Global adoption of renewable energy is increasing due to growing concern over climate change, increasing costs associated with conventional generation, and decreasing capital investment costs of renewable energy technologies. Specifically, wind power represents the most technologically mature renewable alternative and is recognized as a cost effective generation source in both large and small power systems. However, the variability due to the stochastic nature of the wind resource introduces technological limitations to the amount of wind power which can be integrated in a power system. Energy storage is seen as a solution to mitigate the variability in wind power output.

Wind power and energy storage devices have the potential to contribute a substantial amount of renewable generation to meet the electricity demand in remote power systems. Remote power systems are characterized by their self reliance on electrical generation. The basic function of a remote power system is to provide the necessary power to satisfy the community’s electricity demand requirements as economically as possible with an adequate level of continuity and reliability.

In this thesis a probabilistic method for analyzing the integration of wind power and energy storage in a remote power system is developed, extending previous work done by Barton and Infield. The main objective of the method is to use a probabilistic model of a wind-storage system to estimate the required storage capacity and analyze the adequacy of power system components for a specified firm power commitment. A validation study and sensitivity analysis is provided comparing the probabilistic estimates of performance metrics to calculations from a time sequential simulation. The results of the study show the probabilistic method is limited in its general application due to the sensitivity of predicted metrics to system parameters such as installed wind capacity, firm power commitment, and confidence level. A method to reduce the residuals of the probabilistic estimates compared to calculations from a time sequence simulation method is provided. A case study for a remote power system located on Haida Gwaii is included to illustrate how the method can be used in a cost-benefit analysis of wind power and energy storage integration.