20 cm depth of 20 different points in Gülümbe and Aşağıköy series. Soils in the Gülümbe were clay (C) and clay loam (CL...
A new multidisciplinary workflow is suggested to recharacterize the Hamra Quartzite (QH) formation using artificial
neural networks. This approach involves core description,
routine core analysis, special core analysis and raw logs of
fourteen wells. An efficient electrofacies clustering neural
network technology based on a self-organizing map is p...
The working principle of hydraulic system is to use the flow and pressure of liquid in the system for energy transfer and conversion. Hydraulic system realizes various work tasks such as pushing, grasping, lifting, rotating, etc. by controlling the action of hydraulic actuators. Its advantages are high power density, good stability, fast response t...
Citations
... Compared to the strength-based, strain-based, and strain energy-based empirical criteria proposed by different researchers since 1970, which have been developed based on limited data and few input parameters and have relatively low prediction accuracy, machine learning (ML)based models have shown considerably higher prediction and classification performance in recent studies [50,57,71]. These studies have used two categories of ML algorithms, supervised and unsupervised. ...
... For the latter, it is assumed that the output (risk level) is unknown, and by performing a clustering analysis, the data clusters are identified, and new labeling is conducted. According to recent comprehensive review studies [61,71], artificial neural networks (ANN) [58,72], support vector machine (SVM) [47,62], logistic regression (LR) [38], decision tree (DT) [19,58], particle swarm optimization (PSO) [67,69], self-organizing map (SOM) [57,74], Bayesian network (BN) [30,39], and their hybridized versions are among the most common ML algorithms used for rockburst assessment. More information regarding the employed supervised and unsupervised ML algorithms in prior studies, defined input and output variables, and the corresponding performance metrics values can be found in Zhou et al. [71] and Wu et al. [61]. ...
... Classification system for rockburst risk level (modified from[57] and[73] ...
The rockburst hazard induced by the extreme release of the stress concentrated in rock mass in deep underground mines poses a significant threat to the safety and economy of the mining projects. Therefore, properly managing this hazard is critical for ensuring rock engineering projects’ sustainability. This study proposes comprehensible and practical classifiers for rockburst risk level appraisal by hybridizing K-means clustering with gene expression programming, GEP, logistic regression, LR, and classification and regression tree, CART (i.e., K-mean-GEP-LR and K-means-CART classifiers). A database containing 246 rockburst events with four risk levels of none, light, moderate, and severe was compiled from previous practices. Preliminary statistical analyses were conducted to detect the extreme outliers and determine the critical rockburst indicators. The K-means clustering analysis was performed to identify the main clusters within the database and relabel the rockburst events. The GEP algorithm was then utilized to develop binary models for predicting the occurrence of each class. Then, the likelihood of each class occurrence was determined using LR. Furthermore, the K-means clustering was combined with the CART algorithm to provide another visual tree structure model. The classifiers’ performance evaluation showed 96% and 95% accuracy values in the training and testing stages, respectively, for the K-means-GEP-LR model, while the accuracy values of 98.8% and 93.0% were obtained for the foregoing stages for the K-means-CART classifier. The results showed the robustness and high classification capability of both models. MatLab codes were also provided for the K-means-GEP-LR model, which assists other researchers/engineers in implementing the model in practice.
... Because of the high ability of artificial intelligence algorithms and multivariate analytical procedures in analyzing complex problems, these techniques have been used to improve the problems in several fields of mining engineering especially rock fragmentation [21][22][23][24][25][26][27][28][29][30][31][32][33][34][35]. Monjezi et al. [36] used Fuzzy logic (FL) for predicting the size of fragmented rocks in Golgohar iron ore open-pit mine. ...
