Grain size characterization of Ti-6Al-4V titanium alloy based on laser ultrasonic random forest regression.

Abstract

In this paper, a random forest regression (RFR) rain size characterization method based on a laser ultrasound technique is investigated to predict the grain size of titanium alloy (Ti-6Al-4V). The longitudinal wave velocity of the ultrasound signal and the attenuation coefficient at different frequencies are used as the input and the grain size is used as the output. An RFR algorithm was used to develop a grain size prediction model. Meanwhile, the grain size calculation model based on conventional scattering attenuation was established by calibrating the n value in the classical scattering theory using the attenuation coefficients at different frequencies of ultrasonic signals. The results show that the RFR algorithm is feasible for the grain size characterization of titanium alloys.