A Kernel Extreme Learning Machine-Grey Wolf Optimizer (KELM-GWO) Model to Predict Uniaxial Compressive Strength of Rock

Applied Sciences, Год журнала: 2022, Номер 12(17), С. 8468 - 8468

Опубликована: Авг. 24, 2022

Uniaxial compressive strength (UCS) is one of the most important parameters to characterize rock mass in geotechnical engineering design and construction. In this study, a novel kernel extreme learning machine-grey wolf optimizer (KELM-GWO) model was proposed predict UCS 271 samples. Four namely porosity (Pn, %), Schmidt hardness rebound number (SHR), P-wave velocity (Vp, km/s), point load (PLS, MPa) were considered as input variables, output variable. To verify effectiveness accuracy KELM-GWO model, machine (ELM), KELM, deep (DELM) back-propagation neural network (BPNN), empirical established compared with UCS. The root mean square error (RMSE), determination coefficient (R2), absolute (MAE), prediction (U1), quality (U2), variance accounted for (VAF) adopted evaluate all models study. results demonstrate that best predicting performance indices. Additionally, identified parameter by using impact value (MIV) technique.

Язык: Английский

---

Development of a hybrid artificial intelligence model to predict the uniaxial compressive strength of a new aseismic layer made of rubber-sand concrete

Mechanics of Advanced Materials and Structures, Год журнала: 2022, Номер 30(11), С. 2185 - 2202

Опубликована: Март 23, 2022

This study, proposes the use of a novel rubber-sand concrete (RSC) material, which comprises rubber particles, sand, and cement, as an aseismic material in practical engineering construction. The uniaxial compressive strength (UCS) damping materials is important factor that directly affects seismic activity underground structures. To predict UCS RSC, artificial intelligence model back propagation neural network (BPNN), optimized through four swarm optimization (SIO) algorithms: particle algorithm (PSO), fruit fly (FOA), lion (LSO), sparrow search (SSA), used. dataset for prediction models was obtained from compression tests RSC laboratory. performances hybrid were evaluated using six performance indicators: root mean square error (RMSE), correlation coefficient (R), determination (R2), absolute (MAE), (MSE), sum (SSE).The capability these graded based on indicators ranking system. results show ability LSO-BPNN better than three other models, with RMSE (1.0635, 1.2352), R (0.9887, 0.9713), R2 (0.9776, 0.9165), MAE (0.7257, 0.8243), MSE (1.1352, 1.5256), SSE (64.7074, 36.6151), score (24, 24) training testing phases, respectively. Therefore, efficient accurate method predicting RSCs. Sensitivity analysis showed sand most elements affected prediction, followed by lowest relative importance being RPZ. study provides guidance extension application to engineering.

Язык: Английский

---

Ventilation on demand in underground mines using neuro-fuzzy models: Modeling and laboratory-scale experimental validation

Engineering Applications of Artificial Intelligence, Год журнала: 2024, Номер 133, С. 108048 - 108048

Опубликована: Фев. 19, 2024

Язык: Английский

---

Feature selection using metaheuristics made easy: Open source MAFESE library in Python

Future Generation Computer Systems, Год журнала: 2024, Номер 160, С. 340 - 358

Опубликована: Июнь 7, 2024

Язык: Английский

---

An enhanced stability evaluation system for entry-type excavations: Utilizing a hybrid bagging-SVM model, GP and kriging techniques

Journal of Rock Mechanics and Geotechnical Engineering, Год журнала: 2024, Номер unknown

Опубликована: Июль 1, 2024

In underground mining, especially in entry-type excavations, the instability of surrounding rock structures can lead to incalculable losses. As a crucial tool for stability analysis critical span graph must be updated meet more stringent engineering requirements. Given this, this study introduces support vector machine (SVM), along with multiple ensemble (bagging, adaptive boosting, and stacking) optimization (Harris hawks (HHO), cuckoo search (CS)) techniques, overcome limitations traditional methods. The indicates that hybrid model combining SVM, bagging, CS strategies has good prediction performance, its test accuracy reaches 0.86. Furthermore, partition scheme is adjusted based on CS-BSVM 399 cases. Compared previous empirical or semi-empirical methods, new overcomes interference subjective factors possesses higher interpretability. Since relying solely one technology cannot ensure credibility, further genetic programming (GP) kriging interpolation techniques. explicit expressions derived through GP offer probability value, technique provide interpolated definitions two subclasses. Finally, platform developed above three approaches, which rapidly feedback.

