In solar power tower plants, thousands of tracking mirrors (heliostats) concentrate solar radiation on a receiver to produce electricity. An alignment error in these heliostats of only 2 mrad produces around 30% losses in the annual energy production of this type of plants.
Existing techniques for aligning (canting and focusing) heliostat facets generally are time consuming or too imprecise. This project proposes the combination of models based on geometric optics and computer vision techniques. It is intended to develop a new methodology that overcomes the limitations of the state of the art. The objective of this project is to develop and validate a new technique to determine, in a reasonable time and with precision, both alignment errors in heliostats with the aim of improving the efficiency of solar plants.
The proposed system is based on the combination of images and optical models to determine the alignment errors in the heliostat facets. As illustrated in the figure, a camera facing the heliostat to be aligned is placed at fixed points on the rear structure of a heliostat located in front of the former one. This way, in addition to the heliostat to be aligned, the camera sees in reflection the back of the heliostat on which it rests. By comparing the captured image (once the optical deformations of the camera have been treated and corrected) and the theoretical image that should be observed (if the heliostat were perfectly aligned), it is possible to determine the alignment errors, both for canting and focusig, in all its facets.
Existing heliostat alignment techniques can be classified into three categories: on-sun, mechanical, and optical. The following table summarizes the capabilities of existing techniques, in comparison with those developed under VISHELIO project.