The distribution of surface normal deviations from an ideal shape on concentrating solar collector surfaces is used to define the optical quality of a reflector. These distributions can be modeled by considering them to ideally follow a two-dimensional, circular Gaussian distribution. However, the measurement of these deviations in experimental systems usually defines only that component of the surface normal that deviates from the ideal surface normal direction, and ignores the rotational component of the normal vector in the plane perpendicular to the ideal direction. To compare the measured one-dimensional radial distribution with the expected two-dimensional model, the authors must transform the two-dimensional model into the appropriate radial distribution. The following analysis describes this transformation, and presents results gained from an application of the analysis to measured surface normal data from a mirror panel used in the reflecting surface of the 400 m{sup 2} paraboloidal (Big Dish) concentrator constructed at the ANU.
[1]
M. Crowder.
The Engineering Statistician's Guide to Continuous Bivariate Distributions
,
1993
.
[2]
Daniel Feuermann,et al.
Analysis of a two-stage linear Fresnel reflector solar concentrator
,
1991
.
[3]
Tara C. Kandpal,et al.
Concentration characteristics of a two stage solar concentrator: Effect of primary mirror surface errors
,
1992
.
[4]
Mark R. Shortis,et al.
Photogrammetry: An Available Surface Characterization Tool for Solar Concentrators, Part I: Measurements of Surfaces
,
1996
.
[5]
V. J. Romero,et al.
CIRCE2/DEKGEN2: A software package for facilitated optical analysis of 3-D distributed solar energy concentrators
,
1991
.