Figure - available from: Environmental Earth Sciences
This content is subject to copyright. Terms and conditions apply.
Source publication
The accurate determination of rock elasticity modulus is crucial for geomechanical analysis and reliable rock engineering designs. Traditional experimental methods have limitations in estimating elasticity modulus, prompting the adoption of artificial intelligence and data-driven techniques to develop adaptive and accurate predictive models. This s...
Similar publications
Complicated geological conditions such as high geo-stress and high ground temperatures often have a significant impact on the mechanical behavior and failure potential of hard rocks in deep engineering. This study aims to investigate the mechanical behaviors and rockburst proneness of granite under high differential stress and high temperatures in...
Citations
... The mechanical properties of different types of rocks are significantly different due to the control of their rockforming mineral components 42 . For example, sediments with higher quartz particle content and lower clay content are difficult to be compacted during burial. ...
The Cretaceous Formation in Kuqa Depression is the main battlefield for increasing natural gas storage during "the 14th Five-Year Plan" period in Tarim Basin. Engineering problems occur frequently when drilling into very thick gravel layers. In order to clarify the compaction of shallow gravel layer, this paper takes the Bozi-Dabei area as an example, and puts forward the method of "classification of compaction degree based on quadratic clustering". The results show that the method consists of four steps: the selection of characteristic parameters, the clustering effect of single cluster analysis, the classification standard of quadratic cluster construction, and the construction and verification of the classification standard of compaction degree. In the vertical direction, affected by Tianshan uplift and Tuzimazha Salt Wall, the compaction degree of three fan bodies in the study area is significantly different, and the compaction degree of Bozi fan body is stronger than that of Dabei fan body. On the plane, with the increase of source distance, the compaction degree of Bozi 3 fan body decreases due to the influence of gravel composition. The compaction degree of Bozi 1 fan body first increased and then decreased. The compaction degree of Dabei fan body is weak.
... These approaches enhance model adaptability, reduce reliance on traditional calibration processes, and enable real-time estimation across complex, nonlinear systems. Additionally, modern computational techniques now support the simulation of coupled mechanical, geological, and hydrological processes, offering more integrated and robust frameworks for MEMs [52,53]. ...
Accurate determination of rock mechanical parameters is crucial for safe and efficient drilling, production, and reservoir development in these complex formations. While numerous empirical relationships have been proposed to estimate these properties, there is no reliable way to guide the selection of suitable models in new studies. This study develops a one-dimensional mechanical earth model (1D MEM) for a carbonate reservoir in southwestern Iran to systematically evaluate existing empirical relationships for carbonate rocks. First, laboratory tests were performed on the obtained core samples. The results were then integrated with petrophysical data to establish new empirical correlations and convert dynamic measurements into static elastic and rock strength parameters. The study involved estimating Biot's effective stress coefficient, calculating the pore pressure and in-situ stresses profiles, and calibrating them using field data. Additionally, this study reviewed existing methods and empirical relationships for estimating key physico-mechanical rock parameters. The predictions of these relationships were statistically compared with results from the current research as reference data. Using the Taylor diagram as an innovative method the relationships were ranked based on their prediction accuracy, providing a guideline for selecting optimal relationships to use in future studies on similar lithologies. The comprehensive statistical analysis on more than 200 existing empirical relationships showed that the effectiveness of predictive relationships is influenced by factors such as lithology, geological environment, equation forms, and data acquisition. Relationships developed specifically for carbonate rocks particularly limestone produce more reliable results than those intended for mixed or non-carbonate rocks.
... These metrics are derived from the confusion matrix, which compares the actual classes (rows) against the predicted classes (columns). The basics of a confusion matrix and its evaluation principles can be found in (Mao et al. 2024). Table 5 presents a template of the confusion for four class classifications along with the key performance metrics. ...
Extracting rock mass strength properties from existing data like Measurement While Drilling (MWD) is important to reduce the cost of additional geological and geotechnical surveys. This study presents an approach that combines clustering (unsupervised learning) and classification algorithms to identify similar rock groups for their prediction. The dataset comprises 272,272 MWD from 2,790 drill holes, split into 215,401 data points (2,332 drill holes) for cross-validation, and another 215,401 data points, from 558 previously unseen drill holes for testing. Principal component analysis (PCA) and clustering algorithms such as K-means, Gaussian mixture, C Fuzy, and hierarchical clustering were employed to group rocks with similar MWD parameters. The combination of PCA and k-means clustering provides good cluster quality which best describes the different rock strength characteristics (clusters), as revealed by geological investigation and coring data. After identifying the rock categories, Extra Gradient Boosting (XGBoost), and Adaptive Boosting (AdaBoost) approaches were used to develop classification models for rock strength prediction. The XGBoost model achieved the best and most reliable performance with accuracy, precision, recall, and F1 score exceeding 98% on the test set. This study highlights the synergetic benefits of combining unsupervised and supervised machine learning techniques to predict rock mass conditions, especially in scenarios with limited geological information or unavailable labeled data.
... The aim of this study is to investigate the applicability and utilization of gelatin-like simulation mass from the point of view of scalability, and to determine its static parameters using non-destructive testing procedures. One of the main material parameters is the elastic modulus E, which describes the material's behavior under stress [24]. Young's modulus measures the material's stiffness and its ability to deform elastically under stress [25]. ...
