Faculty of Engineering

Permanent URI for this community

https://scholar.sun.ac.za/handle/10019.1/10

The Faculty of Engineering at Stellenbosch University is one of South Africa's major producers of top quality engineers. Established in 1944, it currently has five Engineering Departments.

News

For the latest news click here.

Browse

Browsing Faculty of Engineering by Subject "3D Geometry morphing"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Mapping the relationship between the subjective response of jurors to fan blade parameters
    (Stellenbosch : Stellenbosch University., 2020-03) Chung, Sung bok; Muiyser, Jacques; Venter, Martin P.; Stellenbosch University. Faculty of Engineering. Dept. of Mechanical and Mechatronic Engineering.
    ENGLISH ABSTRACT: The noise generated by computer cooling fans is known to be a factor which may cause discomfort in the workplace. Aerodynamic noise caused by fan blades has been identified to be a significant factor which contributes largely to noise emission by fans. Fan parameters used in the design of fan blades are known to affect noise emission levels significantly. Considerations made for fan noise are known to be centred around taking unweighted and A-weighted sound pressure readings, with fan designs often being designed to limit sound pressure levels. The analysis made towards the unweighted and A-weighted sound pressure levels in the temporal and frequency domain alone is known to sufficiently define objective measures for fan noise. It is of interest to consider whether the subjective evaluation of noise made by jurors can also be accounted for. The presented work is focused on identifying the feasibility of mapping the relationship between fan blade parameters to the subjective response of jurors, using the interim step of psychoacoustic analysis to make predictions for an expected level of annoyance towards fan noise emissions. Three blade parameters were selected for investigation, consisting out of blade chord, blade angle and sweep; these are considered to have significant effects on noise emissions produced by fan blades. Fifteen prototype models with varied blade parameters were created to investigate the effect which blade parameters have on the subjective evaluation of noise. Prototype models were created by 3D scanning and recreating a reference commercial fan, altering its blade parameters by using mesh morphing techniques. Jury evaluation tests were used to collect data towards the perceived annoyance for fan noise using a forced pairwise comparison test, and a bipolar semantic differential test. The jury tests were used to obtain a ranking for prototype models, giving a relative measure of perceived preference, and also a description for noise, as experienced by jurors. Statistical analysis methods were used to fit simple, well-defined regression models, which were used to learn about the interaction effects present. One model, deemed adequate for making predictions on jury preferences, was selected and used for optimisation purposes. This model was optimised to determine the best expected performance, for a specific set of blade parameters. A final model using optimised blade parameters was 3D printed and used as a validation set to evaluate for the model. The optimised model was ranked 3rd in the final, forced pairwise comparison test, where 20 jurors were asked to partake in the test. The presented work determined the relationship between blade parameters, psychoacoustic metrics and data obtained from subjective jury test results to determine the feasibility of determining jury preference from the parameters of a fan blade alone.