CFD-Based Taguchi Optimization of Impeller Geometry to Improve Centrifugal Fan Efficiency: A Review of an Integrated Design Optimization Framework

on

Industrial centrifugal fans are widely used in ventilation, HVAC, and manufacturing systems, where energy efficiency directly influences operational costs and environmental sustainability. Because fan performance is strongly governed by impeller geometry, optimizing design parameters has become an important engineering challenge. Conventional optimization techniques often require numerous Computational Fluid Dynamics (CFD) simulations, resulting in considerable computational expense. The reviewed study proposes an integrated optimization framework that combines CFD simulations with the Taguchi design-of-experiments method to identify efficient impeller configurations while minimizing computational effort. By investigating the effects of inlet blade angle, outlet blade angle, blade number, and flow rate, the research demonstrates how statistical optimization can be effectively integrated with numerical simulation to improve centrifugal fan efficiency and support practical engineering design.

Bibliographic Information

Item Information
Article Title CFD-based Taguchi optimization of impeller geometry to improve centrifugal fan efficiency
Authors Delima Yanti Sari; Bagas Santoso; Hendri Nurdin; Hastuti; Rifelino; Fitrah Qalbina; Tsung-Liang Wu; Dani Harmanto
Journal Teknomekanik
Volume 9
Issue 2
Publication Year 2026
Pages 203–217
DOI https://doi.org/10.24036/teknomekanik.v9i2.53772
Publisher Universitas Negeri Padang
License Creative Commons Attribution 4.0 International (CC BY 4.0)
ISSN e-ISSN: 2621-8720
p-ISSN: 2621-9980
Keywords centrifugal fan; design configurations; HVAC systems; impeller geometry optimization

1. Research Background

Centrifugal fans are indispensable components in industrial ventilation and Heating, Ventilation, and Air Conditioning (HVAC) systems because they provide stable airflow under varying pressure conditions. As industries seek greater energy efficiency and lower operating costs, improving centrifugal fan performance has become an increasingly important engineering objective. One of the primary factors influencing fan efficiency is impeller geometry, which governs airflow characteristics, pressure development, and aerodynamic losses.

Previous studies have demonstrated that several geometric parameters—including the inlet blade angle (β1), outlet blade angle (β2), number of blades, and operating flow rate—substantially affect centrifugal fan performance. However, these parameters interact with one another rather than acting independently. Improper combinations may produce flow separation, recirculation zones, turbulence, and non-uniform pressure distributions, ultimately reducing overall efficiency.

Various optimization techniques such as Response Surface Methodology, Genetic Algorithms, and Particle Swarm Optimization have previously been employed to improve fan performance. Although these approaches are capable of identifying optimal solutions, they generally require numerous CFD simulations, resulting in high computational costs and extended processing times. Consequently, engineers continue to seek optimization methods that provide reliable results while requiring fewer numerical simulations.

To address this challenge, the reviewed study proposes an integrated optimization framework combining Computational Fluid Dynamics (CFD) with the Taguchi method. CFD serves as the numerical tool for evaluating aerodynamic performance, while the Taguchi method employs an orthogonal experimental design to systematically investigate multiple design parameters with a limited number of simulation runs. This integration aims to achieve efficient parameter optimization without sacrificing analytical reliability.

The research focuses on optimizing four major design variables: inlet blade angle, outlet blade angle, number of blades, and flow rate. Through statistical analysis using Analysis of Variance (ANOVA), the study evaluates the relative influence of each parameter and determines the combination that maximizes centrifugal fan efficiency.

Beyond proposing an optimization strategy, the study also contributes practical engineering knowledge regarding the interaction between impeller geometry and aerodynamic performance. By integrating numerical simulation with statistical design optimization, the research offers an efficient framework that can support future centrifugal fan development while reducing computational resources required during product design.

2. Research Objective

  • To develop an integrated optimization framework that combines Computational Fluid Dynamics (CFD) simulations with the Taguchi method for centrifugal fan design optimization.
  • To investigate the influence of four principal design variables—inlet blade angle, outlet blade angle, number of blades, and flow rate—on centrifugal fan efficiency.
  • To minimize computational cost by employing an L9 Taguchi orthogonal array instead of performing exhaustive CFD simulations.
  • To determine the statistical significance of each design parameter using Analysis of Variance (ANOVA).
  • To identify the optimal impeller geometry capable of improving centrifugal fan performance through enhanced aerodynamic behavior.

