Neural network model for the on-line monitoring of a crystallization process

Abstract

This paper presents the results of the application of a recently developed technique, based on Neural Networks (NN), in the recognition of angular distribution patterns of light scattered by particles in suspension, for the purpose of estimating concentration and crystal size distribution (CSD) in a precipitation process based on the addition of antisolvent (a model system consisting of sodium chloride, water and ethanol). In the first step, in NN model was fitted, using particles with different size distributions and concentrations. Then the model was used to monitor the process, thus enabling a fast and reliable estimation of supersaturation and CSD. Such information, which is difficult to obtain by any other means, can be used in the study of fundamental aspects of crystallization and precipitation processes

A Physics-Assisted Data-Driven Framework for Predicting Crystal Chord Length Distributions under Arbitrary Experimental Conditions

“…In particular, artificial neural network (ANN) models have been widely utilized to convert CLDs to PSDs for different types of particles. 1,9,10,13 In reference 1, an ANN was used to solve the forward problem for needle-shaped crystals by transforming a two-dimensional (2D) PSD to a CLD. The data utilized to train the ANN was generated by using first-principles, geometric simulations of high aspect ratio crystals.…”

“…More recently, there has been a growing body of work on the use of data-driven methods for discovering the aforementioned mappings between PSDs and CLDs, casting the forward and inverse problems as learning ones. In particular, artificial neural network (ANN) models have been widely utilized to convert CLDs to PSDs for different types of particles. ,,, In reference , an ANN was used to solve the forward problem for needle-shaped crystals by transforming a two-dimensional (2D) PSD to a CLD. The data utilized to train the ANN was generated by using first-principles, geometric simulations of high aspect ratio crystals.…”

In‐Line Monitoring of Crystallization Processes Using a Laser Reflection Sensor and a Neural Network Model

“…Although forward laser diffraction methods have become increasingly popular due to reproducibility and easiness of operation, they cannot be used in measuring CSD in highly concentrated suspensions, unless the optical model is replaced with empirically fitted models based on neural networks [2,3], thus enabling the on-line monitoring of crystallization processes under certain conditions [4].…”