2014/10/28 by Paul Scott, Scott, Paul, Sylvie Thiébaux +1
Engineering · #Microgrid Control and Optimization #Smart Grid Energy Management #Optimal Power Flow Distribution
paper · pdf · doi:10.48550/arxiv.1410.7868
Smart devices, storage and other distributed technologies have the potential\nto greatly improve the utilisation of network infrastructure and renewable\ngeneration. Decentralised control of these technologies overcomes many\nscalability and privacy concerns, but in general still requires the underlying\nproblem to be convex in order to guarantee convergence to a global optimum.\nConsidering that AC power flows are non-convex in nature, and the operation of\nhousehold devices often requires discrete decisions, there has been uncertainty\nsurrounding the use of distributed methods in a realistic setting. This paper\nextends prior work on the alternating direction method of multipliers (ADMM)\nfor solving the dynamic optimal power flow (D-OPF) problem. We utilise more\nrealistic line and load models, and introduce a two-stage approach to managing\ndiscrete decisions and uncertainty. Our experiments on a suburb-sized microgrid\nshow that this approach provides near optimal results, in a time that is fast\nenough for receding horizon control. This work brings distributed control of\nsmart-grid technologies closer to reality.\n