ePrints

A Modular C++/Eigen Aero-Elastic Simulation Code for Multi-Rotor Wind Turbines

Lookup NU author(s): Professor Zhiqiang HuORCiD

Downloads

• Published version [.pdf]

Licence

This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).

Abstract

© 2026 by the authors.This paper presents AeroelasticQ, a modular, high-performance aeroelastic simulation code for wind turbines, with particular emphasis on future applicability to multi-rotor configurations. The framework is organized into three core components: a flexible-blade structural solver, an airfoil-based aerodynamic solver, and a two-mesh aero-structural mapping module for transferring loads and kinematics between the aerodynamic and structural discretization. The implementation is written in C++17 using the Eigen linear algebra library (v5.0.0), and OpenMP (v5.1) is employed to enable rotor-level parallel execution for multi-rotor applications. The structural dynamics are formulated using Kane’s dynamic method combined with modal superposition, while the aerodynamic loads are computed using three-dimensional blade element momentum theory. The coupled and uncoupled modules are validated in the time domain against OpenFAST (v4.1.2) AeroDyn, ElastoDyn, and the coupled AeroDyn–ElastoDyn configuration using the NREL 5 MW reference wind turbine. The rotor-level aerodynamic validation gives mean absolute errors of 8.94 × 10−4, 2.82 × 10−4, and 2.71 × 10−5 for Ct, Cp, and Cq, respectively, while the coupled aeroelastic cases show close agreement in blade tip deflections, blade root loads, and aerodynamic power. A rigid three-rotor verification confirms the multi-rotor load-aggregation framework, with tower base thrust and overturning moment errors below 1.5% and 2% NRMSE, respectively, in both all rotors operating and one operating/two-parked configurations. In single-thread benchmarks, AeroelasticQ achieves speedups of 5.23×, 19.69×, and 3.65× in the aerodynamic-only, structural-only, and fully coupled modes, respectively. In the multi-rotor benchmark, the five-rotor case achieves a parallel speedup of 2.55× with a parallel efficiency of 51%.

Publication metadata

Author(s): Qiu C, Yoshida S, Hu Z, Zhu H, Ismaiel A

Publication type: Article

Publication status: Published

Journal: Energies

Year: 2026

Volume: 19

Issue: 10

Online publication date: 20/05/2026

Acceptance date: 18/05/2026

Date deposited: 08/06/2026

ISSN (electronic): 1996-1073

Publisher: MDPI

URL: https://doi.org/10.3390/en19102457

DOI: 10.3390/en19102457

Data Access Statement: The dataset is available upon request from the authors

Altmetrics

Share