Publisher DOI: 10.1063/1.4984546
Title: Development of a CSP plant energy yield calculation tool applying predictive models to analyze plant performance sensitivities
Language: English
Authors: Haack, Lukas 
Peniche, Ricardo 
Sommer, Lutz 
Kather, Alfons 
Issue Date: 27-Jun-2017
Publisher: AIP Publishing
Source: AIP Conference Proceedings (1850): 160012 (2017-06-27)
Abstract (english): 
At early project stages, the main CSP plant design parameters such as turbine capacity, solar field size, and thermal storage capacity are varied during the techno-economic optimization to determine most suitable plant configurations. In general, a typical meteorological year with at least hourly time resolution is used to analyze each plant configuration. Different software tools are available to simulate the annual energy yield. Software tools offering a thermodynamic modeling approach of the power block and the CSP thermal cycle, such as EBSILONProfessional®, allow a flexible definition of plant topologies. In EBSILON, the thermodynamic equilibrium for each time step is calculated iteratively (quasi steady state), which requires approximately 45 minutes to process one year with hourly time resolution. For better presentation of gradients, 10 min time resolution is recommended, which increases processing time by a factor of 5. Therefore, analyzing a large number of plant sensitivities, as required during the techno-economic optimization procedure, the detailed thermodynamic simulation approach becomes impracticable. Suntrace has developed an in-house CSP-Simulation tool (CSPsim), based on EBSILON and applying predictive models, to approximate the CSP plant performance for central receiver and parabolic trough technology. CSPsim significantly increases the speed of energy yield calculations by factor ≥ 35 and has automated the simulation run of all predefined design configurations in sequential order during the optimization procedure. To develop the predictive models, multiple linear regression techniques and Design of Experiment methods are applied. The annual energy yield and derived LCOE calculated by the predictive model deviates less than ±1.5 % from the thermodynamic simulation in EBSILON and effectively identifies the optimal range of main design parameters for further, more specific analysis.
Conference: 22nd International Conference on Concentrating Solar Power and Chemical Energy Systems, SolarPACES 2016 
URI: http://hdl.handle.net/11420/4447
ISBN: 978-0-7354-1522-5
ISSN: 1551-7616
Institute: Energietechnik M-5 
Document Type: Chapter/Article (Proceedings)
Project: Transientes Verhalten gekoppelter Energienetze mit hohem Anteil Erneuerbarer Energien 
Part of Series: 
Volume number: 1850
Appears in Collections:Publications without fulltext

Show full item record

Page view(s)

251
Last Week
2
Last month
11
checked on May 29, 2023

Google ScholarTM

Check

Add Files to Item

Note about this record

Cite this record

Export

Items in TORE are protected by copyright, with all rights reserved, unless otherwise indicated.