Bubble size measurement algorithm for optical fiber probe measurement.
The aim of this project is to determine the optimal algorithm for processing the optical probe signals using Visual Basic. The optimised algorithm uses level threshold to differentiate noise signal from bubble signal. Bubble rise and end times are determined using both level and slope threshold. Sub...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
2009
|
Subjects: | |
Online Access: | http://hdl.handle.net/10356/16488 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Summary: | The aim of this project is to determine the optimal algorithm for processing the optical probe signals using Visual Basic. The optimised algorithm uses level threshold to differentiate noise signal from bubble signal. Bubble rise and end times are determined using both level and slope threshold. Subsequently, cross-correlation of the top and bottom probe signals is done based on setting a tolerance level of ± 30% for the difference in the residence times of the bubbles at the top and bottom probe and their chord lengths and bubble velocities are calculated by the algorithm. Manual elimination of the outliers is recommended by identifying them based on the graph of bubble velocities versus chord lengths. The developed algorithm was validated at gas flow rates of 0.588 cm/s and 1.88 cm/s by comparing bubble sizes obtained using a high speed camera to determine the accuracy and reliability of the algorithm. As the algorithm gives chord length distribution, numerical transform from chord length to bubble size distribution as proposed by N.N Clark et al. was used prior to any comparison made. Comparisons were made based on horizontal bubble diameter distribution and the corresponding mean horizontal bubble diameter and most probable horizontal diameter. The optimised algorithm gives a similar probability distribution as that obtained by the high speed camera and a percentage discrepancy of ± 23% and ± 25% for the mean horizontal bubble diameter and most probable bubble diameter respectively. |
---|