Seismic Bumps Detection

Keywords: Seismic, Bumps

Background

Underground coal mining carries numerous and diverse risks, which include dangers like black lung disease, pockets of flammable gas, rock bursts, and tunnel collapses. Among these hazards, seismic activity emerges as a notable threat. Seismic hazard refers to the likelihood of an earthquake occurring within a specific geographic area during a particular period, with ground motion exceeding a specified threshold. To tackle this issue, a commonly used approach involves the utilization of artificial neural networks (ANNs) to anticipate the emergence of hazardous seismic events, commonly known as "bumps," during subsequent shifts. For this study, researchers constructed an ANN model comprising six hidden layers, employing rectified linear unit (ReLU) and sigmoid activation functions. Predicting hazardous bumps relied on analyzing energy readings and bump counts obtained from preceding shifts. Specifically, if the energy of a seismic event surpassed 1,000 joules, it was categorized as a hazardous bump.

Aim & Objectives

Aim
To Detect the Seismic Bumps Using DL Techniques
Objective
Develop an effective Artificial Neural Network (ANN) model to predict hazardous seismic events, commonly known as "bumps," in underground coal mining operations. The objective is to enhance safety by proactively identifying and categorizing seismic risks through energy readings and bump counts, thus enabling timely implementation of safety measures to mitigate potential hazards and ensure the well-being of mining operators and workers.

Research Methodology

ANN Technique used

Software & Tools

Jupyter, Google Collab

— Preview truncated. The complete document (full chapters, references, diagrams, and appendices) is shared with clients as part of project delivery. —