3. Why This Research Matters

  • Addresses industrial energy efficiency. Improving centrifugal fan efficiency directly reduces electrical energy consumption in industrial ventilation and HVAC systems.
  • Reduces computational cost. The integration of CFD with the Taguchi method significantly decreases the number of simulations required during engineering optimization.
  • Supports practical engineering design. The proposed methodology provides engineers with a structured and systematic approach for evaluating multiple interacting design parameters.
  • Improves aerodynamic understanding. The study demonstrates how impeller geometry influences airflow behavior, pressure distribution, and aerodynamic losses inside centrifugal fans.
  • Combines statistical and numerical engineering tools. Integrating CFD, Taguchi experimental design, and ANOVA creates an efficient workflow for engineering optimization.
  • Enhances industrial competitiveness. More efficient fan designs contribute to lower operating costs, improved equipment performance, and sustainable manufacturing practices.
  • Provides a transferable optimization framework. Although developed for centrifugal fans, the integrated methodology can potentially be adapted for optimizing other turbomachinery and fluid engineering systems.

4. Research Methodology

The study employed a quantitative engineering approach that integrates Computational Fluid Dynamics (CFD) with the Taguchi design-of-experiments method to optimize centrifugal fan impeller geometry. Rather than relying solely on repeated numerical simulations, the researchers combined statistical optimization with CFD analysis to evaluate multiple design variables efficiently while reducing computational cost. The overall workflow consisted of geometry construction, mesh generation, CFD simulation, Taguchi experimental design, statistical analysis using Analysis of Variance (ANOVA), and identification of the optimal design configuration.

Research Design

  • An integrated optimization framework combining CFD simulation and the Taguchi method.
  • Numerical performance evaluation followed by statistical optimization.
  • ANOVA employed to determine the significance of each design factor.

Model Construction

The centrifugal fan geometry was developed from a commercially available industrial fan using SOLIDWORKS. The numerical model included the impeller, casing, inlet duct, and outlet duct. The reference geometry consisted of an impeller diameter of 287 mm, ten blades, a blade width of 3 mm, an impeller width of 95 mm, an inlet diameter of 170 mm, and an outlet dimension of 153 × 130 mm. This baseline configuration served as the reference model for subsequent optimization.

Computational Fluid Dynamics (CFD) Simulation

The CFD simulations were performed using Ansys Student R1 2024 under steady-state conditions. The airflow was assumed to be incompressible with constant fluid properties and negligible gravitational effects. A velocity inlet corresponding to the selected operating flow rate and an atmospheric pressure outlet were imposed as boundary conditions. The impeller rotation was represented using the Rotating Reference Frame (RRF) method with a constant angular velocity of 301.069 rad/s.

To model turbulence, the realizable k-ε model with standard wall functions was adopted because of its suitability for rotating and separated flows. A pressure-based solver with second-order discretization schemes was used, and numerical convergence was achieved when all residuals fell below 10−5.

Mesh Validation

Prior to optimization, a mesh dependency analysis was conducted using four mesh densities. The numerical predictions were compared with reference data reported in previous studies to evaluate simulation accuracy. Mesh B demonstrated the smallest deviation from the reference solution while maintaining reasonable computational efficiency and was therefore selected for all subsequent simulations.

Optimization Parameters

Four design parameters were selected for optimization because of their influence on centrifugal fan performance:

  • Inlet blade angle (β1)
  • Outlet blade angle (β2)
  • Number of blades (n)
  • Flow rate (Q)

Each factor was investigated at three levels. Instead of evaluating every possible combination, the researchers employed a Taguchi L9 orthogonal array, enabling systematic exploration of the design space using only nine simulation cases.

Taguchi Optimization

The Taguchi method was implemented using the "Larger-the-Better" Signal-to-Noise (S/N) ratio because the objective was to maximize centrifugal fan efficiency. The orthogonal array allowed simultaneous investigation of multiple design variables while substantially reducing the computational effort typically required by conventional optimization methods.

Statistical Analysis

Analysis of Variance (ANOVA) was applied to quantify the contribution of each design parameter to fan efficiency. By partitioning the total variation into individual factor contributions, the researchers identified the relative importance of inlet blade angle, outlet blade angle, blade number, and flow rate in determining aerodynamic performance.

