Optimal DER allocation in meshed microgrids with grid constraints

Abstract

We develop a new approach to model the linearized power flow in distributed energy resources (DER) allocation and dispatch optimization. The model uses the Decoupled Linearized Power Flow (DLPF) (Yang et al., 2016) approach. DLPF has the advantage of being suitable for meshed networks, while the majority of current models use LinDistFlow, which is only suitable for radial networks. First, we provide a validation of the DLPF voltage magnitude and branch power (in kVA) solutions compared to LinDistFlow and the true AC power flow (ACPF) solution for a meshed benchmark network, a 33-node system. We then formulate the DER allocation and dispatch problem as a mixed integer linear program (MILP) and use DLPF and LinDistFlow to model constraints on the electric power network infrastructure, which limits voltages to ANSI C84 standard limits. The model further uses polygon relaxations to limit kVA flows at the branches to their allowed thermal limit. We demonstrate the DLPF based DER allocation and dispatch model using a 115-node meshed circuit. Results indicate that the DLPF formulation is effective in capturing undervoltages and overvoltages in the meshed network, leading to optimal siting and sizing of photovoltaic (PV) and battery storage capacities. The DLPF model is then compared with a MILP that uses the LinDistFlow formulation, highlighting the importance of selecting the appropriate power flow linearization method.