Theoretical and experimental studies on the optimization of finger seal

Tools

Zhang, Yanchao and Yin, Minghu and Zeng, Quanren and Wang, Rui and Zhao, Tong (2020) Theoretical and experimental studies on the optimization of finger seal. Journal of Advanced Mechanical Design, Systems, and Manufacturing, 14 (7). JAMDSM0107. ISSN 1881-3054 (https://doi.org/10.1299/jamdsm.2020jamdsm0107)

[thumbnail of Zhang-etal-JSME-2020-Theoretical-and-experimental-studies-on-the-optimization-of-finger-seal]
Preview
 Text. Filename: Zhang_etal_JSME_2020_Theoretical_and_experimental_studies_on_the_optimization_of_finger_seal.pdf
Accepted Author Manuscript
Download (4MB)| Preview

Abstract

For the finger seal, the impact of structural parameters on its performance is complicated, and it is difficult and contradictory to lower its leakage and wear synchronously. Therefore, the problems relating to decision-making of optimal results and selection of reference variables still confuse the optimization of finger seal. In this paper, the performance (leakage and wear) of finger seal are calculated with finite element simulation, and a sensitivity analysis is carried out based on grey correlation method to obtain the influence degrees of structural parameters on the performance of finger seal; then a multi-objective optimization model is proposed by the combination of genetic algorithm, BP neural network and Nash equilibrium game theory; finally, the experiments of finger seals with various optimal structures are carried out to verify the theory and method established in this paper. The theoretical and experimental results indicate that the finger seal optimized with the model proposed in this paper has a better synthetical performance, which proves its feasibility. The theory and method established in this paper can be applied to enhance the synthetical performance of finger seals effectively.

ORCID iDs

Zhang, Yanchao, Yin, Minghu, Zeng, Quanren ORCID logoORCID: https://orcid.org/0000-0002-9955-9669, Wang, Rui and Zhao, Tong;
CORE (COnnecting REpositories)