Application of artificial intelligence in coal mine ultra-deep roadway engineering—a review DOI Creative Commons
Bingbing Yu, Bo Wang, Yuantong Zhang

и другие.

Artificial Intelligence Review, Год журнала: 2024, Номер 57(10)

Опубликована: Авг. 19, 2024

The deep integration of computer field and coal mining is the only way to mine intellectualization. A variety artificial intelligence tools have been applied in open-pit shallow mines. However, with geometric increase demand, contradiction between supply demand becoming more serious, exploitation resources from layer (> 600 m) has become an inevitable trend. Well then, as a new engineering scene, harsh conditions "three high one disturbance" seriously threaten safety personnel. superposition complex environment makes number input factors sharply, which leads application roadway engineering. guidance not mature, construction various databases missing, there are still some problems universality applicability. To this end, paper starts introduction operating characteristics tools, conducts comprehensive study relevant high-level articles published top journals. It systematically sorts out research progress that successfully solved five directions rock mechanics strength, surrounding stability, rock-burst, roof fall risks micro-seismic events. While objectively evaluating performance different it also expounds its own views on key results. Literature review shows whether development tool or comparative model, ANN than 98%, performs extremely well direction stability risk, accuracy rate 90%. As most mature AI application, mechanical strength experienced process "SVM → DL XGBoost RF". dataset small samples (< 100) big 1000), R2 tree-based models can be stabilized at 95%. rock-burst prediction mainly focuses monitoring data. Whether sample large-scale data BN remains above 85%. evaluation events recent years. image processing CNN important. signal recognition classification accounts for 90%, potential source location needs further explored. In general, nature itself first choice almost all influencing factors. At same time, update iteration methods (micro-seismic, ground sound, separation, deformation, etc.) expands database, making possible obtain due threat life cost equipment, very difficult before. parameter selection method combining lithology conditions, geological will gradually research. Finally, follow-up work collation on-the-spot investigation, existing mines, explores engineering, puts forward focus challenging future, gives opinions.

Язык: Английский

Stability prediction of underground entry-type excavations based on particle swarm optimization and gradient boosting decision tree DOI Creative Commons
Jian Zhou, Shuai Huang, Ming Tao

и другие.

Underground Space, Год журнала: 2022, Номер 9, С. 234 - 249

Опубликована: Дек. 19, 2022

The stability of underground entry-type excavations will directly affect the working environment and safety staff. Empirical critical span graphs traditional statistics learning methods can not meet requirements high accuracy for assessment excavations. Therefore, this study proposes a new prediction method based on machine to scientifically adjust graph. Accordingly, particle swarm optimization (PSO) algorithm is used optimize core parameters gradient boosting decision tree (GBDT), abbreviated as PSO-GBDT. Moreover, classification performance eight other classifiers including GDBT, k-nearest neighbors (KNN), two kinds support vector machines (SVM), Gaussian naive Bayes (GNB), logistic regression (LR) linear discriminant analysis (LDA) are also applied compare with proposed model. Findings revealed that compared models, PSO-GBDT undoubtedly most reliable, its up 0.93. model has great potential provide more scientific accurate choice In addition, each predict category several grid points divided by graph, updated graph discussed in combination previous studies. results show advantages being scientific, efficient updating output boundary strict theoretical support, which help mine operators make favorable economic decisions.

Язык: Английский

Процитировано

29
Microseismic Location in Hardrock Metal Mines by Machine Learning Models Based on Hyperparameter Optimization Using Bayesian Optimizer DOI
Jian Zhou,

Xiaojie Shen,

Yingui Qiu

и другие.

Rock Mechanics and Rock Engineering, Год журнала: 2023, Номер 56(12), С. 8771 - 8788

Опубликована: Сен. 3, 2023

Язык: Английский

Процитировано

22
Prediction of Flyrock Distance in Surface Mining Using a Novel Hybrid Model of Harris Hawks Optimization with Multi-strategies-based Support Vector Regression DOI
Chuanqi Li, Jian Zhou, Kun Du

и другие.

Natural Resources Research, Год журнала: 2023, Номер 32(6), С. 2995 - 3023

Опубликована: Сен. 4, 2023

Язык: Английский

Процитировано

19
Investigation on crack propagation and reasonable wall thickness of supercritical CO2 pipeline DOI
Dong Zhang, Xiaoben Liu, Yaru Fu

и другие.

Engineering Fracture Mechanics, Год журнала: 2024, Номер 298, С. 109951 - 109951

Опубликована: Фев. 14, 2024

Язык: Английский

Процитировано

7
A comparative analysis of hybrid RF models for efficient lithology prediction in hard rock tunneling using TBM working parameters DOI
Jian Zhou, Peixi Yang,

Weixun Yong

и другие.

Acta Geophysica, Год журнала: 2024, Номер 72(3), С. 1847 - 1866

Опубликована: Апрель 15, 2024

Язык: Английский

Процитировано

7
Refined Approaches for Open Stope Stability Analysis in Mining Environments: Hybrid SVM Model with Multi-optimization Strategies and GP Technique DOI
Shuai Huang, Jian Zhou

Rock Mechanics and Rock Engineering, Год журнала: 2024, Номер 57(11), С. 9781 - 9804

Опубликована: Июль 11, 2024

Язык: Английский

Процитировано

7
Estimating the mean cutting force of conical picks using random forest with salp swarm algorithm DOI Creative Commons
Jian Zhou, Yong Dai, Ming Tao

и другие.

