University of Birmingham
Online monitoring of Rheology using Passive Acoustic Emission sensing with machine learning
Abstract
dc:description.abstractAdvancement in technological practices in the manufacturing industry continuously provides opportunities for producing more efficient and higher quality products. For industries such as personal care, home care, pharmaceutical and cosmetics that rely heavily on rheology for the manufacture of liquid products, these opportunities present in the implementation of real-time measurement and monitoring of rheology using Industry 4.0 principles. An experimental setup was developed using a passive acoustic emission sensor placed online on a continuous, closed loop flow system to collect acoustic data in real-time from the fluid flowing through the closed loop. The acoustic data was then processed using mathematical models and machine learning algorithm to develop rheological fingerprints specific to a distinct fluid under distinct process conditions. Solutions of Glycerol, Carboxymethyl Cellulose (CMC) and Carbopol® were chosen to represent Newtonian, Non-Newtonian and Non-Newtonian with yield stress respectively. Results of the acoustic data processing showed that visual monitoring of rheology can be achieved using the tools developed and a predictive model using Artificial Neural Network (ANN) can estimate rheological values for unknown fluid samples.
Degree
thesis:*- Name dc:type.qualificationname
- m_rs
- Level dc:type.qualificationlevel
- m_rs
- Grantor dc:publisher.institution
- University of Birmingham
- Year dc:date.issued
- 2022
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Ayinde-Tukur, Temilade
- ORCID person:identifier.orcid
Subjects
dc:subject × 4Rights
- Language dc:language.dcterms_iso639_2
- eng