Is the Actuator line method able to reproduce the interaction between closely-spaced Darrieus rotors? a critical assessment on wind and hydrokinetic turbines

Abstract

The deployment of multiple closely-spaced Darrieus turbines has recently gained interest in the academic and industrial sectors due to their potential to increase power output and wake recovery. However, the computational cost of accurate three-dimensional, blade-resolved CFD analyses rapidly becomes unfeasible as long as multiple rotors are added, putting emphasis on the need for a more practical, yet accurate, simulation method. Based on recent assessments, the Actuator Line Method (ALM) has been shown to represent the most interesting solution for Darrieus turbines’ simulations. However, since the blade-flow interaction is not solved, no evidence was available to date on whether ALM can capture all the physical phenomena related to turbine mutual interaction.In this study, the reliability and limitations of the ALM in simulating closely-spaced Darrieus turbines were assessed using two pairs of turbines with varying geometrical and operational characteristics, focusing on turbine loads, local flow field, and wake development. The results demonstrate that the ALM is effective for simulating twin-rotor Darrieus turbines, particularly at medium and high tip-speed ratios, providing satisfactory predictions of instantaneous blade loads and capturing the increase in torque due to the mutual interaction between adjacent rotors. The ALM is also able to accurately reproduce the flow field across the twin rotors, as demonstrated through Proper Orthogonal Decomposition (POD) analysis of the wake flow field. With an average simulation time of 7 h/rev/core, compared to 508 h/rev/core for blade-resolved CFD, this study provides evidence that the ALM is a low-cost and reliable tool for simulating closely spaced Darrieus turbines, thus representing the most suitable tool to develop new VAWT applications like floating wind turbines or hydrokinetic installations.