Dissertation Defence: A Multiscale Computational Framework for Large-Scale Wind Farm Flow Simulations in Complex Terrain

Arjun Ajay, supervised by Dr. Joshua Brinkerhoff, will defend their dissertation titled “A Multiscale Computational Framework for Large-Scale Wind Farm Flow Simulations in Complex Terrain” in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Mechanical Engineering.

An abstract for Arjun Ajay’s dissertation is included below.

Examinations are open to all members of the campus community as well as the general public. Registration is not required for in-person exams.

Abstract

The rapid expansion of wind energy is central to achieving global decarbonization and energy security goals. On land, wind turbines are increasingly installed in mountainous regions. Wind-farm performance is governed by interacting physical processes spanning a wide range of spatial and temporal scales, from synoptic and mesoscale atmospheric forcing to boundary-layer turbulence, terrain-induced flow heterogeneity, and turbine-scale wake dynamics. Understanding these multiscale interactions remains a major scientific and computational challenge, particularly in thermally-stratified atmospheric boundary layers and topographically complex environments where conventional modelling approaches exhibit significant limitations.

Aiming to overcome these limitations, this thesis develops a multiscale simulation framework for large eddy simulations (LES) of onshore wind farms under realistic ground topography and mesoscale atmospheric variability. Complex terrain is modelled using a hybrid wall-modelled immersed boundary method that captures the dominant influence of complex ground topography while maintaining numerical stability and computational efficiency. Mesoscale atmospheric variability is introduced through a profile-assimilation-based coupling strategy using multiresolution analysis with wavelet basis functions and a hybrid geostrophic–wavelet forcing formulation, enabling time-varying meteorological forcing and diurnal boundary-layer evolution to be consistently imposed. Turbulence representation on coarse grids is enhanced by developing a mixed subgrid-scale modelling approach combined with higher-order numerical discretization, enabling improved representation of anisotropy and non-equilibrium effects in stratified and terrain-influenced boundary layers. All methods are implemented in the open-source LES code TOSCA (Toolbox fOr Stratified Convective Atmospheres), providing a unified and extensible platform for wind energy simulations in complex environments.

The developed framework is used to model the performance of a realistic windfarm simulation over an entire diurnal period, validated against measurements from the American WAKE ExperimeNt (AWAKEN) field campaign. Results show that during unstable atmospheric conditions, power production experiences limited spatial variability across the farm, while stable conditions produce pronounced spatial and temporal variability in both flow properties and turbine power. Moreover, while upstream turbines produce higher power than downstream rows during unstable and transition regimes due to turbine wake effects, under stable conditions downstream turbines outperform the first row despite experiencing stronger wakes. This behaviour originates from a dynamic interaction between the complex terrain and a nocturnal low-level-jet, and highlights the benefit of the simulation framework developed in this thesis for elucidating the multi-scale flow physics of onshore wind farms.