Date of Award

Spring 1-1-2012

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Mechanical Engineering

First Advisor

John Daily

Second Advisor

Chuck Kutscher

Third Advisor

Keith Gawlik

Fourth Advisor

William Pfeffer

Fifth Advisor

Jean Hertzberg

Abstract

As the concentrating solar power industry competes to develop a less-expensive parabolic trough collector, assurance is needed that new parabolic trough collectors maintain accurate optical alignment. Previous optical characterization techniques are either too slow, ill-suited for field testing, or do not allow the collector to be tested in realistic orientations. The Observer method presented here enables the rapid optical characterization of parabolic trough collectors in any orientation in the field.

The Observer method directly measures the combined optical angular errors in the reflector surface shape and the absorber position, which can be separated into its two components: reflector surface slope and absorber misalignment. The data acquisition requires the placement of photogrammetry targets on and around the collector. Multiple photographs of the absorber and its reflection are taken with a digital camera from different angles with respect to the collector. The images are processed to determine the camera location of each image using photogrammetry bundle analysis. The absorber and its reflection are found in the photographs using image-processing techniques.

A Monte Carlo uncertainty model was developed to determine the uncertainty in the Observer measurements. The uncertainty was estimated for a wide array of measurement test scenarios to demonstrate the user's control over the measurement uncertainty. To validate the Observer method, the absorber alignment technique was compared to traditional photogrammetry; the absorber position measured with the two methods compared with a root-mean-square difference of 1.5 mm in the transverse direction and 0.86 mm along the optical axis. The reflector surface slope error measurement was compared to both VSHOT and SOFAST, two well-established optical characterization tools, by measuring a single reflector panel in the laboratory. The VSHOT and SOFAST measurements agreed with the Observer with a root-mean-square difference of 1.6 mrad and 2.1 mrad, respectively. In the field, the Observer method's capability to test collectors in any orientation was demonstrated by mounting the camera on a radio-controlled helicopter and measuring a collector oriented at 90 degrees above the horizon. The absorber measurement capability was demonstrated in the field for a collector facing both horizontally and vertically.

Share

COinS