The investigation of wave propagation in the geological environment is warranted, and will ultimately help to provide a better understanding of the response of subsoil to excitation. Frequently utilized physical modeling represents a simplification of the global natural system for the needs of the investigation of static and dynamic phenomena with regard to the time domain. The determination of appropriate model materials is probably the most important task for physical model creation. Considering that subsoil represents a crucial medium for wave propagation, an evaluation of suitable model materials was carried out. A plate load test with a circular plate is a non-destructive method for determining the static bearing capacities of soils and aggregates, which are usually expressed by the deformation modulus Edef,2 (MPa) and the static modulus of elasticity E (MPa). A lightweight deflectometer test was used to characterize the impact modulus of deformation Evd (MPa), which is determined based on the pressure under the load plate due to the impact load. A representative propagation of the load–settlement curve for the PLT and the acceleration–time curve for the hammer drop test were investigated. The calculated E values were found to be in the interval between 2.6 and 5.7 kPa, and depending on the load cycle, the values of E ranged from 2.6 to 3.1 kPa. The modulus E from the hammer drop test was significantly larger than the interval between 10.6 and 40.4 kPa. The values of the dynamic multiplier, as a ratio of the hammer drop value to the PLT value, of the modulus E ranged from 4.1 to 13.0. The output of the plate load testing was utilized for the calibration of the finite element method (FEM) numerical model. Both the physical and numerical models showed practically ideal linear behavior of the mass. However, the testing of gelatin-like materials is a complex process because of their viscoelastic nonlinear behavior.
... These include the prediction of triaxial compressive strength, uniaxial compressive strength and in-situ plate bearing tests using artificial intelligence. (Wang 2016;Grover 2022b, 2024;Mao et al. 2023bMao et al. , 2024. ...
... Similarly, rounded gravel complicates compaction efforts, necessitating more costly soil improvement methods. The liquid and plastic limits of soils also play a critical role in determining bearing capacity, assessing liquefaction risk, and selecting improvement methods, all of which directly impact the cost of engineering projects (Hoek 2006;Hoek and Bray 2009;Genç 2011;Wu et al. 2018;Azarafza et al. 2021;Mao et al. 2024). In addition, many studies utilize artificial intelligence techniques across various branches of engineering to determine shear strength in mine blasting, assess soil bearing capacity, and evaluate soil permeability Grover 2021d, 2023;Rabbani et al. 2022Rabbani et al. , 2023aHosseini et al. 2023;Mao et al. 2023a;Nanehkaran et al. 2023;Rabbani et al. 2024a, b). ...
The Casagrande chart is traditionally used for determining soil classes. However, processing samples individually on this chart is time-consuming, and human error, particularly at the classification boundaries, can lead to incorrect soil classification. To address these issues, this study employs machine learning algorithms to classify different soil types more efficiently and accurately. The primary goal is to integrate machine learning into engineering geology studies, leveraging technological advancements. As part of the study, field and experimental work was conducted, beginning with the collection of 272 soil samples from the designated study area to represent the entire region. The initial physical properties of these samples were then determined. The soil samples were carefully double-bagged and transported to the laboratory to prevent any degradation. Upon arrival, the water content of the samples was determined first. Subsequently, sieve analysis, consistency limits, and specific gravity tests were conducted in sequence. For the classification of soil types, machine learning methods, including Decision Tree (DT), Support Vector Machines (SVM), K-Nearest Neighbor (k-NN), and Artificial Neural Networks (ANN), were employed. Among these, the DT model demonstrated the highest performance, achieving a success rate of 91.5% across five different soil classifications. The ANN, SVM, and k-NN models followed, with success rates of 89.0%, 89.0%, and 86.0%, respectively. In addition, the hyperparameter optimisation of the models used in the study was provided and the minimum classification error analysis was presented. This study significantly contributes to the effective and efficient analysis of experimental data, demonstrating that soil classification can be successfully performed by integrating machine learning methods with numerical data.
... Because it does not consider the complex basic characteristics of the frequency component of blasting vibration signal, the proposed frequency prediction formula is not scientific (Zhang et al., 2023). This paper mainly studies the frequency characteristics and propagation laws of blasting seismic waves, but in engineering practice, the impact of blasting seismic waves on structures is the result of the joint action of peak vibration velocity and frequency (Mao et al., 2024). In the future, the research results should be combined with the peak vibration velocity to evaluate the impact of blasting earthquakes. ...
The dynamic response characteristics of high and steep slopes under the action of earthquakes and blasting was focused on, especially the frequency distribution and propagation laws, which are crucial for slope stability assessment. Using stress wave theory as the theoretical basis and advanced FLAC3D numerical simulation technology, we systematically analyze the frequency response of slope under different joint conditions under seismic waves. The nonlinear characteristics of reflected P-wave coefficient and the significant sensitivity of joint to incident wave frequency are revealed when the Angle of incident P-wave changes. The results show that with the increase of the incidence Angle of the incident P-wave, the reflection coefficient of the reflected P-wave decreases slowly at first and then increases sharply to 1.0. The reflection coefficient of the wave at the joint is more sensitive to the frequency of the incident wave. In a biplanar rock mass, multiple reflections of waves between structural planes produce transmitted waves with different time differences.
