Control of Microgrids with Renewable Energy Sources and Battery Energy Storage Systems

Abstract

Renewable energy has a significant potential to reduce global carbon emissions and secure long-term sustainable power supply. Higher penetration of renewable energy, however, is largely hindered by the inherent variability and intermittency of renewable resources, and renewable energy is generally considered as non-dispatchable. In other words, rather than control and dispatch the renewable energy economically, grid operators must arrange other generation units in response to what produced by wind turbines or solar panels in order to maintain network stability and meet reliability parameters.

This motivates the development of microgrids supplied by renewable energy resources. Either grid-connected or islanded, these decentralized power systems are believed to be a promising solution to achieve higher penetration of clean energy in the future. Given proper control of storage units and communications with the energy market, the non-dispatchable energy can be regulated to reduce the difficulty of power scheduling in the grid operation.

Battery energy storage systems (BESS) are used in many research to make renewable energy more controllable and usable on demand. Assisted by appropriate prediction techniques and control algorithms, BESS can store part of the generated renewable energy and supply some stored energy at different periods, so that the actual power dispatched can meet the required operating criteria.

The main focus of this thesis is the optimal control strategy of microgrids with renewable energy systems and battery energy storage systems. The control algorithms introduced in this thesis are developed to maximize the profit and minimize the energy cost in microgrids, which require predicted data on renewable power, temperature, electricity price, and load demand. In comparison with many previous studies in this area, we developed some novel algorithms that have an emphasis on the battery lifetime degradation and computation speed, which can be critical in achieving the optimal dispatch in microgrids.