Use this URL to cite or link to this record in EThOS:
Title: Automated reflection photoelasticity : digital data acquisition and use
Author: Zhang, Su Juan
ISNI:       0000 0001 2416 4446
Awarding Body: University of Sheffield
Current Institution: University of Sheffield
Date of Award: 2000
Availability of Full Text:
Access from EThOS:
Access from Institution:
Automation of reflection photoelasticity has simplified the stress and strain analysis of real engineering components and reduced analysis time. However, images obtained from automated reflection photo elasticity contain noise and accuracy of the analysis will be affected by the degraded intensity images. In the present study, major sources of noise in automated reflection photoelasticity have been found to be the photoelastic coating and the electronic instrumentation. An automated reflection polariscope PSIOS developed by Patterson and Wang (1998) for the simultaneous observation and capture of four phase-stepped photo elastic images was used as an example. The majority of the noise is in the high spatial frequency domain. The zero-phase, low pass Butterworth filter was found to be the most effective and flexible smoothing method for reducing the effect of noise in the intensity images. Results from experiments performed for assessing the ability of the PSIOS indicated that it is capable of yielding accurate results for the stress analysis of real components in both static and dynamic conditions and that it is fast and easy to use. Full-field experimental methods are often used to validate the stress distribution generated from numerical analysis. A common practice is to plot data along a line across the maps and to include both experimental and numerical results on the same axes. This approach is used widely and usually a reasonable, quantitative conclusion can be made. However, it cannot obtain more information about the relationship between the stress maps. Another method is to compare hot spots on experimental maps to the numerical maps. If the hot spots on the two maps match well, the numerical method is considered valid. However, when designs are being optimised for weight or crack paths are being investigated, comparison of the positions of the hot spots alone will not be enough and the correlation elsewhere in the data field should be taken into account. It has been shown that fit between the stress map from an experimental method and the stress map from the numerical analysis can be represented by a statistical parameter, the scaled standard deviation. An evaluation of the method was performed using stress maps from transmission photoelasticity, thermoelasticity and the finite element method as examples. The results from experiments using a curved tiebar, a circular ring and a real engineering component in this case, a race car hub carrier indicated that the scaled standard deviation represents the fitness between the two stress maps. If the scaled standard deviation is smaller than 0.1, then the experimental map and the numerical map can be considered to be in good agreement.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Stress analysis