2024 : 4 : 30
Jafar Nejadali

Jafar Nejadali

Academic rank: Assistant Professor
ORCID:
Education: PhD.
ScopusId:
Faculty: Faculty of Technology and Engineering
Address: University of Mazandaran
Phone: 011-35305108

Research

Title
Shape optimization of regenerative flow compressor with aero‑foil type blades using response surface methodology coupled with CFD
Type
JournalPaper
Keywords
Regenerative compressor; Shape optimization; Aerofoil blades; Blade angles; Response surface method
Year
2021
Journal STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION
DOI
Researchers Jafar Nejadali

Abstract

Regenerative flow compressors or blowers are rotodynamic machines with the ability to produce high head at low flow rates. This paper describes shape optimization of the aerofoil type blades for the regenerative compressor. A regenerative compressor with accessible experimental data was modeled as a base geometry. A three-dimensional numerical analysis combined with response surface method has been performed to determine the optimized shape. The polynomial-based RSM having a second order was applied. Isothermal efficiency and Pressure coefficient were selected as objective functions. The RSM is adopted with two design variables comprising blade inlet and blade outlet angles. The Preliminary numerical simulations for obtaining the optimum number of blades revealed that the highest efficiency occurs for the pitch to chord ratio (solidity) of 0.36. For isothermal efficiency, the optimal positions of β1 and β2 occurred at 33.8° and 50.7°, respectively. The results of sensitivity analysis showed that both design variables have the same sensitivity on the efficiency but the outlet angle (β2) is more impactful on the pressure coefficient as compared to that of the inlet blade angle (β1). Throughout the optimization, the isothermal efficiency for the optimized model at the design flow coefficient was enhanced up to 3.17% compared to the reference one. In addition, the pressure coefficient for the optimal RFC was successfully increased up to 8% compared with that of the reference model.