Accuracy on the time-of-flight estimation for ultrasonic waves applied to non-destructive evaluation of standing trees: a comparative experimental study

Time-of-flight measurement is a critical step to perform ultrasonic non-destructive testing of standing trees, with direct influence on the precision of defect detection. Aiming to increase the accuracy on the estimation, the characteristics of the ultrasonic measurement chain should be adapted to the constraints of wood testing in living condition. This study focused on the excitation signal parameters, such as shape, temporal duration, and frequency response, and then the selection of a suitable time-of-flight determination technique. A standing plane tree was tested, placing ultrasonic receivers at four different positions, with five different excitation signals and three time-of-flight detection methods. The proposed ultrasonic chain of measurement resulted in high signal-to-noise ratios in received signals for all configurations. A time-frequency analysis was used to determine the power distribution in the frequency domain, showing that only chirp signal could concentrate the power around the resonant frequency of the sensor. Threshold and Akaike information criterion method performed similar for impulsive signals with decreasing uncertainty as sensor position approached to the radial direction. Those two methods failed to accurate determine time-of-flight for Gaussian pulse and chirp signals. Cross-correlation was only suitable for the chirp signal, presenting the lower uncertainty values among all configurations