Energy management of microgrids

The innovative shape of traditional power systems has been developing towards more intelligent, flexible and efficient entities. In the name of microgrids, the transformation of physical structure has gradually blurred the boundaries between generation, transmission and distribution network from centralization to decentralization topologies. With much smaller scales, microgrids are comprised of local distributed generation, energy storage systems (ESSs) and load, making it quite possible to enhance reliability with distributed generation, increase efficiency with reduced transmission distance, and implement capability in ease use of alternative energy resources and storages. However, public concern has been raised due to progressive stability and reliability issues regarding different aspects related with fast developing system infrastructure and emerging electricity market. On the other hand, it might sometimes become a serious issue for microgrids on the trade-off between economy and reliability, since operating reserves are much incomparable to variations from both generation and consumption sides in most cases. Against the background of exploring better configuration for microgrids to be not only reliable but also economic, many conducted research has mainly focused on autonomous operation, droop control, hierarchical control, low-voltage protections, state estimations and power electronics studies and so forth. In terms of modeling and control strategy, this thesis gives an overall comprehensive methodology about effective planning and efficient operation of microgrids. Firstly, optimal power flowalgorithms are investigated in a novelway that provides the minimum operating cost and maintains the highest possible reliability levels. In order to deal with stochastic behaviors of RESs and loads in microgrids, scenario reduction techniques are extended by imposing auxiliary constraints so that the optimal solution space is narrowed down. By implementing the fine-tuned algorithm, case studies with benchmarks have well validated the reliability has been significantly enhanced at little cost of additional expenditure. Large-scale energy storage systems have provided promising potentials for multiple applications in microgrids by providing additional spinning reserve, regulating voltage, frequency and power factors, as well as offering economic benefits by participating into demand response and facilitating the integration of renewable energy sources (RESs). A dynamic optimal power flow (OPF) model with adaptive operation costs is thus proposed to address influential consequences of ESS implementation in microgrids, specifying the dynamic characteristics of energy storage in multiple time periods. Conclusive results have shown that the proposed model and algorithm can be utilized not only to meet system ramping requirements, but also to help flatten the load profile in peak hours. In microgrids, participation of renewable energy sources in combination of ESSs has further expanded the potential benefits to end uses as well as to system operators. In consideration of economic operation, a two-layer predictive energy management system (EMS) for microgrids with hybrid ESS is proposed incorporating respective degradation costs, so that high system robustness at minimum operational cost is well maintained. The hierarchical dispatch model is well designed, that total operational cost is achieved in the upper layer EMS while the lower layer EMS deals with more cumbersome tasks resulted in intermittencies and forecast errors. The novelty of the proposed algorithm has been demonstrated by intensive scenario tests incorporating different pricing schemes, prediction horizon lengths and forecast accuracies. Clustering microgrids in different distribution networks, microgrid commu- nity has enabled individual members to be more operationally flexible by resource sharing with minimizing dependency on the main power grid during normal conditions and more self-sustainable with enhanced power system reliability under extreme events. This thesis further investigated the hierarchical EMS in the context of interconnected microgrid community, in which pairing strategy is proposed so that community-level EMS to explicitly determine the power flow between microgrids and with the upstream distribution grid. Such the profit can be fairly shared by each participating microgrid in the community, without user privacy being compromised. All the proposed methodologies and algorithms throughout the research have been verified in MATLAB, with regards to different system topologies involving multiple data of RESs, ESSs and loads.Doctor of Philosophy (IGS