Performance Evaluation

The optimization process evaluated fan efficiency together with aerodynamic characteristics such as pressure distribution and velocity contours. CFD visualization was used to interpret flow behaviour under different impeller configurations, enabling the researchers to relate statistical optimization results to physical airflow phenomena inside the centrifugal fan.

5. Key Findings

CFD–Taguchi Integration Successfully Optimized Fan Performance

The integrated framework combining Computational Fluid Dynamics and the Taguchi method successfully identified an impeller configuration that substantially improved centrifugal fan efficiency while requiring only a limited number of numerical simulations. The results demonstrate that statistical experimental design can effectively complement CFD-based engineering optimization.

Substantial Improvement in Efficiency

The optimized impeller configuration increased centrifugal fan efficiency from 39.79% for the reference design to 63.26%. This considerable improvement indicates that appropriate combinations of geometric parameters can significantly reduce aerodynamic losses and enhance overall fan performance.

Design Parameters Influence Fan Performance Differently

ANOVA revealed that the investigated design parameters contributed differently to centrifugal fan efficiency. The statistical analysis enabled the researchers to identify which geometric variables exerted the greatest influence on performance, thereby providing quantitative guidance for future impeller design.

Improved Aerodynamic Flow Behaviour

The CFD simulations of the optimized configuration exhibited more favourable airflow characteristics than the reference model. Improved velocity distribution and pressure contours indicated reduced flow separation and lower aerodynamic losses, explaining the observed increase in efficiency.

Taguchi Method Reduced Computational Cost

By employing the L9 orthogonal array, the study evaluated four design variables at three levels using only nine simulation cases. This represents a substantially more efficient optimization strategy than exhaustive numerical experimentation while still producing statistically meaningful results.

Mesh Validation Improved Numerical Reliability

The mesh dependency analysis confirmed that the selected computational mesh produced stable and accurate simulation results. Selecting the appropriate mesh ensured that subsequent optimization outcomes reflected genuine design improvements rather than numerical artefacts.

Integrated Statistical and Numerical Analysis Strengthened Design Decisions

The combination of CFD visualization, Taguchi optimization, and ANOVA enabled the researchers to interpret both the physical mechanisms governing airflow and the statistical significance of design variables. This integrated methodology provides engineers with a robust framework for developing high-performance centrifugal fan designs.

6. Scientific Contribution

  • Introduces an integrated engineering optimization framework. The study demonstrates how Computational Fluid Dynamics (CFD) can be effectively integrated with the Taguchi design-of-experiments method to optimize centrifugal fan impeller geometry while substantially reducing computational effort.
  • Provides an efficient alternative to conventional optimization techniques. Unlike optimization methods that require extensive numerical simulations, the proposed approach employs an L9 orthogonal array, enabling systematic evaluation of multiple design variables with relatively few simulation cases.
  • Clarifies the interaction among key impeller design parameters. By simultaneously investigating inlet blade angle, outlet blade angle, blade number, and flow rate, the research illustrates how multiple geometric variables collectively influence aerodynamic performance rather than acting independently.
  • Integrates statistical and numerical engineering analysis. The combined application of CFD, Taguchi optimization, and Analysis of Variance (ANOVA) provides both physical interpretation of airflow behaviour and statistical evaluation of parameter significance.
  • Demonstrates the effectiveness of simulation-based engineering design. The optimized impeller configuration achieved a substantial improvement in efficiency while maintaining a systematic and reproducible optimization procedure.
  • Provides a transferable optimization methodology. Although developed for centrifugal fan design, the proposed framework can potentially be adapted to other turbomachinery and fluid engineering applications involving multiple interacting design variables.

7. Industrial Implications

  • Supports energy-efficient industrial ventilation systems. Improved centrifugal fan efficiency contributes directly to lower electricity consumption in HVAC installations and industrial ventilation networks.
  • Reduces engineering design time. The integration of CFD with the Taguchi method decreases the number of numerical simulations required during product development, allowing engineers to identify optimal designs more rapidly.
  • Lowers computational cost. The proposed optimization strategy minimizes computational resources compared with exhaustive simulation-based optimization approaches while maintaining reliable engineering results.
  • Improves product competitiveness. Manufacturers can use the optimization framework to develop centrifugal fans with higher aerodynamic efficiency, potentially reducing operating costs and increasing market competitiveness.
  • Supports digital engineering workflows. The methodology aligns with modern engineering practices that integrate CAD modelling, CFD simulation, statistical optimization, and numerical validation into a unified product development process.
  • Enhances sustainable manufacturing. More efficient centrifugal fan designs contribute to reduced energy consumption, improved equipment performance, and lower environmental impacts throughout operational service life.
  • Provides guidance for engineering design practice. The statistical evaluation of design parameters offers practical information that engineers can incorporate during impeller design and performance optimization.