Uniaxial compressive strength is one of the most important mechanical characteristics of rocks, and its prediction is essential in most rock engineering projects. Recently, several types of research have been conducted regarding the relationship between the petrophysical and geomechanical properties of rocks and geophysical parameters, among which geoelectric and seismic methods have been the most utilized and efficient. In this research, the prediction of uniaxial compressive strength (UCS) has been investigated utilizing the geoelectrical method. In the present study, after sampling from the studied area, the core samples were saturated and by installing special electrodes, the changes in electrical resistivity during compressive stress loading were measured and monitored in the laboratory environment. Laboratory samples with various structures and textures indicated different electrical behaviors during the application of compressive stress which stresses the significant effect of the volume proportion of aggregates on petrophysical and geomechanical properties. In core samples prepared from fault breccias and alluvial materials with a volumetric block proportion (VBP) of 25–75%, a significant correlation was observed between the UCS and the measured electrical resistivity values. The neuro-fuzzy model presented with a decision factor (R2) of 99.7%, root sum of square error (RMSE) of 2.511, geometric mean error ratio (GMER) of 0.865, and relative improvement index (RI) of 71.08 can predict uniaxial compressive strength (UCS) more accurately than artificial neural network and multivariable regression methods.
... Deep rock masses, defined as rock masses with depths exceeding 3000 m, present more complex geological origins and occur in unique environments characterized by high geo-tress, high temperature, high pore pressure, and strong mining disturbances. These factors contribute to the frequent and intense occurrence of rockburst in deep rock mass engineering projects, highlighting the critical importance of understanding the factors that contribute to rockbursts for the safety of future deep underground construction projects [11][12][13][14][15][16][17][18]. Such incidents pose threats to personnel, property, and engineering stability. ...
In order to ensure the successful construction and stable operation of deep engineering projects, significant progress has been made in researching deep underground rockburst issues from various perspectives. However, there have been few systematic analyses of the overall research status of deep rockburst to date. In this study, a bibliometric approach using CiteSpace software (version 6.2.R3) was employed to visualize and analyze knowledge maps of 353 research articles on deep rockburst collected from the Web of Science core database from 1996 to 2022. The results show that the number of publications experienced exponential growth after an initial stage of budding and peaked in 2016. In terms of collaboration, China plays an absolute central role. The top three highly cited journals were the International Journal of Rock Mechanics and Mining Sciences, Rock Mechanics and Rock Engineering, and Tunneling and Underground Space Technology. In the keyword co-occurrence analysis, the keyword “prediction” had the highest frequency of occurrence in the past two decades, indicating it as the major research focus in deep rockburst studies. The keyword co-occurrence clustering analysis revealed eight clusters, including conventional criteria, acoustic emission, geology, seismic velocity tomography, dynamic disturbance, and others, representing the primary research topics. This study provides a comprehensive analysis of the current research progress and development trends of deep underground rockburst, helping to understand the key areas of focus in this field and providing potential prospects for future investigations for researchers and practitioners.
... Artificial neural network (ANN) as an artificial intelligence technique is a system developed for information processing, where it works similarly to biological neural systems and it was developed based on the human brain (Taheri et al. 2017;Mohammadi et al. 2018;Dormishi et al. 2019;Mikaeil et al. 2019;Faradonbeh et al. 2020;Noori et al. 2020). This technique is capable of processing information that is complex, nonlinear, and able to work in parallel, distributed, and local processing and adaptation. ...
The determination of uniaxial compressive strength (UCS) is one of the most important mechanical properties of rocks. The direct measurement of UCS using laboratory methods is difficult, costly, and time-consuming. The main purpose of this research is to estimate the UCS of some sedimentary rocks by using multiple regression analysis (MRA), artificial neural network (ANN), and adaptive neuro-fuzzy inference system (ANFIS). Sandstone, limestone, travertine, and conglomerate samples were collected as studied rocks and some geotechnical tests including dry density, porosity, ultrasonic P-wave velocity (UPV), point load strength (PLS), Brazilian tensile strength (BTS), block punch strength (BPS) and uniaxial compressive strength were determined. Results show the average values of γdry, n, UPV, PLS, BTS, BPS, and UCS were obtained 2.42 g/cm³, 5.77%, 4.30 km/s, 7.91, 5.19, 6.37 and 29.70 MPa, respectively. To evaluate the performance of MRA, ANN, and ANFIS models, some statistical coefficients, including R, RMSE, VAF, MAPE, and PI were calculated. Based on the results, the average values of R were obtained 0.82, 0.96, and 0.99 and RMSE were 7.43, 3.74, and 0.50 for the MRA, ANN, and ANFIS models, respectively. Comparing the results demonstrated that the ANFIS models have higher validity and the ANN models are more efficient than MRA. Also, by comparing the results of this research with previous studies, it was found that constructed models with better performance were obtained. Finally, concluded that the MRA, ANN, and ANFIS methods can be successfully employed for predicting the UCS of rocks.
