Use this URL to cite or link to this record in EThOS:
Title: Particle size distribution estimation of a powder agglomeration process using acoustic emissions
Author: Nsugbe, Ejay
ISNI:       0000 0004 7968 8732
Awarding Body: Cranfield University
Current Institution: Cranfield University
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Access from Institution:
Washing powder needs to undergo quality checks before it is sold, and according to a report by the partner company, these quality checks include an offline procedure where a reference sieve analysis is used to determine the size distributions of the powder. This method is reportedly slow, and cannot be used to measure large agglomerates of powders. A solution to this problem was proposed with the implementation of real time Acoustic Emissions (AE) which would provide the sufficient information to make an assessment of the nature of the particle sizes. From the literature reviewed for this thesis, it was observed that particle sizes can be monitored online with AE but there does not appear to be a system capable of monitoring particle sizes for processes where the final powder mixture ratio varies significantly. This has been identified as a knowledge gap in existing literature and the research carried out for this thesis contributes to closing that gap. To investigate this problem, a benchtop experimental rig was designed. The rig represented limited operating conditions of the mixer but retained the critical factors. The acquired data was analysed with a designed hybrid signal processing method based on a time domain analysis of impact peaks using an amplitude threshold approach. Glass beads, polyethylene and washing powder particles were considered for the experiments, and the results showed that within the tested conditions, the designed signal processing approach was capable of estimating the PSD of various powder mixture combinations comprising particles in the range of 53-1500 microns, it was also noted that the architecture of the designed signal processing method allowed for a quicker online computation time when compared with other notable hybrid signal processing methods for particle sizing in the literature.
Supervisor: Starr, Andrew Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Signal processing ; process monitoring ; time domain ; online monitoring ; non-invasive monitoring ; amplitude threshold