Язык: Английский

---

Numerical Simulation of Gas Extraction in Coal Seam Strengthened by Static Blasting

Sustainability, Год журнала: 2022, Номер 14(19), С. 12484 - 12484

Опубликована: Сен. 30, 2022

For mines with low permeability and high gas emissions, static blasting technology is used to pre-split the coal seam increase strengthen extraction, which will significantly reduce occurrence of accidents in mines. Taking Wangjialing Coal Mine as research object, mathematical model fluid-solid established. The numerical simulation software COMSOL simulate established model. Simultaneously, factors affecting efficiency extraction are analyzed by adjusting parameters. results reveal a more significant drop pressure increasing time. At 10 d, 30 90 d 180 increases 11.80%, 18.67%, 22.22% 24.13% comparison conventional extraction. In studying influence expansion other on during blasting, it found that change negative has little effect Static can achieve safe mining, providing basis field application efficient

Язык: Английский

---

Research and Application of an Intelligent Prediction of Rock Bursts Based on a Bayes-Optimized Convolutional Neural Network

International Journal of Geomechanics, Год журнала: 2023, Номер 23(5)

Опубликована: Март 2, 2023

Intelligent prediction of rock bursts has great significance in mechanics research and a high value engineering applications. An intelligent rockburst method based on Bayes-optimized convolutional neural network (BOCNN) was proposed. First, an exploratory analysis data conducted using joint distribution diagrams the heat map correlation matrix to establish high-quality set cases parameter system for prediction. Second, six models were built by combining machine learning algorithms, such as random forest, k-nearest neighbor (KNN), Bayes, deep (CNN1d CNN2d), BOCNN. In addition, we used accuracy, precision, recall, F1 score, receiver operating characteristic curve, Taylor diagram, probability indicator results indicators evaluate accuracy models. A comparative explore with good robustness, generalization performance, accuracy. Moreover, 11 established analysis. Then, MATLAB tool build applied findings Jiangbian Hydropower Station Sichuan Province, China. The study show that can provide technical support predicting hazards mining, transportation, water conservancy hydropower projects scientific basis later construction design structures.

Язык: Английский

---

Hybrid method for analyzing air thermal conditions in underground mines

Expert Systems with Applications, Год журнала: 2024, Номер 245, С. 123026 - 123026

Опубликована: Янв. 1, 2024

Язык: Английский

---

Deformation extent prediction of roadway roof during non-support period using support vector regression combined with swarm intelligent bionic optimization algorithms

Tunnelling and Underground Space Technology, Год журнала: 2024, Номер 145, С. 105585 - 105585

Опубликована: Янв. 11, 2024

Язык: Английский

---

Regulation and Optimization of Air Quantity in a Mine Ventilation Network with Multiple Fans

Archives of Mining Sciences, Год журнала: 2023, Номер unknown

Опубликована: Июль 20, 2023

Regulation and optimization of aiR Quantity in a mine Ventilation netwoRk with multiple fans the ventilation system underground is an important guarantee for workers' safety environmental conditions.As mining activities continue, constantly changing.therefore, to ensure on demand, network regulation are very important.in this paper, path method based graph theory studied.however, existing algorithms do not meet needs actual optimization.therefore, algorithm optimized improved from four aspects.First, depth-first search algorithm, independent proposed solve problem false paths searched when there unidirectional circuit network.Secondly, calculation formula amended that number downcast upcast shaft, multi-downcast multi-upcast shaft circuits calculated accurately.thirdly, avoid both increase control points multi-fan disturbances airflow distribution by determining reference through all paths, shared fan must be identified.Fourthly, position regulators determined optimized, final air quantity realized.the case study shows can effectively accurately realize network.

Язык: Английский

---