Session: 11-02-03: CFD Applications for Optimization and Control II

Paper Number: 166645

Design, Simulation and Optimization of a Centrifugal Fan 

The analytical design, numerical simulation, and optimization of centrifugal fans are essential for improving the efficiency and performance of ventilation, cooling, and other air-moving systems across various industries. These fans are widely used in applications such as HVAC systems, industrial processing, and automotive engineering, where precise aerodynamic performance and energy efficiency are critical. The flow behavior within centrifugal fans is complex, characterized by three-dimensional, viscous, unsteady, and turbulent phenomena. These complexities pose significant challenges in accurately predicting performance characteristics, making the development of advanced design methodologies essential.

Traditionally, centrifugal fan design has relied on a combination of analytical, physics-based methods and empirical correlations derived from experimental data. While these methods provide valuable insights, they often require extensive prototype testing to refine the design and ensure compliance with performance criteria. Experimental validation, although crucial, is time-consuming and costly, often necessitating multiple iterations before achieving the desired performance. However, the advent of computational fluid dynamics (CFD) and the rapid advancement of high-performance computing (HPC) have revolutionized the design process by offering a highly efficient and accurate alternative to physical testing.

This study focuses on the numerical simulation and optimization of centrifugal fans using the commercial solver ANSYS CFX 2024R1 on an HPC-cluster (High-Performance Computing). A detailed grid independence study based on Richardson extrapolation is conducted to ensure that the numerical results are mesh-converged and sufficiently accurate for further analysis. The turbulence modeling approach is carefully selected to capture the complex flow structures within the fan, ensuring that both steady-state and transient flow phenomena are accurately represented. Special attention is given to secondary flow structures, separation regions, and pressure recovery zones, which are critical for optimizing fan efficiency.

To enhance fan performance, an optimization approach combining Design of Experiments (DOE), Response Surface Methodology (RSM), and Multi-Objective Genetic Algorithms (MOGA) is employed. This framework enables the systematic exploration of design parameters while accounting for multiple conflicting objectives, such as maximizing aerodynamic efficiency and minimizing losses. By integrating these methodologies, an improved design is identified that outperforms traditional empirically derived geometries.

The final stage of this study involves the experimental validation of simulation and optimization results. A physically manufactured prototype is tested according to ISO 5801:2018, ensuring that the numerical predictions align with real-world performance data. This validation step is crucial for confirming the reliability of CFD-based optimization and its applicability in industrial design processes.

The integration of CFD-based simulation and optimization techniques significantly enhances centrifugal fan development. By reducing the reliance on extensive physical prototyping, this approach accelerates the design process, minimizes development costs, and enables the creation of highly efficient, high-performance fan systems that meet the increasing demands of modern engineering applications.

Presenting Author: Philipp Epple Coburg University of Applied Sciences

Presenting Author Biography: Full Professor in Fluid Mechanics

Authors:

Manuel Fritsche Coburg University of Applied Sciences
Philipp Epple Coburg University of Applied Sciences
Ivana Milanovic University of Hartford