CONFERENCE PROCEEDINGS
Advanced Search
Home > Proceedings > Volume PC12671 > Article
Presentation
5 October 2023 Reliable neural network heliostat calibration on low data
Max Pargmann, Moritz Leibauer, Daniel Maldonado Quinto, Vincent Nettelroth, Robert Pitz-Paal
Author Affiliations +
Proceedings Volume PC12671, Advances in Solar Energy: Heliostat Systems Design, Implementation, and Operation; PC1267106 (2023) https://doi.org/10.1117/12.2676698
Event: SPIE Optical Engineering + Applications, 2023, San Diego, California, United States
Abstract
Heliostat calibration is crucial for efficient solar tower power plant operation. Geometry-based models are reliable but yield moderate accuracy, while neural networks promise higher accuracy but need high amounts of data. We introduce a hybrid model that combines a reliable geometric model with a neural network disturbance model over a regularization sweep. Using real measurement data from the Jülich solar tower, we achieve higher accuracies than rigid body models starting from the first measurement, with a top performance below 0.7 milliradians. Our approach reduces data requirements of deep learning models, making them promising for heliostat calibration in solar tower power plants.
Conference Presentation
© (2023) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Citation Download Citation
Max Pargmann, Moritz Leibauer, Daniel Maldonado Quinto, Vincent Nettelroth, and Robert Pitz-Paal "Reliable neural network heliostat calibration on low data", Proc. SPIE PC12671, Advances in Solar Energy: Heliostat Systems Design, Implementation, and Operation, PC1267106 (5 October 2023); https://doi.org/10.1117/12.2676698
ACCESS THE FULL ARTICLE
ORGANIZATIONAL
Sign in with credentials provided by your organization.
INSTITUTIONAL
Select your institution to access the SPIE Digital Library.
SELECT YOUR INSTITUTION
PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.
PERSONAL SIGN IN
No SPIE Account? Create one
PURCHASE THIS CONTENT
SUBSCRIBE TO DIGITAL LIBRARY
50 downloads per 1-year subscription
Members: $195
Non-members: $335 ADD TO CART
25 downloads per 1 - year subscription
Members: $145
Non-members: $250 ADD TO CART
PURCHASE SINGLE ARTICLE
Includes PDF, HTML & Video, when available
Members: $17.00
Non-members: $21.00 ADD TO CART
PROCEEDINGS
 PRESENTATION
WATCH
PRESENTATION SAVE TO MY LIBRARY
GET CITATION
Advertisement
Advertisement
KEYWORDS
Calibration
Data modeling
Neural networks
Performance modeling
Solar energy
Education and training
Efficient operations
RELATED CONTENT
Subscribe to Digital Library
Receive Erratum Email Alert
Back to Top