| Autor: |
Jian Zhang, Min Zhang, Juan Du, Kai Yue, Xinyi Wang, Chen Yang, Hongwu Zhang |
| Jazyk: |
angličtina |
| Rok vydání: |
2024 |
| Předmět: |
|
| Zdroj: |
Propulsion and Power Research, Vol 13, Iss 4, Pp 534-552 (2024) |
| Druh dokumentu: |
article |
| ISSN: |
2212-540X |
| DOI: |
10.1016/j.jppr.2024.02.007 |
| Popis: |
Gas turbine is a promising device for power generation and propulsion either using traditional or renewable energy fuels. One of its key problems is the flow instability of compressors especially with the increase in blade load and changeable working environment. To intelligently and efficiently inhibit flow separation and enhance the pressure rise ability of highly loaded compressors under variable operating conditions, a novel flow control technique termed as adaptive Coanda jet control (ACJC) is proposed in this paper for a compressor stator cascade with a high diffusion factor of 0.66. To realize the ACJC strategy, an incidence angle (IA) prediction model and an optimal injection mass flow rate (OIMFR) prediction model are established by adopting single factor analysis of variance, principal component analysis and Back Propagation Neural Network (BPNN) methods. Two inlet Mach numbers including 0.1 and 0.4 are considered to represent incompressible and compressible flow conditions, and different inlet incidence angles are involved to model various off-design working situations of the real compressor. Effectiveness of the ACJC system is evaluated using numerical simulations are performed to understand the effects of the injection mass flow ratio on the flow field and aerodynamic performance of the blade cascade. Results indicate that the ACJC system can accurately predict the optimal injection mass flow ratio that can achieve the minimum flow loss at each incidence angle. Compared to the cascade without ACJC under the incidence angel of 5°, the optimal injection mass flow ratio being 1.27% and 1.20% can reduce the total pressure loss coefficient by 18.88% and 21.56% for incoming Mach number being 0.1 and 0.4, respectively. |
| Databáze: |
Directory of Open Access Journals |
| Externí odkaz: |
|