8. Research Limitations

  • The optimization framework evaluates only four design variables: inlet blade angle, outlet blade angle, number of blades, and flow rate. Other geometric characteristics that may influence centrifugal fan performance were beyond the scope of the study.
  • The investigation is based exclusively on numerical simulations. Although mesh validation was conducted using published reference data, the study does not report experimental validation of the optimized impeller configuration.
  • The simulations were performed under steady-state operating conditions using predetermined boundary conditions. Fan performance under transient operating conditions was not investigated.
  • The optimization employed three levels for each design factor through an L9 orthogonal array. Additional parameter levels or wider design ranges may reveal alternative optimal configurations.
  • The computational model utilized a realizable k-ε turbulence model. Different turbulence models may produce slight variations in predicted aerodynamic characteristics.
  • The study focuses primarily on aerodynamic efficiency. Other engineering considerations such as structural integrity, manufacturing constraints, vibration characteristics, acoustic performance, durability, and lifecycle cost were not included in the optimization process.

9. Future Research Opportunities

  • Conduct experimental validation of the optimized impeller design to verify the numerical predictions under practical operating conditions.
  • Investigate additional geometric parameters, including blade curvature, blade thickness, impeller diameter, hub geometry, and casing configuration, to further improve centrifugal fan performance.
  • Compare the Taguchi optimization framework with other engineering optimization techniques, such as Genetic Algorithms, Particle Swarm Optimization, Response Surface Methodology, and machine learning-based optimization.
  • Extend the CFD analysis to transient operating conditions and variable rotational speeds to better represent real industrial applications.
  • Incorporate structural, vibration, acoustic, and thermal analyses into a multidisciplinary optimization framework for centrifugal fan design.
  • Evaluate the influence of different turbulence models and advanced numerical approaches on optimization accuracy.
  • Investigate optimization strategies for various centrifugal fan configurations used in industrial ventilation, HVAC systems, and other turbomachinery applications.
  • Integrate lifecycle cost analysis and energy consumption assessment to support economically and environmentally sustainable fan design.
  • Develop automated optimization workflows by combining CFD simulations with artificial intelligence or machine learning algorithms to accelerate engineering design.
  • Explore the applicability of the proposed optimization methodology to compressors, pumps, blowers, and other rotating fluid machinery with complex aerodynamic interactions.

10. Potential for Public Policy Citation

Although this study is primarily an engineering design investigation, it has meaningful potential to inform public policy related to energy efficiency, sustainable industrial development, and engineering innovation. Centrifugal fans are widely used in manufacturing facilities, commercial buildings, and HVAC systems, making improvements in their efficiency directly relevant to national energy conservation strategies.

The proposed integration of Computational Fluid Dynamics (CFD) with the Taguchi optimization method demonstrates a systematic approach to improving equipment performance while reducing computational resources during product development. Such methodologies align with broader governmental initiatives promoting industrial digitalization, advanced manufacturing, and sustainable engineering practices.

The findings may therefore serve as useful technical references for policymakers, engineering standard organizations, industrial innovation programs, and research funding agencies interested in supporting energy-efficient mechanical systems, simulation-driven engineering design, and environmentally responsible manufacturing technologies.

11. Who Should Read This Paper?

  • Mechanical engineers involved in centrifugal fan and turbomachinery design.
  • Researchers working in Computational Fluid Dynamics (CFD) and numerical simulation.
  • Industrial engineers responsible for HVAC and ventilation system optimization.
  • Graduate students studying fluid mechanics, turbomachinery, or engineering optimization.
  • Manufacturing companies developing energy-efficient industrial equipment.
  • Researchers applying Taguchi methods and Design of Experiments (DOE) in engineering.
  • Academics interested in simulation-based engineering design and statistical optimization.
  • Government agencies and engineering consultants involved in industrial energy-efficiency initiatives.

