Johan Utomo

Numerical studies of wavelet-based method as an alternative solution for population balance problems in a batch crystalliser

Abstract Abstract This paper deals with the numerical simulation studies of the nucleation and crystal growth process in a batch crystalliser. The population density functions may extend over orders of several magnitudes and the size distribution can be very sharp, thus accurate numerical solution of the population density functions can be challenging. The main interest for finding a more representative population balance solutions has motivated many researchers to develop many methods for the last four decades. In this paper, four methods have been implemented including wavelet-based method. The results obtained in three case studies, have demonstrated the wavelet scheme to be an alternative in providing accurate, fast and robust solutions.

CHALLENGES OF MODELLING A POPULATION BALANCE USING WAVELET

Abstract Crystallization is one of the oldest separation technologies due to its ability to produce a range of bulk products to high purity chemicals. Aspect of controlling the size distribution is important for downstream operations and characteristics of products. Two cases of population balance problems are considered in this paper to show limitations of some utilized methods. Those cases present the sharp transition phenomena in the particle size distribution. A wavelet-based method by Liu and Cameron (2001) is applied and compared with other conventional methods based on Finite Difference, Orthogonal Collocation and Orthogonal Collocation with Finite Elements. The result show that the wavelet method is faster, more accurate and more efficient in solving the population balance problems.