Optimization of Centrifugal Fan Design Using Genetic Algorithm and CST Method

Abstract Centrifugal fans used in constrained spaces, particularly in civil aviation ovens, are expected to deliver high performance. Centrifugal fans operate by initially drawing air axially, then redirecting it 90 degrees before it passes through the blades, where it is guided in a spiral pattern. This redirection, coupled with centrifugal forces, significantly influences airflow direction. One notable feature is the higher total pressure they generate, making them ideal for high-pressure applications. In traditional design approaches, meeting performance expectations often relies on trial-and-error methods and certain dogmatic assumptions. This study aims to identify and optimize the geometric and design parameters that influence the performance of centrifugal fans used in civil aviation ovens. To achieve this, the research integrates parameterization techniques based on the Class-Shape Transformation (CST) method with optimization methods grounded in Genetic Algorithms (GA), complemented by fluid analysis. The optimization process utilized the Class-Shape Transformation method due to its precision in capturing complex geometries with fewer variables. Numerical simulations were conducted using ANSYS Fluent, and performance data were recorded for further analysis. A communication loop between MATLAB Genetic Algorithm Toolbox and ANSYS Fluent was established, and relevant parameters were defined to conduct the necessary calculations for optimizing the centrifugal fan design.