Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Modelling and simulation of parabolic trough plants based on real operating data

: Rohani, S.; Fluri, T.; Dinter, F.; Nitz, P.

Fulltext urn:nbn:de:0011-n-4774935 (2.5 MByte PDF)
MD5 Fingerprint: 6dbc73027232d88b5498d1bab32d2830
(CC) by-nc-nd
Created on: 2.2.2018

Solar energy 158 (2017), pp.845-860
ISSN: 0038-092X
ISSN: 0375-9865
Journal Article, Electronic Publication
Fraunhofer ISE ()
Solarthermie und Optik; Solarthermie; Materialforschung und Optik

In recent years, parabolic trough power plants have been built in a large number and on a very large scale. Operating data from these plants are rarely published, however. In the present work detailed operating data from a commercial parabolic trough power plant (Andasol 3, Guadix, Spain) are analysed and compared with simulation results. Ten consecutive days of three different months (July, September and November) representing different seasons were selected. The simulation results have been obtained using the Fraunhofer in-house tool ColSimCSP. In the analysis, special attention is given to the operating strategy e.g. storage discharge strategy and start time of focusing for the various seasons. Understanding the real operating boundaries for more realistic yield estimation by implementing them in the performance model is a novelty of this work.
The existing simulation model in ColSimCSP is expanded and further developed according to the results of the evaluation of measured data. A detailed comparison on component level is performed between simulation and measured data for plant performance parameters such as solar field thermal power and net electrical energy yield. The lowest mean deviation is seen in solar field thermal power (0.59%), which shows the accuracy and reliability of the solar field modelling approach. The mean deviation of the net electrical energy yield is 2.29%, which still shows a relatively good match between simulation result and operating data.
Since the validation is performed against a wide range of real operating data of 30 days from a variety of months and days with different weather conditions, the results are highly relevant for developing large-scale concentrating solar power (CSP) plant models and validating related modelling and simulation software.