Fuzzy-grey wolf optimization for energy storage sizing and power management in microgrids

Escalating fossil fuel prices, pressure of more stringent environmental regulations, and deregulation in the electricity market provide opportunities and motives for renewable energy sources (RESs), such as wind, solar, tidal and wave energy to be integrated into existing power grids. Nevertheless, due to the distributed nature of RESs, the traditional centralized power system architecture, as well as control mechanisms, are not adequate to support the integration of renewable energy systems. Therefore, the concept of a microgrid emerged, where a group of distributed energy sources and loads, form a network within clearly defined electrical boundaries. Microgrids can be operated independently in the 'islanded' mode or connected to a bigger power network to operate in the 'grid-connected' mode. In addition, due to the intermittent nature of the RESs, and the non-linear behaviour of load demands, high levels of reserves are required to smooth over the resultant power fluctuations. Therefore, energy storage systems are becoming an integral part of microgrids which improve reliability issues, demand management and provide support for voltage and frequency control, and thereby increase system stability and reliability. The importance and inevitability of energy storage in microgrids have extensively been demonstrated in recent research. Nevertheless, optimal sizing of energy storage systems in microgrids is an area which requires further developments. This thesis critically reviews the gaps in existing optimal sizing strategies and proposes a powerful hybrid optimization method based on artificial intelligence techniques to achieve maximum efficiency, enhance the economic dispatch, and provide optimal performance for the microgrid modules. The hybrid optimization method implemented in real-time energy management within microgrids power system to guarantee optimal and safe operation of the system. Moreover, planning and operating generation sources in an efficient manner has a significant influence on microgrid stability, reliability, and profitability. This thesis will subsequently present power management system (PMS) for reliable and economic operation of microgrids in both grid-connected and islanded modes. In addition, the proposed PMS has the ability to schedule the microgrid generation with minimum information shared by generation units. Due to the comparable sizes of loads and sources, operating a microgrid in islanded-mode is more challenging compared to operating in a grid-connected mode. This could be seen from the inertia point of view as well, where lack of inertia or traditional synchronous generators leads to frequency deviations. Moreover, the intermittent nature of RES results in power imbalances and instabilities. Thus, energy storage has been used in various ways to solve both issues in islanded microgrids, and thus is becoming an industrial norm. The addition of a new generation source to an already existing power grid is not easily accomplished, due to the high impact on both the system stability as well as the economic dispatch of the grid. Subsequently this thesis presents frequency control strategies for islanded microgrids based on the optimal size of Battery Energy Storage System (BESS), as well as efficient modelling of BESS' controller. Through the utilization of both conventional generation as well as RES, the proposed approach enhances the system frequency response in cases of differing penetration levels of renewable sources. This thesis also has explored the impact of dynamic operations of both emerging technologies, such as distributed generation with battery storage, as well as islanded microgrids on system frequency performance. In addition, this thesis investigated the impact of adding new generation sources, such as tidal energy, to the existing power grid as well as addressed the impact of the new source on determining the optimal size of BESS and system economic dispatch. Accordingly, in order to evaluate the impact of the proposed methods in this research, different scenarios are defined and simulated in a real case study with the use of different software tools, such as MATLAB Simulink, m-file, and DIgSILENT PowerFactory. In conclusion, the obtained results of this research demonstrated that significant economic and environmental benefits can be achieved with the use of an artificial intelligence-based hybrid optimization method in managing generation sources in a microgrid. Furthermore, the results showed that adding the tidal energy source to the islanded microgrid has outstanding performance for minimising the operating cost of the microgrid as well as minimising the optimal size of BESS that required to stabilise the power grid. Moreover, selecting an appropriate size for the BESS, along with a robust BESS controller, has a significant impact on the frequency response stabilisation of microgrid in the islanded mode, which in turn increases the power system reliability and security.