The team, led by Hyeokseong Hong, Inho Jeong and Haeseong Cho, set out to build what aerodynamicists call a mid-fidelity aeroelastic analysis tool. The term aeroelastic refers to the interplay between aerodynamic forces and elastic structural deformation: air loads bend the blade, the bending changes the blade’s shape and angle of attack, and the altered shape in turn changes the air loads. When this feedback loop is strong, as it is for helicopter rotors, neither the fluid nor the structure can be analyzed in isolation. High-fidelity approaches that resolve every detail of the airflow with computational fluid dynamics can capture this coupling exquisitely well, but they demand such enormous computing resources that iterating on a design — or even converging on a single trimmed flight condition — becomes impractical. Low-fidelity methods, meanwhile, run quickly but often miss the wake-driven physics that dominate rotor loads.

The Korean team’s answer was to pair two solvers of intermediate fidelity, each chosen for what it does best. On the structural side, they employed a solver built on Geometrically Exact Beam Theory, often abbreviated GEBT. Unlike simpler beam formulations that assume small deflections, GEBT makes no such concession: it tracks large rotations and displacements of the blade exactly, which matters enormously for rotors whose blades can flex by substantial fractions of their radius and twist by several degrees under load. The formulation, rooted in the intrinsic equations of moving beams developed by Hodges and refined in earlier work by the same group, allows the blade to deform realistically without the artificial stiffness that plagues linearized models.

On the aerodynamic side, the researchers coupled their structural solver to DUST, an open-source code built on the Vortex Particle Method. Rather than meshing the entire volume of air around the rotor as a conventional CFD solver must, the vortex particle method represents the wake — the sheet of rotating air shed from each blade — as a cloud of discrete particles carrying vorticity. As the rotor turns, these particles are transported, stretched and diffused according to the equations of fluid motion, allowing the wake to evolve freely and naturally. This is a crucial advantage in forward flight, where the wake is swept behind the rotor and can loop back to interact with the blades in the phenomenon known as blade-vortex interaction, a major source of noise and vibration. Because the method tracks vorticity directly rather than resolving every eddy in the flow field, it captures wake dynamics at a fraction of the computational cost of full CFD.

Gluing the two solvers together is its own engineering challenge, and here the team turned to preCICE, an open-source coupling library that has become a workhorse for fluid-structure interaction studies. preCICE handles the delicate business of exchanging data between codes that may use different meshes, different time steps and different mathematical descriptions of the same physical surface. Forces computed by the aerodynamic solver must be mapped onto the structural mesh, and the resulting deformations must be mapped back so the aerodynamics can be updated. Getting this mapping right — conserving forces and moments across the interface — is essential for the coupled system to converge to a physically meaningful answer rather than oscillating between the two solvers.

The coupling strategy itself is deliberately loose rather than tight. In a tightly coupled scheme, the fluid and structural solvers exchange information many times within each time step, which guarantees stability but multiplies the computational cost. The Jeonbuk team instead used a loosely coupled approach in which the solvers exchange data once per step, and they stabilized the process with what is known as the delta-airload method. This technique, well established in rotorcraft analysis, feeds only the change in aerodynamic load between iterations into the structural solver, damping out the spurious oscillations that loose coupling can otherwise introduce. Crucially, the framework also performs trim analysis: it iterates the rotor’s control inputs — collective and cyclic pitch — until the blade loads and vehicle forces match the target flight condition. Trim is what turns a simulation from a curiosity into an engineering tool, because real helicopters must fly at equilibrium, and loads measured on an untrimmed rotor are meaningless for design.

To prove the framework works, the researchers turned to HART-II, the Second Helicopter Heroic Aeroacoustic Rotor Test, an international benchmark conducted by the German Aerospace Center DLR with a four-bladed model rotor fitted with higher harmonic control. HART-II is arguably the most thoroughly instrumented rotor test ever performed, with laser and pressure measurements capturing blade motions, airloads and wake geometry in extraordinary detail. The benchmark defines three canonical cases: the baseline configuration, a Minimum Noise case in which blade pitch is actively modulated to reduce blade-vortex interaction noise, and a Minimum Vibration case tuned to suppress hub vibrations. Each case stresses the simulation in a different way, and all three demand accurate prediction of both the aerodynamic pressures on the blade and the elastic response of the structure.

The results, according to the study, show good agreement between the framework’s predictions and the HART-II reference data for all three configurations, capturing the key aeroelastic characteristics of the rotor in forward flight. The structural responses — blade bending and torsion — and the aerodynamic responses — sectional airloads — both track the experimental trends, which is precisely the dual validation an aeroelastic tool needs. Achieving this at mid-fidelity computational cost is the real prize: it means the framework can be used inside the iterative loops of design and trim analysis, where high-fidelity CFD-based coupling would be prohibitively expensive. The work builds on the group’s earlier mid-fidelity framework for medium utility helicopters and on their research into model-order reduction for composite rotor blades, suggesting a sustained research program aimed at making rotor load prediction both accurate and affordable.

The broader implications reach well beyond one benchmark. As the rotorcraft industry pivots toward electric vertical-takeoff-and-landing aircraft, tiltrotors and other unconventional configurations, the ability to predict blade loads quickly and reliably becomes a bottleneck for certification and design. Mid-fidelity tools of the kind developed here occupy a sweet spot: accurate enough to capture wake-driven physics and geometric nonlinearity, fast enough to explore thousands of design variants and flight conditions. The work was supported by South Korea’s Defense Acquisition Program Administration through the Korea Research Institute for defense Technology planning and advancement, reflecting the strategic importance of rotorcraft technology. If frameworks like this one continue to mature, the helicopters and air taxis of the coming decades may be quieter, smoother and safer — engineered with the help of simulations that finally see the wake clearly.

Subject of Research: Mid-fidelity aeroelastic fluid-structure interaction framework for helicopter rotor blade load prediction in forward flight

Article Title: A Wake-Resolved Aeroelastic Framework for Rotor Blade Load Prediction in Forward Flight

Article References: Hong, H., Jeong, I., & Cho, H. (2026). A Wake-Resolved Aeroelastic Framework for Rotor Blade Load Prediction in Forward Flight. International Journal of Aeronautical and Space Sciences. https://doi.org/10.1007/s42405-026-01275-8

Image Credits: AI Generated

DOI: 10.1007/s42405-026-01275-8

Keywords: rotorcraft, aeroelasticity, geometrically exact beam theory, vortex particle method, fluid-structure interaction, rotor wake, blade-vortex interaction, HART-II, trim analysis, preCICE, helicopter loads, mid-fidelity simulation