An Energy Meteorology Approach to Wind Power Optimization and Management under the Energy-Environment-Sustainability Nexus


Student thesis: Doctoral Thesis

View graph of relations

Related Research Unit(s)


Awarding Institution
Award date24 Aug 2018


The energy-environment-sustainability nexus is peculiar, with some elements of climate very much influencing our energy demand and supply (especially renewables). When coupled with the impacts of climate change, the task of sustainable energy development is one that requires urgent and lasting solutions. This has led to the emergence of energy meteorology. Over the last 2 decades, the development of wind energy meteorology as a potent solution has yielded positive and significant results globally, especially for installed wind energy capacity. However due to wind variability, the management, diffusion and optimization of wind power is still hindered. There exists a potential to maximize wind power despite its inherent limitations, by exploring the meteorology behind it.

By using high resolution reanalysis data wind seasonal variability over specified regions in Asia was examined. Wind power was relatively higher in winter than summer (JJA) driven by the North-easterly winds of the mid-latitudes during winter (DJF), due to relatively higher wind speeds occurring less frequently in winter. Wind capacity factor ranged between 0.1 – 0.5 and varied by season and by region, agreeable with surveyed studies of between 0.2 – 0.5 globally. The wind vector characteristics results over the study area are consistent with some other regions reported in literature that have relatively higher mean wind speeds during the cooler months than summer months. There exists an anti-correlation in wind power between the South West Asia and South East Asia regions which could be exploited for regional grid connection. Insight into seasonal variation of available wind power and its influence on demand-net-wind (DNW) is provided at this stage which forms a framework upon which grid connection, supply optimization, cost distribution and sustainability must commence. However, for actual turbine performance based on atmospheric and climatic conditions, real time data is required.

Thus in the second stage of the study, wind speed and wind power pairs from an operational wind turbine were used to investigate turbine power output. By averaging to 3 different time series, 3 principal phases of operations viz-a-viz wind speed prediction (atmospheric physicists), turbine control (turbine operators) and energy distribution/load scheduling (energy industry) were prioritized. Compared to the turbine power curve that theoretically gives the output, estimation errors of up to 53% were observed depending on the time interval considered and wind conditions. An effective power curve (EPC) was proposed and tested based on the polynomial function method of wind turbine power curve parametric modeling technique, generating a better fit and average estimation error less than 4% under very good wind conditions. Although the errors could be minimized, it is important to identify possible causes.

Therefore, the concluding part of the study considered the impact of topography and environmental conditions on wind turbine performance via two approaches. Firstly, the impact of wind shear on turbine power output was conducted using extrapolation laws to stratify the wind speed across of a 90 m blade diameter turbine based on the multiple element model analysis. 3 estimation approaches (micro-wind turbine, overall mean and turbine power curve) were compared, accounting for only a maximum of 25% dispersion that is highly dependent on terrain type. Secondly, a Computational Fluid Dynamic approach is used to investigate Annual Energy Production for different terrains under specified climatological, domain and boundary conditions. Differences of up to 35% are recorded depending on terrain topography, roughness, turbine face direction and prevalent wind direction. The vertical profiles of simulated wind field also confirm the change of wind speed over the turbine blade diameter. The results explain some factors responsible for wind power curve dispersion bellies.

These results reduce some knowledge gaps in energy meteorology, and provide necessary input of dependent variables in wind power development and management. These find direct application in technologies set to usher renewable energy development into the future such as support vector mechanism, machine learning, monitoring and evaluation systems and data science for wind energy optimization.