Blast-Induced Ground Vibration Prediction and Control

Blast-induced ground vibration arises from the rapid release of energy during rock fragmentation in mining, quarrying and civil engineering operations. The resulting seismic waves may compromise structural integrity, disturb local communities and disrupt sensitive ecosystems. Prediction and control of these vibrations hinge on understanding charge characteristics, geological context, blast design parameters and propagation dynamics. Traditional approaches rely on empirical scaled-distance relationships that link charge weight and distance to peak particle velocity (PPV), yet they often lack site-specific precision. Recent advances combine high-resolution sensor measurements with machine learning, optimisation algorithms and probabilistic simulation to improve prediction accuracy and support adaptive blast design. Integrating expert judgement through methods such as fuzzy Delphi selection further refines input variables, while hybrid and deep-learning architectures accommodate non-linear interactions among parameters. These developments not only enhance risk assessment and regulatory compliance but also foster more efficient, environmentally sensitive blasting practices worldwide.

Research from Nature Portfolio

Recent studies have demonstrated the potential of advanced learning frameworks to elevate prediction performance beyond conventional techniques. A black-hole-optimised long short-term memory model was shown to outperform traditional neural networks, support vector machines and ensemble trees by effectively capturing temporal and non-linear features in ground vibration data. Another investigation employed a two-stage procedure in which a fuzzy Delphi method identified the most influential blast parameters, followed by hybrid artificial neural networks incorporating evolutionary optimisers; this approach delivered high coefficients of determination in both training and validation. A further work introduced a particle swarm-tuned k-nearest neighbours ensemble, exploring multiple kernel functions to reveal that optimised swarm dynamics can significantly reduce root mean squared error in PPV estimation across diverse blasting scenarios.

Blast-Induced Ground Vibration Prediction and Control publication trend

The graph below shows the total number of articles in blast-induced ground vibration prediction and control across all publications each year (not limited to Nature Index journals).

Technical terms

Peak particle velocity (PPV): The maximum ground-particle speed recorded during a blasting event, used as the principal measure of vibration intensity.

Scaled distance: An empirical metric relating the charge weight and blast-to-receiver distance, employed in traditional vibration attenuation equations.

Fuzzy Delphi method (FDM): A structured expert-elicitation technique that combines fuzzy logic with Delphi rounds to select the most influential variables for modelling.

Monte Carlo simulation: A computational approach that uses repeated random sampling to characterise the uncertainty and distribution of model outputs.

Hybrid machine learning model: A predictive framework that integrates two or more algorithms—often combining optimisation techniques with regression or classification learners—to harness complementary strengths.

References

1. A combination of fuzzy Delphi method and hybrid ANN-based systems to forecast ground vibration resulting from blasting. Scientific Reports (2020).
2. A Novel Hybrid Model for Predicting Blast-Induced Ground Vibration Based on k-Nearest Neighbors and Particle Swarm Optimization. Scientific Reports (2019).
3. A Combination of Feature Selection and Random Forest Techniques to Solve a Problem Related to Blast-Induced Ground Vibration. Applied Sciences (2020).
4. Effective Assessment of Blast-Induced Ground Vibration Using an Optimized Random Forest Model Based on a Harris Hawks Optimization Algorithm. Applied Sciences (2020).
5. Predicting Blast-Induced Ground Vibration in Open-Pit Mines Using Vibration Sensors and Support Vector Regression-Based Optimization Algorithms. Sensors (2019).