Monitoring of blast-induced ground vibration using WSN and prediction with an ANN approach of ACC dungri limestone mine, India

Abstract

Blast-induced ground vibration (BIGV) is an undesirable environmental issue in and around mines. Usage of a high amount of explosive causes ground vibrations that are harmful to the nearby habitats and dwellings. In this paper, an attempt has been made to monitor the BIGV with low-cost wireless sensor network (WSN) and prediction of peak particle velocity (PPV) using an artificial neural network (ANN) technique at ACC Dungri limestone mine, Bargarh, Odisha, India. Eleven blasts PPV were recorded at different locations using wireless sensor network prototype system. The data has been transmitted by ZigBee (IEEE 802.15.4) protocol. The results are very promising and the recorded PPV varies from 0.191 mm/s to 8.60 mm/s. A three-layer, feed-forward back propagation neural network consists of 6 input parameters, 5 hidden neurons, and one output parameters were trained. Obtained results were compared based on correlation of determination (R2) and standard error between recorded and predicted values of PPV.