Sensitivity analysis of the performance of a Darrieus hydrokinetic turbine in uncertain operating conditions

Abstract

Unlike the previous scientific works on the vertical axis turbines which were mostly limited to design improvements, performance metrics and flow dynamics but significantly ignored the effect of operational uncertainties on the power output, this paper presents how the performance of a low-speed vertical axis hydrokinetic turbine can be attributed to the inherent uncertainties in the operating conditions. While a complete set of uncertainty sources in the design, development and operation of a water turbine is pictured with real and hypothetical relationships, this study focuses mainly on two uncertainties. Starting from the methodological illustration of a towing tank experiment, the sources of uncertainty are defined as ±0.7% in free-stream velocity V0 and ±1.06% in the turbine’s rotational velocity Ω. The stochastic space is constructed on a two-dimensional Clenshaw-Curtis grid and finite-volume-based deterministic computations are performed at each grid node. The sensitivity of the turbine’s performance to these input uncertainties is quantified using probability methods, including response surface, PDF and CDF. Flow-related non-linearities such as vortex release and blade-vortex interaction are directly reported in stochastic space and corresponding variations in the force coefficients of the turbine’s blade are identified. It is observed that the uncertainty in the turbine’s rotation is more influential on the power output compared to free-stream velocity.