Isolated microgrids have been shown to be a reliable and efficient solution to provide energy to remote communities. From the primary control perspective, due to the low system inertia and fast changes in the output power of wind and solar power sources, isolated microgrids' frequency can experience large excursions and thus easily deviate from nominal operating conditions, even when there is sufficient frequency control reserves;​ hence, it is challenging to maintain frequency around its nominal value. From the secondary control perspective, the generation scheduling of dispatchable units obtained from a conventional Unit Commitment (UC) are considered fixed between two dispatch time intervals, yielding a staircase generation pro file over the UC time horizon;​ given the high variability of renewable generation output power, committed units participating in frequency regulation would not remain fixed between two time intervals. The present work proposes techniques to address these issues in primary and secondary frequency control in isolated microgrids with high penetration of renewable generation.

In this thesis, first, a new frequency control mechanism is developed which makes use of the load sensitivity to operating voltage and can be easily adopted for various types of isolated microgrids. The proposed controller offers various advantages, such as allowing the integration of significant levels of intermittent renewable resources in isolated/islanded microgrids without the need for large energy storage systems, providing fast and smooth frequency regulation with no steady-state error, regardless of the generator control mechanism. The controller requires no extra communication infrastructure and only local voltage and frequency is used as feedback. The performance of the controller is evaluated and validated using PSCAD/EMTDC on a modified version of the CIGRE benchmark;​ also, small-perturbation stability analysis is carried out to demonstrate the contribution of the proposed controller to system damping.

In the second stage of the thesis, a mathematical model of frequency control in isolated microgrids is proposed and integrated into the UC problem. The proposed formulation considers the impact of the frequency control mechanism on the changes in the generation output using a linear model. Based on this model, a novel UC model is developed which yields a more cost e efficient solution for isolated microgrids. The proposed UC is formulated based on a day-ahead scheduling horizon with Model Predictive Control (MPC) approach. To test and validate the proposed UC, the realistic test system used in the first part of the thesis is utilized. The results demonstrate that the proposed UC would reduce the operational costs of isolated microgrids compared to conventional UC methods, at similar complexity levels and computational costs.