Blast Noise Management and Prediction

Manuel, Gethin

Abstract

The aim of this work is to manage adverse occupational and environmental impacts from industrial blast noise sources at the DNV Spadeadam Testing & Research Site. Situated in Cumbria, UK, the site carries out crucial major hazards work including improving safety concerns within industry decarbonization sectors and government agencies and performs a variety of explosives and blast testing. This PhD project is concerned with two issues related to blast noise: occupational blast noise impacts to personnel in the near field, and environmental blast noise impacts on communities at long-range.
Part I of this thesis concerns assessments of hearing protection suitability for the protection of site personnel against two differing blast operations carried out at the DNV Spadeadam site. Field measurements of real-world personnel exposures were found to exist beyond the scope of the current national legislative guidance for the selection of hearing protection against impulsive noise, DEF-STD 27:2015. Analysis of waveforms showed that both personnel exposures contain frequency and temporal characteristics not currently represented by the scope of the legislative guidance.
Part II of this thesis is dedicated to implementing tools for the management of blast noise impacts at long-range on residential communities, for a variety of industrial blast and explosion testing carried out at the site. Currently, the operational decisions regarding large explosion trials rely upon computationally expensive prediction models. A Live Noise Monitoring System (LNMS) was deployed across a number of sensitive residential receptors, to monitor environmental noise levels from the site's activities, and correlate the noise measurements with measured and forecast meteorological data. The monitoring network identified that smaller but more frequent blast operations most adversely impacted long-range communities. The database of measurements has been used to assess the performance of existing heuristic and computational models for the prediction of noise impacts up to a number of days in the future. Furthermore, a data-driven model in the form of a deep neural network has been trained and validated for the prediction of noise impacts from a unique explosive process using surface meteorological data. Further measurements characterising the source terms of the operation are required to improve the model. It is concluded that blast noise impacts are best managed by a combination of live monitoring networks and a mixture of predictive tools.

Citation

Manuel, G. (2024). Blast Noise Management and Prediction. (Thesis). University of Salford

• Thesis Type: Thesis
• Deposit Date: Jan 25, 2024
• Publicly Available Date: Mar 5, 2024
• Award Date: Jan 26, 2024