Premium
In‐Line Monitoring of Crystallization Processes Using a Laser Reflection Sensor and a Neural Network Model
Author(s) -
Giulietti M.,
Guardani R.,
Nascimento C.A.O.,
Arntz B.
Publication year - 2003
Publication title -
chemical engineering and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.403
H-Index - 81
eISSN - 1521-4125
pISSN - 0930-7516
DOI - 10.1002/ceat.200390039
Subject(s) - crystallization , suspension (topology) , materials science , dispersity , particle (ecology) , particle size distribution , chord (peer to peer) , particle size , analytical chemistry (journal) , reflection (computer programming) , artificial neural network , optics , chemical engineering , chromatography , chemistry , physics , mathematics , computer science , engineering , distributed computing , oceanography , homotopy , polymer chemistry , pure mathematics , programming language , geology , machine learning
Laboratory‐scale experiments were carried out for measuring the chord length distribution of different particle systems using a laser reflection sensor. Samples consisted of monodisperse, polydisperse and bimodal FCC catalyst and PVC particles of different sizes, ranging from about 20 to 500 μm. The particles were dispersed in water, forming suspensions with solid‐phase mass fractions ranging from ca. 0.2 % until ca. 30 %. The experimental results, consisting of the particle number counting per chord length class, were used in fitting a neural network model for estimating the mass concentration of particles in the suspension and the volume‐based size distribution, eliminating the effects of suspension concentration and particle shape. The results indicate the feasibility of using such a model as a software sensor in crystallization processes monitoring.