Degree Name

Doctor of Philosophy


School of Electrical, Computer and Telecommunications Engineering


The development of cooperative control strategies for microgrids has become an area of increasing research interest in recent years, often a result of advances in other areas of control theory such as multi-agent systems and enabled by emerging wireless communications technology, machine learning techniques, and power electronics. While some possible applications of the cooperative control theory to microgrids have been described in the research literature, a comprehensive survey of this approach with respect to its limitations and wide-ranging potential applications has not yet been provided. In this regard, an important area of research into microgrids is developing intelligent cooperative operating strategies within and between microgrids which implement and allocate tasks at the local level, and do not rely on centralized command and control structures. Multi-agent techniques are one focus of this research, but have not been applied to the full range of power quality problems in microgrids. The ability for microgrid control systems to manage harmonics, unbalance, flicker, and black start capability are some examples of applications yet to be fully exploited. During islanded operation, the normal buffer against disturbances and power imbalances provided by the main grid coupling is removed, this together with the reduced inertia of the microgrid (MG), makes power quality (PQ) management a critical control function.

This research will investigate new cooperative control techniques for solving power quality problems in voltage source converter (VSC)-based AC microgrids. A set of specific power quality problems have been selected for the application focus, based on a survey of relevant published literature, international standards, and electricity utility regulations. The control problems which will be addressed are voltage regulation, unbalance load sharing, and flicker mitigation. The thesis introduces novel approaches based on multi-agent consensus problems and differential games. It was decided to exclude the management of harmonics, which is a more challenging issue, and is the focus of future research. Rather than using model-based engineering design for optimization of controller parameters, the thesis describes a novel technique for controller synthesis using off-policy reinforcement learning. The thesis also addresses the topic of communication and control system co-design. In this regard, stability of secondary voltage control considering communication time-delays will be addressed, while a performance-oriented approach to rate allocation using a novel solution method is described based on convex optimization.

FoR codes (2020)

400802 Electrical circuits and systems, 400803 Electrical energy generation (incl. renewables, excl. photovoltaics), 460202 Autonomous agents and multiagent systems



Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.