Results in Engineering, Год журнала: 2023, Номер 17, С. 100892 - 100892

Опубликована: Янв. 13, 2023

Conical picks are widely used as cutting tools in shearers and roadheaders, the mean force (MCF) is one of important parameters affecting conical pick performance. As MCF depends on a number due to that existing empirical theoretical formulas numerical modelling not sufficient enough reliable predict proficient manner. So, this research, novel intelligent model based random forest algorithm (RF) heuristic called salp swarm (SSA) have been applied determine optimal hyper-parameters RF, root square error fitness function. A total 188 data samples including 50 rock types seven (tensile strength σt, compressive σc, cone angle θ, depth d, attack γ, rake α back-clearance β) were collected develop an SSA-RF for prediction. The prediction results compared with influential four classical models, such forest, extreme learning machine, support vector machine radial basis function neural network. absolute (MAE), (RMSE), percentage (MAPE) Pearson correlation coefficient (R2) employed evaluation indexes compare capability different predicting models. MAE (0.509 0.996), RMSE (0.882 1.165), MAPE (0.146 0.402) R2 (0.975 0.910) values between measured predicted training testing phases clearly demonstrate superiority other tools. sensitivity analysis has also performed understand influence each input parameter MCF, which indicates d σt most variables

Язык: Английский

Процитировано

15
Predicting dynamic compressive strength of frozen-thawed rocks by characteristic impedance and data-driven methods DOI Creative Commons
Shengtao Zhou, Zong‐Xian Zhang, Xuedong Luo

и другие.

Journal of Rock Mechanics and Geotechnical Engineering, Год журнала: 2023, Номер 16(7), С. 2591 - 2606

Опубликована: Дек. 14, 2023

In cold regions, the dynamic compressive strength (DCS) of rock damaged by freeze-thaw weathering significantly influences stability engineering. Nevertheless, testing under conditions is often both time-consuming and expensive. Therefore, this study considers effect characteristic impedance on DCS aims to quickly determine frozen-thawed rocks through application machine-learning techniques. Initially, a database for rocks, comprising 216 specimens, was compiled. Three external load parameters (freeze-thaw cycle number, confining pressure, impact pressure) two (characteristic porosity) were selected as input variables, with predicted target. This research optimized kernel scale, penalty factor, insensitive loss coefficient support vector regression (SVR) model using five swarm intelligent optimization algorithms, leading development hybrid models. addition, statistical prediction equation multiple linear techniques developed. The performance models comprehensively evaluated error indexes trend indexes. A sensitivity analysis based cosine amplitude method has also been conducted. results demonstrate that proposed SVR-based consistently provided accurate predictions. Among these models, SVR chameleon algorithm exhibited best performance, metrics indicating its effectiveness, including root mean square (RMSE) = 3.9675, absolute (MAE) 2.9673, determination (R2) 0.98631, variance accounted (VAF) 98.634. suggests yielded most optimal enhancing Notably, pressure emerged influential in prediction. anticipated serve reliable reference estimating subjected weathering.

Язык: Английский

Процитировано

15
Applications of Microseismic Monitoring Technique in Coal Mines: A State-of-the-Art Review DOI Creative Commons
Fei Liu, Yan Wang, Miaomiao Kou

и другие.

Applied Sciences, Год журнала: 2024, Номер 14(4), С. 1509 - 1509

Опубликована: Фев. 13, 2024

China’s coal mines have to extend greater depths for the exploitation of more mineral resources, and they suffered catastrophic mining-induced disasters, such as rockbursts, water inrushes, gas outbursts, roof fall accidents. The microseismic monitoring technique is a practical tool mine safety management, which extensively utilized in many Chinese mines. Microcracks coal/rock masses are recorded microseismicities field, potential instabilities can be assessed by in-depth analysis parameters. This study provides state-of-the-art review achievements developments It also presents some prospects improving location accuracy microseismicity, efficient intelligent processing data, comprehensive assessment instabilities, development new equipment. valuable management may contribute deep mining production.

Язык: Английский

Процитировано

6
NiOA: A Novel Metaheuristic Algorithm Modeled on the Stealth and Precision of Japanese Ninjas DOI Open Access

El-Sayed M. El-kenawy,

Faris H. Rizk,

Ahmed Mohamed Zaki

и другие.

Journal of Artificial Intelligence in Engineering Practice., Год журнала: 2024, Номер 1(2), С. 17 - 35

Опубликована: Окт. 17, 2024

This paper presents a new metaheuristic optimization algorithm called the Ninja Optimization Algorithm (NiOA) owing to its characteristics such as stealth, precision, and adaptability of ninjas Japan. NiOA is proposed avoid high exploration exploitation costs within complex search spaces problem getting trapped in local optima. The imitates ninja searching techniques because it has scanning phase, adapted large areas look for answers, while more specific phase used refine answers found. performance compared with other benchmark functions some frequently CEC 2005 benchmarks. These benchmarks are well suited test unimodal multimodal problems good quality. Experimental results prove that can significantly provide better regarding solution quality, convergence rate, time complexity, suggesting robust solving high-dimensional large-scale problems. Furthermore, reveals applicable solve different kinds spaces, signifying be practice on scientific engineering

Язык: Английский

Процитировано

6