Improved moth-flame algorithm based on cat chaotic and dynamic cosine factor DOI
Chenhua Xu, Wenjie Zhang,

Zhicheng Tu

и другие.

Review of Scientific Instruments, Год журнала: 2024, Номер 95(2)

Опубликована: Фев. 1, 2024

The moth-flame algorithm shows some shortcomings in solving the complex problem of optimization, such as insufficient population diversity and unbalanced search ability. In this paper, an IMFO (Improved Moth-Flame Optimization) is proposed to be applied optimization function. First, cat chaotic mapping used generate initial position moth improve diversity. Second, cosine inertia weight introduced balance global local abilities algorithm. Third, memory information particle swarm into iterative process speed up convergence population. Finally, Gaussian mutation strategy current optimal solution avoid from falling optimum. Simulation experiments are conducted on 11 benchmark test functions, compared with other improved MFO (Moth-Flame algorithms classical algorithms. results show that has higher accuracy stability above-mentioned functions. experimented verified by optimizing KELM (Kernel Extreme Learning Machine) engineering example exhibits a better performance.

Язык: Английский

Short-Term Rockburst Damage Assessment in Burst-Prone Mines: An Explainable XGBOOST Hybrid Model with SCSO Algorithm DOI

Yingui Qiu,

Jian Zhou

Rock Mechanics and Rock Engineering, Год журнала: 2023, Номер 56(12), С. 8745 - 8770

Опубликована: Сен. 2, 2023

Язык: Английский

Процитировано

61
Occurrence mechanism and prevention technology of rockburst, coal bump and mine earthquake in deep mining DOI Creative Commons
Kun Du,

Ruiyang Bi,

Manoj Khandelwal

и другие.

Geomechanics and Geophysics for Geo-Energy and Geo-Resources, Год журнала: 2024, Номер 10(1)

Опубликована: Май 29, 2024

Abstract Rockburst, coal bump, and mine earthquake are the most important dynamic disaster phenomena in deep mining. This paper summarizes differences connections between rockburst, bumps earthquakes terms of definition, mechanism, phenomenon, evaluation index, etc. The definition evolution progress three categories summarized, as well monitoring, early warning, prevention measures also presented. Firstly, by combining theoretical research with specific technologies engineering field cases, main failure mechanisms introduced. Then, indexes bump a new index rockburst is given. Finally, characteristics warning methods bumps, discussed technology application. At last, future directions put forward.

Язык: Английский

Процитировано

11
CEEMDAN-BILSTM-ANN and SVM Models: Two Robust Predictive Models for Predicting River flow DOI
Elham Ghanbari-Adivi,

Mohammad Ehteram

Water Resources Management, Год журнала: 2025, Номер unknown

Опубликована: Янв. 22, 2025

Язык: Английский

Процитировано

2
Classifying rockburst with confidence: A novel conformal prediction approach DOI Creative Commons
Bemah Ibrahim, Isaac Ahenkorah

International Journal of Mining Science and Technology, Год журнала: 2024, Номер 34(1), С. 51 - 64

Опубликована: Янв. 1, 2024

The scientific community recognizes the seriousness of rockbursts and need for effective mitigation measures. literature reports various successful applications machine learning (ML) models rockburst assessment; however, a significant question remains unanswered: How reliable are these models, at what confidence level classifications made? Typically, ML output single grade even in face intricate out-of-distribution samples, without any associated value. Given susceptibility to errors, it becomes imperative quantify their uncertainty prevent consequential failures. To address this issue, we propose conformal prediction (CP) framework built on traditional (extreme gradient boosting random forest) generate valid while producing measure its output. proposed guarantees marginal coverage and, most cases, conditional test dataset. CP was evaluated case Sanshandao Gold Mine China, where achieved high efficiency applicable levels. Significantly, identified several "confident" from model as unreliable, necessitating expert verification informed decision-making. improves reliability accuracy assessments, with potential bolster user confidence.

Язык: Английский

Процитировано

8
Research on gas tunnel prediction in Central Sichuan using energy valley optimizer and support vector machine DOI
Yuxuan Liu,

Peidong Su,

Peng Qiu

и другие.

Bulletin of Engineering Geology and the Environment, Год журнала: 2025, Номер 84(1)

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

1
Fault Prediction Modeling for High-Impact Recorders Based on IPSO-SVM DOI Creative Commons

Linyu Li,

You Wenbin,

Yonghong Ding

и другие.

