Ultrasound waves (UT) propagating at a frequency of 2.25 MHz through the melt flow inside
an extruder die were used to monitor the density of the foam extrudate. Acoustic properties
of the polymer melt, manifested by the behavior of the ultrasound waves inside the die, were
empirically correlated to the density of the extruded foam. Wavelet packet analysis was used
to filter the ultrasound echo signals. A three layer multi-layer perceptron artificial neural
network was used to correlate the response of the ultrasound sensor to the measured foam …