Enhancing Ground Vibration Prediction in Mine Blasting: A Committee Machine Intelligent System Optimized with Metaheuristic Algorithms

Ground vibrations resulting from mine blasting pose significant risks to nearby structures and the environment. This paper introduces an advanced framework employing Committee Machine Intelligent Systems (CMIS) optimized by cutting-edge metaheuristic algorithms to predict ground vibrations. Initiall...

Full description

Saved in:
Bibliographic Details
Published in:Natural resources research (New York, N.Y.) Vol. 34; no. 6; pp. 3449 - 3475
Main Authors: Hasanipanah, Mahdi, Amnieh, Hassan Bakhshandeh
Format: Journal Article
Language:English
Published: New York Springer Nature B.V 01.12.2025
Subjects:
Accuracy
Algorithms
Artificial neural networks
Bayesian analysis
Blasting
Error analysis
Explosives
Genetic algorithms
Ground motion
Heuristic methods
Mathematical models
Neural networks
Optimization
Optimization algorithms
Predictions
Quantiles
Regularization
Sensitivity analysis
Support vector machines
Swarm intelligence
Uncertainty analysis
Variables
Velocity
Vibration
Vibration analysis
Vibrations
ISSN:1520-7439, 1573-8981
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Ground vibrations resulting from mine blasting pose significant risks to nearby structures and the environment. This paper introduces an advanced framework employing Committee Machine Intelligent Systems (CMIS) optimized by cutting-edge metaheuristic algorithms to predict ground vibrations. Initially, cascaded feedforward neural networks (CFNN) were developed using Levenberg-Marquardt algorithm and Bayesian regularization (BR). These neural networks were integrated into a CMIS to enhance prediction accuracy. Optimization of the CMIS was carried out using four metaheuristic algorithms, including the grey wolf optimizer (GWO), Harris hawks optimization (HHO), artificial bee colony, and genetic algorithm. Comprehensive evaluation metrics such as quantile-quantile plots, Taylor diagrams, and error analysis were employed to assess model performance. Results indicated that the CMIS model optimized by HHO provided the most accurate ground vibration predictions, surpassing existing models in the literature. Sensitivity analysis identified weight charge per delay as the most critical factor influencing ground vibration. Moreover, the uncertainty analysis indicated that the CFNN-BR model had the narrowest uncertainty band, followed by the CMIS-GWO model. The study underscores the potential of integrating CMIS with metaheuristic algorithms for precise and reliable vibration prediction in mining operations.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1520-7439
1573-8981
DOI:10.1007/s11053-025-10518-6