Applied Sciences, Год журнала: 2025, Номер 15(3), С. 1343 - 1343

Опубликована: Янв. 27, 2025

The challenge in reusing high-impact recorders lies developing an efficient and accurate failure prediction model under small-sample conditions. To address this issue, study proposes IPSO-SVM model. First, the particle swarms IPSO algorithm were grouped based on their exploration exploitation functions, dynamic inertia weight mechanisms designed accordingly. grouping ratio was dynamically adjusted during iterations to enhance optimization performance. Tests using benchmark functions verified that approach improves convergence accuracy stability compared conventional PSO algorithms. Subsequently, 5-fold cross-validation of SVM used as fitness value, employed optimize penalty kernel parameters Trained experimental data, achieved a 90.5%, outperforming PSO-SVM model’s 85%. These results demonstrate potential addressing challenges

Язык: Английский

Процитировано

1
Application for Identifying the Origin and Predicting the Physiologically Active Ingredient Contents of Gastrodia elata Blume Using Visible–Near-Infrared Spectroscopy Combined with Machine Learning DOI Creative Commons
Jinfang Ma, Xue Zhou,

Baiheng Xie

и другие.

Foods, Год журнала: 2023, Номер 12(22), С. 4061 - 4061

Опубликована: Ноя. 8, 2023

Gastrodia elata (G. elata) Blume is widely used as a health product with significant economic, medicinal, and ecological values. Due to variations in the geographical origin, soil pH, content of organic matter, levels physiologically active ingredient contents G. from different origins may vary. Therefore, rapid methods for predicting origin these ingredients are important market. This paper proposes visible-near-infrared (Vis-NIR) spectroscopy technology combined machine learning. A variety learning models were benchmarked against one-dimensional convolutional neural network (1D-CNN) terms accuracy. In identification models, 1D-CNN demonstrated excellent performance, F1 score being 1.0000, correctly identifying 11 origins. quantitative outperformed other three algorithms. For prediction set eight ingredients, namely, GA, HA, PE, PB, PC, PA, GA + total, RMSEP values 0.2881, 0.0871, 0.3387, 0.2485, 0.0761, 0.7027, 0.3664, 1.2965, respectively. The Rp2 0.9278, 0.9321, 0.9433, 0.9094, 0.9454, 0.9282, 0.9173, 0.9323, study that showed highly accurate non-linear descriptive capability. proposed combinations Vis-NIR have potential quality evaluation elata.

Язык: Английский

Процитировано

14
Method and Validation of Coal Mine Gas Concentration Prediction by Integrating PSO Algorithm and LSTM Network DOI Open Access
Guangyu Yang, Quanjie Zhu,

Dacang Wang

и другие.

Processes, Год журнала: 2024, Номер 12(5), С. 898 - 898

Опубликована: Апрель 28, 2024

Gas concentration monitoring is an effective method for predicting gas disasters in mines. In response to the shortcomings of low efficiency and accuracy conventional prediction, a new prediction based on Particle Swarm Optimization Long Short-Term Memory Network (PSO-LSTM) proposed. First, principle PSO-LSTM fusion model analyzed, analysis constructed. Second, data are normalized preprocessed. The PSO algorithm utilized optimize training set LSTM model, facilitating selection model. Finally, MAE, RMSE, coefficient determination R2 evaluation indicators proposed verify analyze results. comparison verification research was conducted using measured mine as sample data. experimental results show that: (1) maximum RMSE predicted 0.0029, minimum 0.0010 when size changes. This verifies reliability effect (2) predictive performance all models ranks follows: > SVR-LSTM PSO-GRU. Comparative with demonstrates that more concentration, further confirming superiority this prediction.

Язык: Английский

Процитировано

6
Exploration and Improvement of Fuzzy Evaluation Model for Rockburst DOI
Qiwei Wang, Chao Wang, Yu Liu

и другие.

Mining Metallurgy & Exploration, Год журнала: 2024, Номер 41(2), С. 559 - 587

Опубликована: Фев. 28, 2024

Язык: Английский

Процитировано

5
An ensemble model of explainable soft computing for failure mode identification in reinforced concrete shear walls DOI

Yingui Qiu,

Chuanqi Li, Shuai Huang

и другие.

Journal of Building Engineering, Год журнала: 2023, Номер 82, С. 108386 - 108386

Опубликована: Дек. 26, 2023

Язык: Английский

Процитировано

10