A User's Manual for DELSOL3 - prod.sandia.gov - Sandia National ...

epresentation of heliostat errors and to incorporate analytical scaling of the im-
ages as the tower height is varied (Reference 5). DELSOL also employs a method
for optimizing heliostat densities similar to the Houston approach (Reference 4).
The primary difference in the two codes is in their design/optimization capabili-
ties. The Houston approach considers only one tower height and receiver size at
a time. These variables must be optimized by manually rerunning the Houston
codes until an optimum is located. In contrast, DELSOL automatically optimizes
the tower height and receiver dimension(s), saving considerable user and com-
puter time. (The user is cautioned, however, to provide his own values for the
appropriate input variables if his system of interest differs significantly in size or
cost/performance from the default system description in the code. See Appendix
Am)
DELSOL is a FORTRAN IV code, developed on Sandia’s CDC6600 and
CDC7600, and adapted to execute on a CRAY-1. Typical execution times for
performance calculations on the CRAY-1 are 45-60 seconds, while design and op-
timization calculations may take 5-300 seconds. Execution times on a VAX 11-
780 will typically be about ten times as long as on the CRAY-1. This manual de-
scribes the status of DELSOL as of January 1986. The code is intended to evolve
with the development of central receiver technology and revised versions of the
code and manual will be released as needed.
LA. Differences from Earlier Versions of DELSOL
Several corrections and additions have been made in DELSOLS which dis-
tinguish it from its predecessors. Known errors which were fixed included cor-
recting the logic for scaling flux point positions during the optimization of cavity
receivers, correcting the flux calculation for cavity receivers and extending the
calculation to allow general flux maps in front of the aperture, and correcting the
logic when operating from storage and receiver power simultaneously. Also, the
operating parasitic loss calculation was fixed and modified, and numerous typo-
graphical errors in the code and in the outputs were corrected.
At least four major enhancements were added to DELSOL3. First, receiver
loss algorithms were modified based on References 6 and 7 to more closely model
experimental and test facility data of receiver losses, including measured losses
at the Solar One facility near Barstow, California. Next, algorithms for non-
operating parasitic losses were added to the code, thus accounting for the en-
ergy which is consumed by a power plant at the times when power is not being
produced. Third, the sizing algorithm for a cavity heat absorber area, which de-
termines receiver cost, was modified so that the new area is based on aperture
height rather than on minimum and maximum field dimensions. The previous
algorithm resulted in an extremely large cavity which had very low incident flux
levels on the upper portion of the heat absorber area. This modification assumes
that some internal spillage is acceptable, and is intended to be an empirical model
based on present and past detailed cavity designs.
22