Abstract
Solid can be divided into three main categories such as bulk, particles and molecules. However, the particle productions, as well as distribution of size, are very crucial to achieve partic-ular properties for any specific industrial uses. This research aim is to develop empirical relations using an artificial neural network (ANN) for the sonofragmentation experimental results in terms of number fractions for different lengths and widths of particles corresponding to ultrasonic ampli-tude equal to 10%. For the ANN model, three and four hidden layers are chosen for lengths and widths models, respectively. Then it empirically calculated the number of fractions for different lengths and widths of a particle was compared with the population balance equation (PBE). To fur-ther strengthen the empirical expression, the ANN result was also compared with the population balance model and found to be agreed well with the PBE simulations. The model was trained and validated using the measured (experimental) data of the number fraction. The model was found to be valid and reliable, with R2 equal to or greater than 0.95 in both training and validation.(c) 2023 The Author(s). Published by Elsevier B.V. on behalf of King Saud University. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).