In India, several underground coal mines operate in close proximity to active surface mines where regular blasting is being practiced to remove overburden rocks and coal seam. Due to this, safety and stability of underground mine openings and structures viz. coal pillars, roof, water dams, ventilation, isolation stoppings and others may be compromised. In order to study the effects of surface blasting on adjacent underground workings, peak particle velocity (PPV) had been monitored at various locations in an underground Bord and Pillar mine. The monitored data were analyzed to develop a predictor equation of PPV based on distance (R) from the source of blasting and charge (explosive quantity) per delay (Q). The study finds that Q factor, generally used in predictor equations of PPV, may not be applicable while estimating PPV at an underground location. Depending on the rock strata through which stress waves transmits from the surface to underground, 0.35 Q factors may be appropriate while estimating the scale distance which is considered as a measure of PPV. This paper describes monitoring of ground vibration at underground locations and data analysis of PPVs using regression and neural network techniques. The paper also compares the results obtained using these two techniques.
There are several underground mines in India which operate in close proximity to an operating surface mines. Under such working conditions, there exists an immense danger to the safety and stability of underground mine openings and various underground structures viz. coal pillars, roof, water dams, ventilation and isolation stoppings situated in close proximity to operating surface mines. The prediction of the peak vibration level normally measured by Peak Particle Velocity (PPV) at the various underground structures caused by neighbouring surface mine blasting is pivotal for the safety of underground structures in terms of pillar spalling, roof collapse and junction failure. In addition, the prediction of vibration will help the surface mine operators to carry out optimum size controlled surface blasting with regard to the safety of the underground mine structures. The attenuation of vibration depends mainly on the charge weight, frequency content of wave motion and geotechnical properties of the transmitting medium. The interrelationship between the charge weight, distance and the amplitude of the motion form the basis of describing an attenuation law. Over the years, several predictor equations (attenuation law) of PPV were developed based on quantity of charge per delay and distance from the source of blasting (Langefors and Kihlstrom, 1958, Duvall and Fogelson, 1962). In most of the predicator equations, square root of charge per delay is assumed to be related with scale distance which is a measure of PPV. A few studies have been conducted to forecast PPVs at an underground location due to surface blasting. For the same charge and distance, attenuation of particle vibration (acceleration and velocity) is more at an underground point as compared a surface point where free surfaces are available (Fourie and Green,1993, Singh 2000, Deb and Jha, 2010, Jha, 2010).