Although horizontal axis wind turbines (HAWTs) have been deployed around the world with typically positive results, vertical axis wind turbines (VAWTs) may offer some advantages in cost and performance if deployed correctly. Unlike HAWTs which must actively face the incoming wind to perform optimally, VAWTs perform at the same level regardless of the incoming wind direction. Another advantage is that generators can be placed on the ground beneath a VAWT rotor; this incurs lower structural costs than HAWTs which require a sturdy tower to support the rotor and generator.
In an array of HAWTs, the velocity deficit and increased turbulence levels in the wake of turbines have been shown to reduce the performance and lifespan of the wind farm. Understanding the way the wake develops downstream in the boundary layer is key to planning the optimal layout of turbine arrays. Such an understanding of wakes in boundary layer flows has not been developed equivalently for VAWTs. As such, this study aims to identify some of the defining characteristics of VAWT wakes and how they may influence downstream VAWTS.
A small scale vertical axis wind turbine is placed in a neutral boundary layer produced in a wind tunnel. The velocity and turbulence statistics in the full volume of the wake are quantified using stereo particle image velocimetry at multiple heights. This experiment is repeated for a range of tip speed ratios in order to relate the velocity deficit and levels of turbulence in the wake to the performance of the turbine. Due to the incoming boundary layer the velocity of the unperturbed flow is higher above the wake than below. The resulting high level of shear in this region enhances turbulence and entrainment, leading to a faster recovery of the wake. The configuration of vertical axis wind turbines has an additional influence on the wake as the blades in the rotor continuously change their orientation with respect to the incoming flow. This causes the velocity deficit and turbulence to be strongest behind the blades travelling upstream. The flow varies less with respect to the blades as the tip speed ratio increases, and causes the wake to become more symmetric in the crosswind direction.
The flow observed behind the VAWT in this experiment behaves very differently than behind a typical HAWT. Generally, HAWT wakes can be described as symmetric about the turbine axis of rotation with the velocity deficit and turbulence principally dependent on downstream and radial position. By contrast, the wake behind a VAWT is asymmetric across the wind and strongest downstream of the region of the rotor which turns into the wind. This has significant implications on both the potential layout of VAWT arrays compared to HAWT arrays as well as the sensitivity of the array to the direction of the incoming wind.