Abstract: The bubble flow is a flow regime present in various fields. A usual equipment that operates with the bubble flow is the bubble column. Many papers deal with determining flow regime, hold up, and mass transfer coefficient in a bubble column. However, little attention has been given to the measurement of the gas flowrate. In this context, this work presents a technique to predict the gas flowrate in a bubble column from ultrasound signals using a new set of ultrasonic variables combined with process variables using a feedforward artificial neural network (ANN). Experiments were carried out in a 2-meters-long-bubble column, with inner diameter of 52.5 mm using a single 2.25 MHz ultrasound transducer. The gas flow rate varied in the range of 0 to 4 liters per minute. The best architecture obtained for ANN has one hidden layer with 36 neurons. The test mean absolute error was 0,1 LPM Ver menos