Solar Extinction Measurements
Upscaling the power of STPs is conceptually straightforward; however, a couple of pitfalls have to be avoided. In a best-case scenario, all of the direct irradiation from sunlight captured by the heliostats would be utilized for power generation. A complication of increasing the size of STPs is a comparative increase in solar extinction between the heliostats and the receiver, which limits the efficiency of energy collection. In modern, high power STPs, the heliostat-receiver distance may exceed 1 kilometer, so the solar extinction caused by local parameters such as humidity or particle concentration (i.e. dust) has to be considered when assessing STP performance.
Established methods to analyze solar extinction aim to determine the transmission or scattering coefficients at the local site. There are significant drawbacks to using these methods, including the comparably small measurement volume, or the requirement to use monochrome light sources. Deriving the solar extinction – as it is occurring in STPs – from such measurements would therefore require extrapolation in spatial and spectral dimensions, which can create large errors.
A new method proposed by Ballestrin et al, employs scientific cameras (ORCA-Flash4.0) as detectors to measure solar extinction. By using the solar signal and a broadband detector, spectral extrapolation can be avoided, yielding more accurate measurements of solar extinction. A simplified diagram of the measurement set up is shown in Figure 2.
Fig. 2: A Lambertian target, consisting of a white part of high solar reflectance and a black part of high solar absorbance is observed by 2 cameras simultaneously from different distances. The white part of the target serves as the signal source, while the black part serves as a dark reference to consider the effect of scattered light into the measurement distance. Both cameras observe the same area on the target.
Taking simultaneous images of a Lambertian target (a target with uniform scattering properties, as determined in [Ballestrin 2018] with two cameras at different distances, the solar extinction can be derived by the following formula:
With I2 being the intensity observed by camera 2 and I1 by camera 1, the areas observed on the target are identical for both cameras. By knowing the distance ‘D’ between both cameras, the extinction coefficient can be determined using the Beer-Lambert law. From this coefficient, solar extinction can be obtained for each particular heliostat receiver distance applying the law. In the system developed at PSA, the distance ‘D’ is 742 m which is representative in a low extinction environment such as PSA. Experimental preconditions for employing this method are the availability of Lambertian targets of high homogeneity and diffusivity, as well as highly linear cameras.
It is desirable to use this method in the process of determining appropriate locations for STPs, as well as during operation, to include the effect of solar extinction into the routine assessment of efficiency (see fig. 3).
Fig 3.: Solar extinction is routinely measured at PSA
The method has been applied over one year at Plataforma Solar de Almería, Spain. The results in figure 4 indicate that the solar extinction reached its minimum value and minimum variability during the winter months, while the absolute extinction and its variability peaked during the summer months [Ballestrin 2019].
Fig. 4: Solar extinction measured using the described method at Plataforma Solar de Almeria over one year