Integrated optimization for design and operation of turbomachinery in a solar-based Brayton cycle based on deep learning techniques

Yuqi Wang, Tianyuan Liu, Yue Meng, Di Zhang and Yonghui Xie

Energy, 2022, vol. 252, issue C

Abstract: The design efficiency and operating conditions of turbomachinery are important for ensuring the reliability of an energy system. A data-driven design and operation optimization network for turbomachinery is proposed, which can also present the physical field distributions. The blade design optimization and the off-design operation optimization are integrated aiming at the turbine in a solar-based Supercritical Carbon Dioxide Brayton cycle. For prediction effect of the network, the mean relative errors of the trained Dual Convolutional Neural Network for field and performance prediction are mostly less than 0.75% and 3%, respectively. The R-squared of all concerned parameters are above 0.95. For multi-objective optimization process, Automatic Differentiation method is used to obtain the Pareto solutions. The blade design is conducted and the operation of the turbine is optimized under the three scenarios of insufficient pressure ratio, insufficient heat source, and insufficient rotation speed. The field distributions are acquired for different operation schemes. The presented network only needs 0.028s to obtain the turbine performance of a certain condition in comparison with 6min using Computational Fluid Dynamics method. Users can select a certain operation condition according to efficiency and power requirements, thus adapting to a wide range of off-design conditions.

Keywords: Turbomachinery; Off-design operation optimization; Field and performance prediction; Automatic differential method; Convolutional neural network (search for similar items in EconPapers)
Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (3)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544222008830
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:252:y:2022:i:c:s0360544222008830

DOI: 10.1016/j.energy.2022.123980

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().