... The prediction and evaluation of rockburst are important prerequisites for formulating prevention and control measures. In recent years, machine learning techniques have been applied well in the field of classification and recognition, and they have become promising methods in rockburst prediction, such as support vector machine [7,8], neural network [9,10], random forest [11,12], cloud model [13,14], extreme learning machine [15,16], Bayesian classifier [17,18], decision tree [19,20], fuzzy C-means clustering [21] and logistic regression model [22]. As a classical supervised learning algorithm, the BP (ack propagation) neural network has been widely used in rockburst prediction [23][24][25]. ...
Rockbursts are serious threats to the safe production of mining, resulting in great casualties and property losses. The accurate prediction of rockburst is an important premise that influences the safety and health of miners. As a classical machine learning algorithm, the back propagation (BP) neural network has been widely used in rockburst prediction. However, there are few reports about the influence study of different training sample sizes, optimization algorithms and index dimensionless methods on the prediction accuracy of BP neural network models. Therefore, 100 groups of typical rockburst engineering samples were collected locally and abroad, and considering the relevance, scientificity and quantifiability of the prediction indexes, the ratio of the maximum tangential stress of surrounding rock to the rock uniaxial compressive strength (σθ/σc), the ratio of the rock uniaxial compressive strength to the rock uniaxial tensile strength (σc/σt) and the elastic energy index (Wet) were chosen as the prediction indexes. When the number of samples was 40, 70 and 100, sixty improved BP models were established based on the standard gradient descent algorithm and four optimization algorithms (momentum gradient descent algorithm, quasi-Newton algorithm, conjugate gradient algorithm, Levenberg–Marquardt algorithm) and four index dimensionless methods (unified extreme value processing method, differentiated extreme value processing method, data averaging processing method, normalized processing method). The prediction performances of each improved model were compared with those of standard BP models. The comparative study results indicate that the sample size, optimization algorithm and dimensionless method have different effects on the prediction accuracy of BP models, which are described as follows: (1) The prediction accuracy value A of the BP model increases with the addition of sample size. The average value Aave of twenty improved models under three kinds of sample sizes increases from Aave (40) = 69.7% to Aave (100) = 75.3%, with a maximal value Amax from Amax (40) = 85.0% to Amax (100) = 97.0%. (2) The value A and comprehensive accuracy value C of the BP model based on four optimization algorithms are generally higher than those of the standard BP model. (3) The improved BP model based on the unified extreme value processing method combined with the Levenberg–Marquardt algorithm has the highest value Amax (100) = 97.0% and value C = 194, and the prediction results of five engineering cases are completely consistent with the actual situation at the site, so this is the best BP neural network model selected in this paper.
... As artificial intelligence and big data advance, machine learning algorithms have been widely accepted to estimate rockburst [13,[33][34][35][36][37][38][39][40][41][42][43]. The ML models only consider the input parameters and rockburst intensity levels and do not give an insight into the rockburst mechanism. ...
The occurrence of rockburst can seriously impact the construction and production of deep underground engineering. To prevent rockburst, machine learning (ML) models have been widely employed to predict rockburst based on some related variables. However, due to the costs and complicated geological conditions, complete datasets to evaluate rockburst cannot always be obtained in rock engineering. To fill this limitation, this study proposed an ensemble tree model suitable for incomplete datasets, i.e., the histogram gradient boosting tree (HGBT), to build intelligent models for rockburst prediction. Three hundred fourteen rockburst cases were employed to develop the HGBT model. The hunger game search (HGS) algorithm was implemented to optimize the HGBT model. The established HGBT model had an excellent testing performance (accuracy of 88.9%). An incomplete database with missing values was applied to compare the performances of HGBT and other ML models (random forest, artificial neural network, and so on). HGBT received an accuracy of 78.8% in the incomplete database, and its capacity was better than that of other ML models. Additionally, the importance of input variables in the HGBT model was analyzed. Finally, the feasibility of the HGBT model was validated by rockburst cases from Sanshandao Gold Mine, China.
