Engineering Efficient Airflow with Simulation: Structural, Optimization & Acoustic Insights for Ceiling Fan Design

In Part 1 of this blog series, we explored how simulation using Ansys Fluent enables accurate airflow prediction and IS 374-compliant performance assessment for ceiling fans — helping engineers optimise designs for better energy efficiency and air delivery without costly physical prototyping.

In Part 2 below, we dive deeper into fluid–structure interactions, blade optimization, acoustics, and simulation automation to complete the digital design workflow.

Section 01Fluid–Structure Interaction (FSI) Analysis of Ceiling Fan Blades

During operation, ceiling fan blades are not only subjected to aerodynamic loads but also to self-weight and centrifugal forces. If blade design lacks structural capacity, these loads can cause undesired deformation or failure. Under fluid loading, blades deform — changing their shape and, consequently, the airflow patterns they generate. This creates a coupled behaviour where the modified flow affects further blade deformation, making it critical to capture these interactions early in the design cycle.

One-Way vs Two-Way FSI

• 1-Way FSI — Transfers pressure data from the fluid solver to the structural solver to evaluate deformation due to fluid loads. It does not capture how the newly deformed shape alters the airflow characteristics.
• 2-Way FSI (System Coupling) — Through Ansys System Coupling, Ansys Fluent and Ansys Mechanical exchange data iteratively, predicting blade deformation under equilibrium conditions that include fluid pressure, self-weight, and centrifugal forces — offering a more realistic simulation of operational behaviour.

Section 02Shape Morphing Using Gradient-Based Optimisation

Designers seek blade profiles that maximise airflow and efficiency while maintaining structural integrity. Traditional parametric optimisation (e.g., DOE with Ansys optiSLang) can require significant computational resources. Gradient-based methods offer a more targeted path.

Gradient-Based Optimization

A gradient-based solver, typically using an adjoint method, computes the sensitivity of performance metrics (such as torque or drag) relative to geometric changes. This identifies the regions of the blade most sensitive to performance objectives — such as reducing torque or improving downward flow. Once sensitivity regions are known, shape morphing modifies the mesh locally to explore improved geometries.

The process iteratively:

1. Identifies sensitive regions
2. Morphs blade geometry
3. Re-simulates performance
4. Repeats until optimal criteria are met

This method often reveals design improvements that traditional parameter sweeps might miss.

Section 03Blade Optimisation through Parametric Studies

In addition to gradient-based methods, multi-objective parametric optimization (e.g., with Ansys optiSLang) enables systematic exploration of geometric variations against performance goals. These studies can reveal trade-offs between torque, air delivery, and structural constraints to guide robust blade designs.

Section 04Acoustics Analysis of Ceiling Fans

Acoustics has become a key differentiator in modern consumer appliances. Ceiling fans are used for long durations, and prolonged exposure to unwanted noise can be uncomfortable for users.

From a physics perspective, noise is an acoustic pressure wave that propagates through air and is perceived by the human ear. Fan noise can originate from multiple sources, including air cutting by blades, vibration of fan components, and electric motor excitation. Noise generated due to airflow is referred to as airborne noise, while vibration-induced noise is known as structure-borne noise.

Blade profile and shape have a significant influence on airborne noise, whereas assembly quality and component fitment largely affect structure-borne noise. In some cases, fluid-induced pressure fluctuations excite structural vibrations, which then radiate as acoustic waves.

1. Tonal and Broadband Noise Characteristics

• Fan noise can be classified as tonal or broadband. Tonal noise is characterized by discrete, high-amplitude signals at specific frequencies, while broadband noise is a lower-amplitude, continuous sound spread across a wide frequency range.
• Ceiling fans operate at relatively low rotational speeds, which means most of their broadband noise falls within the human audible range of 20 Hz to 20 kHz.

Simulation-Driven Aeroacoustics

Conducting experimental acoustics studies can be time-consuming and expensive due to multiple test iterations and physical prototyping. Simulation provides an efficient alternative by offering early-stage insights into acoustic performance.

Within CFD simulations, pressure monitors record time-dependent pressure fluctuations. These signals can be post-processed using Fast Fourier Transformation (FFT) to analyze noise levels across frequencies. To capture very small pressure variations—down to 2 × 10⁻⁵ Pa—scale-resolving turbulence models are required. Although computationally intensive, the use of GPU-accelerated Ansys Fluent solvers significantly reduces turnaround time.

A steady-state simulation is typically performed first to establish a converged flow field, which then serves as the starting point for scale-resolving simulations.

Sound Propagation Modelling Approaches

Several methodologies are available for sound propagation analysis, each offering a different balance between accuracy and computational cost. Computational Aeroacoustics (CAA) resolves sound propagation directly by solving the Navier–Stokes equations, providing high fidelity but requiring substantial computational resources. Integral methods, such as the Ffowcs-Williams and Hawkings (FWH) model, reduce computational cost by solving wave equations for sound propagation, making them suitable for far-field noise prediction. Coupled approaches use dedicated acoustic solvers with CFD-derived source data, while broadband noise models offer fast, steady-state-based directional noise estimates.

Outcomes of Aeroacoustics Analysis:

• Pressure fluctuations with respect to time data which can be further processed in Fast Fourier Transformation (FFT) to get the data in frequency domain
• Sound pressure level (SPL) dB v/s frequency data as shown in Fig, which can help in identifying the critical frequencies where the SPL is high
• A-weighted SPL (dBA) v/s octave bands & 1/3rd Octave band frequencies as shown in Fig.
• Acoustic source information using DFT analysis for certain frequencies
• For FWH method source FFT analysis can be performed for identifying acoustic sources
• Audio data of the acoustic signal from simulation. One can hear these audio files to understand how the sound is perceived in real time.

Section 05IS 374 Air Delivery Testing Automation Using PyFluent

As simulation workflows grow more complex, automation becomes essential for productivity. IS 374 air delivery testing must be conducted for multiple blade designs, making manual setup repetitive and time-consuming.

Using PyFluent, the entire IS 374 simulation workflow can be automated — from model setup and execution to post-processing — ensuring consistency while significantly reducing overall analysis time. This approach is particularly valuable when evaluating design variants, where the same boundary conditions and solver settings must be applied repeatedly with precision.

Conclusion

Designing an efficient and reliable ceiling fan requires more than optimizing airflow alone. By integrating fluid–structure interaction, blade optimization, aeroacoustics, and simulation automation, engineers can realistically evaluate blade deformation, improve performance, reduce noise, and accelerate design validation. Together with the aerodynamic insights from Part 1, these advanced Ansys simulation techniques enable a holistic, physics-based approach to ceiling fan design, supporting better engineering decisions with reduced reliance on physical prototyping.