Baseline Corrections for Mining Blast Vibrations

Blast vibration records used in mining contain unwanted information in the form of noise due to long-period drifts, background noise, and noise due to the conversion of the analogue signal to a digital signal. When recorded signals from blasting are used to find other physical variables, such as going from particle velocity to acceleration or displacement, the additional information in the vibration signal can result in inaccuracies in the estimation of the peak values and more dramatically, the results in displacement or acceleration can be nonzero near the end of the record. There are different proposed methods to clean or filter the unwanted information from the signal. All of the methods are based on finding the best fit of the record average value (baseline) and then subtracting from the record under analysis to produce the desirable adjustment. Different criteria and functions are used to define the best fit. The function to use will depend on the shape of the record average values. The main characteristic of the baseline function is that it represents a low frequency wave. Such a low frequency wave can be a periodic function or a polynomial expression of nth order. According to the order of the polynomial expression, the baseline correction can be linear, parabolic or high order. In this document the detailed procedure and the results for the baseline correction of a blast vibration record using a periodic function and a polynomial expression of different order is presented. The results show that the drift problem when displacement is calculate using velocity records increases with the increase of the distance from the blast.