12. Final Thoughts

This study presents a well-structured engineering optimization framework that successfully integrates Computational Fluid Dynamics with the Taguchi design-of-experiments method to improve centrifugal fan efficiency. Rather than relying on computationally intensive optimization techniques, the proposed methodology demonstrates that statistically designed numerical experiments can effectively identify optimal impeller configurations while significantly reducing computational effort.

A notable strength of the research lies in its integration of numerical simulation, statistical optimization, mesh validation, and variance analysis into a coherent engineering workflow. The substantial improvement in fan efficiency illustrates the practical value of combining physical modelling with statistical design methods to solve complex engineering problems.

Beyond centrifugal fan optimization, the study provides a practical example of simulation-driven product development that can inspire similar applications across turbomachinery and fluid engineering disciplines. Its emphasis on efficient optimization, computational reliability, and engineering practicality makes this work a valuable contribution to modern mechanical engineering research.

13. Suggested Citations

Teknomekanik (UNP) Style

Sari DY, Santoso B, Nurdin H, Hastuti, Rifelino, Qalbina F, Wu TL, Harmanto D. CFD-based Taguchi optimization of impeller geometry to improve centrifugal fan efficiency. Teknomekanik. 2026;9(2):203–217. https://doi.org/10.24036/teknomekanik.v9i2.53772

APA (7th Edition)

Sari, D. Y., Santoso, B., Nurdin, H., Hastuti, Rifelino, Qalbina, F., Wu, T.-L., & Harmanto, D. (2026). CFD-based Taguchi optimization of impeller geometry to improve centrifugal fan efficiency. Teknomekanik, 9(2), 203–217. https://doi.org/10.24036/teknomekanik.v9i2.53772

IEEE Style

D. Y. Sari, B. Santoso, H. Nurdin, Hastuti, Rifelino, F. Qalbina, T.-L. Wu, and D. Harmanto, "CFD-based Taguchi optimization of impeller geometry to improve centrifugal fan efficiency," Teknomekanik, vol. 9, no. 2, pp. 203–217, 2026, doi:10.24036/teknomekanik.v9i2.53772.

Harvard Style

Sari, D.Y., Santoso, B., Nurdin, H., Hastuti, Rifelino, Qalbina, F., Wu, T.-L. & Harmanto, D., 2026. CFD-based Taguchi optimization of impeller geometry to improve centrifugal fan efficiency. Teknomekanik, 9(2), pp.203–217. https://doi.org/10.24036/teknomekanik.v9i2.53772

Vancouver Style

Sari DY, Santoso B, Nurdin H, Hastuti, Rifelino, Qalbina F, Wu TL, Harmanto D. CFD-based Taguchi optimization of impeller geometry to improve centrifugal fan efficiency. Teknomekanik. 2026;9(2):203-217. doi:10.24036/teknomekanik.v9i2.53772.

Chicago (Author–Date)

Sari, Delima Yanti, Bagas Santoso, Hendri Nurdin, Hastuti, Rifelino, Fitrah Qalbina, Tsung-Liang Wu, and Dani Harmanto. 2026. "CFD-based Taguchi Optimization of Impeller Geometry to Improve Centrifugal Fan Efficiency." Teknomekanik 9 (2): 203–217. https://doi.org/10.24036/teknomekanik.v9i2.53772.

MLA (9th Edition)

Sari, Delima Yanti, et al. "CFD-based Taguchi Optimization of Impeller Geometry to Improve Centrifugal Fan Efficiency." Teknomekanik, vol. 9, no. 2, 2026, pp. 203–217. https://doi.org/10.24036/teknomekanik.v9i2.53772.

14. Editorial Note

This review has been prepared exclusively from the scientific content presented in the published article and verified bibliographic metadata obtained from the official journal publication. The review summarizes the research objectives, methodology, principal findings, scientific contributions, and practical implications without introducing interpretations that extend beyond the original study. It is intended to provide researchers, engineers, educators, and industry practitioners with an accessible overview of the article while encouraging consultation of the original publication for complete technical details.

15. SEO Meta Description

Discover how CFD simulation and the Taguchi optimization method improve centrifugal fan efficiency through optimized impeller geometry, statistical design of experiments, and engineering analysis.

16. SEO Keywords

CFD optimization, Taguchi method, centrifugal fan, impeller geometry, computational fluid dynamics, engineering optimization, ANOVA, HVAC systems, fluid mechanics, turbomachinery, design of experiments, fan efficiency, numerical simulation, mechanical engineering

Comments

Post a Comment