... The shortterm rockburst risk was predicted using t-distributed stochastic neighbor embedding (t-SNE), K-means clustering, and extreme gradient boosting (XGBoost) algorithms (28 successfully predict the rockburst classification ranks based on 165 rockburst cases. In deep underground projects, a self-organizing map and fuzzy c-mean clustering techniques were used to cluster rockbursts events (33). Even though several rockburst estimation models have been described and compared by previous researchers (34)(35)(36)(37)(38)(39)(40), developing an accurate and reliable predictive model still poses a significant challenge for the ground, which is likely to experience frequent rock bursts. ...
The rockburst phenomenon is the major source of the high number of casualties and fatalities during the construction of deep underground projects. Rockburst poses a severe hazard to the safety of employees and equipment in subsurface mining operations. It is a hot topic in recent years to examine and overcome rockburst risks for the safe installation of deep urban engineering designs. Therefore, for a cost-effective and safe underground environment, it is crucial to determine and predict rockburst intensity prior to its occurrence. A novel model is presented in this study that combines unsupervised and supervised machine learning approaches in order to predict rockburst risk. The database for this study was built using authentic microseismic monitoring occurrences from the Jinping-II hydropower project in China, which consists of 93 short-term rockburst occurrences with six influential features. The prediction process was succeeded in three steps. Firstly, the original rockburst database's magnification was reduced using a state-of-the-art method called isometric mapping (ISOMAP) algorithm. Secondly, the dataset acquired from ISOMAP was categorized using the fuzzy c-means algorithm (FCM) to reduce the minor spectral heterogeneity impact in homogenous areas. Thirdly, K-Nearest neighbor (KNN) was employed to anticipate different levels of short-term rockburst datasets. The KNN's classification performance was examined using several performance metrics. The proposed model correctly classified about 96% of the rockbursts events in the testing datasets. Hence, the suggested model is a realistic and effective tool for evaluating rockburst intensity. Therefore, the proposed model can be employed to forecast the rockburst risk in the early stages of underground projects that will help to minimize casualties from rockburst.
... In deep underground projects, a self-organizing map and fuzzy c-mean clustering techniques were used to cluster rockbursts events [34]. Even though several rockburst estimation models have been described and compared by previous researchers [35][36][37][38][39][40][41], developing an accurate and reliable predictive model still poses a signi cant challenge for the ground, which is likely to experience frequent rock bursts. ...
The rockburst phenomenon is the major source of the high number of casualties and fatalities during the construction of deep underground projects. Rockburst poses a severe hazard to the safety of employees and equipment in subsurface mining operations. It is a hot topic in recent years to examine and overcome rockburst risks for the safe installation of deep urban engineering designs. Therefore, for a cost-effective and safe underground environment, it is crucial to determine and predict rockburst intensity prior to its occurrence. A novel model is presented in this study that combines unsupervised and supervised machine learning approaches in order to predict rockburst risk. The database for this study was built using authentic microseismic monitoring occurrences from the Jinping-II hydropower project in China, which consists of 93 short-term rockburst occurrences with six influential features. The prediction process was succeeded in three steps. Firstly, the original rockburst database's magnification was reduced using a state-of-the-art method called isometric mapping (ISOMAP) algorithm. Secondly, the dataset acquired from ISOMAP was categorized using the fuzzy c-means algorithm (FCM) to reduce the minor spectral heterogeneity impact in homogenous areas. Thirdly, K-Nearest neighbour (KNN) was employed to anticipate different levels of short-term rockburst datasets. The KNN's classification performance was examined using several performance metrics. The proposed model correctly classified about 96% of the rockbursts events in the testing datasets. Hence, the suggested model is a realistic and effective tool for evaluating rockburst intensity. Therefore, the proposed model can be employed to forecast the rockburst risk in the early stages of underground projects that will help to minimize casualties from rockburst.
