OPTIMIZATION DESIGN OF HIGH-EFFICIENCY AXIAL FLOW FAN USING ARTIFICIAL NEURAL NETWORK ALGORITHM
Advances and Applications in Fluid Mechanics · 25.03.2026 · 33 (1) · 111-132
DOI: 10.17654/0973468626006 ↗Kurz gefasst
Fachartikel in der Zeitschrift Advances and Applications in Fluid Mechanics, online erschienen am 25.03.2026. Verfasst von Libo Wang und 3 weiteren Autor:innen.
Abstract (Originalsprache)
This study designs and optimizes a high-efficiency axial fan. The fan is designed to have a total pressure of 230 Pa at a rotational speed of 1100 rpm. Fourteen parameters were selected as optimization variables, including the stagger angle, chord length, and blade sweep/dihedral characteristics at the hub, mean, and tip blade sections. The initial sample space was generated using Latin hypercube sampling. A surrogate model was constructed using a neural network, with the highest efficiency at the design point as the optimization objective with a constraint of greater than 230 Pa total. The non-dominated sorting genetic algorithm II (NSGA-II) was used for optimization. The optimized fan had a total pressure of 234 Pa with an efficiency of 87.8%, representing a 17.96 Pa increase in pressure and a 2% improvement in efficiency compared with the initial fan.
Zitieren
APA
Wang, L., Cheng, K., Lv, Q., & Li, J. (2026). OPTIMIZATION DESIGN OF HIGH-EFFICIENCY AXIAL FLOW FAN USING ARTIFICIAL NEURAL NETWORK ALGORITHM. Advances and Applications in Fluid Mechanics, 33(1), 111-132. https://doi.org/10.17654/0973468626006
Vancouver
Wang L, Cheng K, Lv Q, Li J. OPTIMIZATION DESIGN OF HIGH-EFFICIENCY AXIAL FLOW FAN USING ARTIFICIAL NEURAL NETWORK ALGORITHM. Advances and Applications in Fluid Mechanics. 2026;33(1):111-132. doi:10.17654/0973468626006
BibTeX
@article{Wang2026,
author = {Wang, Libo and Cheng, Kai and Lv, Qian and Li, Jingyin},
title = {OPTIMIZATION DESIGN OF HIGH-EFFICIENCY AXIAL FLOW FAN USING ARTIFICIAL NEURAL NETWORK ALGORITHM},
journal = {Advances and Applications in Fluid Mechanics},
year = {2026},
volume = {33},
number = {1},
pages = {111--132},
doi = {10.17654/0973468626006}
}
EUVIMED