Optimal virtual microgrid design using community energy storage for distribution networks

• Main Author: Alawami, Hasan Yahya Ali
• Format: Thesis
• Language: English; English
• Published: 2023
• Subjects: T Technology (General); TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utem.edu.my/id/eprint/28282/1/Optimal%20virtual%20microgrid%20design%20using%20community%20energy%20storage%20for%20distribution%20networks.pdf; http://eprints.utem.edu.my/id/eprint/28282/2/Optimal%20virtual%20microgrid%20design%20using%20community%20energy%20storage%20for%20distribution%20networks.pdf; http://eprints.utem.edu.my/id/eprint/28282/; https://plh.utem.edu.my/cgi-bin/koha/opac-detail.pl?biblionumber=124152

Virtual microgrid is concerned with upgrading the traditional distribution network (DN) into smart DN using distributed energy resources. This research develops a framework for designing optimal virtual microgrid (VM) in two steps: boundary identification considering both structural characteristics and operating states of PV residential networks and sizing and locating Community Energy Storage (CES). The CES sizing and locating procedure is done in each VM with the goal of maximizing the economic benefit of CES deployment. The methodological tackling of the VM boundaries problem is achieved by identifying boundaries using two inputs: distribution line resistivity and transmitted power. Louvain heuristic algorithm is used for optimizing the VM partitioning. The CES optimal placement and sizing in the VMs are identified by maximizing the overall net present value (NPV) of CES deployment over a specific planning horizon. To tackle the CES allocation issue, Genetic Algorithm (GA) toolbox in MATLAB is utilized. The VM partitioning strategy was tested on IEEE 33-bus and to verify the algorithm robustness on a larger system, IEEE-118 bus system was used. When compared to previous work, the results of the VM partitioning using the proposed strategy showed lower line losses and higher electrical modularity. The impact of PV penetration level and its distribution to the VM partitioning was investigated on IEEE 33-bus and IEEE 69-bus distribution systems. Finally, VM partitioning for a 19-bus Malaysian distribution network is done. With respect to the second objective, the results of CES allocation strategy deployed on 19-bus Malaysian distribution network demonstrated that the proposed approach is capable of determining the optimal size, location and operating characteristics of CES in each VM, as well as maximising CES' total profit. This scenario is expected to achieve a revenue of 16.822 million MYR within the 20 years of CES deployment. Furthermore, it is observed that dispatching CES active power in each VM can lower peak demand, eliminate the violation of reverse power flow limits and reduce the cost of purchasing electricity from the grid. Sensitivity analysis shows that 40% penetration level of PV generation yields the highest NPV from CES deployment and starting the VM project in the next 20 years, i.e., year 2042 yields the highest BCR which is 5.22 due to ES cost reduction. The proposed framework can be used as a guideline for the power utilities and policy makers in developing a smart and reliable distribution network with higher renewable energy penetration.