... The clustering algorithms are also classified into two general forms: fuzzy and classical. In classical clustering, the members belong to only one class, while in fuzzy clustering, each member can belong to different classes based on different membership degrees [74]. There are several fuzzy clustering algorithms that are considered flexible clustering techniques and have successful applications in solving machine learning problems. ...
The noise of drilling in the dimension stone business is unbearable for both the
workplace and the people who work there. In order to reduce the negative effects
drilling has on the health of the environment, the drilling noise has to be measured,
assessed, and controlled. The main purpose of this work is to investigate an
experimental-intelligent method to predict the noise value of drilling in the dimension
stone industry. For this purpose, 135 laboratory tests are designed on five types of
rocks (four types of hard rock and one type of soft rock): and their results are measured
in the first step. In the second step, due to the unpredicted and uncertain issues in this
case, artificial intelligence (AI) approaches are applied, and the modeling is conducted
using three intelligent systems (IS): namely an adaptive neuro-fuzzy inference system-
SCM (ANFIS-SCM): an adaptive neuro-fuzzy inference system-FCM (ANFIS-FCM):
and the radial basis function network (RBF) neural network. 75% of the samples are
considered for training, and the rest for testing. Several models are constructed, and
the results indicate that although there is no significant difference between the models
according to the performance indices, the proposed construction of ANFIS-SCM can
be considered as an efficient tool in the evaluation of drilling noise. Finally, several
scenarios are designed with different input modes, and the results obtained prove that
the types of rock and the drill bits are more important than the operational
characteristics of the machine.
... The AI approaches are Geotech Geol Eng suitable and reliable systems for most problems that deal with uncertainty (Rad et al. 2012;Rad et al. 2014;Mikaeil et al. 2016;Haghshenas et al. 2017a;Haghshenas et al. 2017b;Naderpour et al. 2019;Mikaeil et al. 2018a;Aryafar et al. 2019;Hosseini et al. 2019;Mikaeil et al. 2019a;Naderpour et al. 2019;Naderpour et al. 2020). Artificial intelligence, as a branch of computer science, includes a wide variety of methods and algorithms that have led to tremendous progress in science (Dormishi et al. 2019;Mikaeil et al. 2019b;Mikaeil et al. 2019c;Faradonbeh et al. 2020;Guido et al. 2020a;Noori et al. 2020;Hosseini et al. 2020;Sheffiee Haghshenas et al. 2020;Behnood, and Golafshani., 2020;Tang et al. 2020). As one of the sub-branches of artificial intelligence, the Group Method of Data Handling (GMDH) type of neural network is an inductive algorithm for computer-based mathematical modeling that is considered for multi-parametric Armaghani et al. 2020). ...
Using the tunnel boring machine (TBM) in tunneling projects contributes significantly to increased efficiency and reducing the time of project implementation in comparison with the classical methods. Since the scheduled deadline is a major issue in the mechanized tunneling project, factors that affect the performance of TBM must be deeply considered in the assessment of tunneling operations. In the implementation of the mechanized tunneling project, a key variable is to predict the penetration rate of TBM. The main aim of this study is to predict the penetration rate of TBM in a novelty framework based on binary classification. For this purpose, the two most effective artificial intelligence (AI) techniques, namely a combination of support vector machine (SVM) and the grasshopper optimization algorithm (GOA) and also the group method of data handling (GMDH) were applied, and a valuable database composed of 2838 was collected from the Kerman water conveyance tunnel project. The values of three parameters including the rotation speed, torque, and thrust force were measured that were considered as input data, and the values of penetration rate were measured as output data. Finally, the best-developed models were able to predict the binary classification of the TBM penetration rate with a testing accuracy of 92% and 91.6% for GMDH and GOA-SVM, respectively. In addition, the results obtained from the sensitivity analysis indicated that the rotation speed had the highest impact on the predicted penetration rate and torque and thrust force had the subsequent maximum impact in descending order, respectively.
