Uncertainty quantification pertinent to wind farm flow physics and modeling

This paper attempts to identify and quantify the uncertainties associated with the wind farm modeling and flow physics in wind farms. Different aspects and types of uncertainties in modeling the atmosphere and rotor using different fidelity methods are presented. A description of the uncertainty quantification algorithms is also presented. This is followed by some case studies on the identification and quantification of relevant uncertainties. The first one pertains to the input parameters and model constants for a wind resource assessment problem on a complex terrain. The second one explains the uncertainties associated with a turbine-turbine interaction problem in atmospheric boundary-layer flow using different actuator line modeling approaches for rotors. This is further exemplified using the aleatoric and epistemic uncertainties in blade loads as a result of using an actuator line method, followed by a discussion on the physical propagation of these uncertainties downstream. The third type of case study is that of a 3-D cosine hill where the polynomial chaos algorithm is applied to propagate the uncertainties in model constants for Reynolds-averaged Navier-Stokes equations and to investigate their effect on the flow-field.