Linear optimal power flow using cycle flows

摘要: Abstract Linear optimal power flow (LOPF) algorithms use a linearization of the alternating current (AC) load equations to optimize generator dispatch in network subject loading constraints branches. Common voltage angles at buses as optimization variables, but alternatives can be computationally advantageous. In this article we provide review existing methods and describe new formulation that expresses directly terms flows themselves, using decomposition graph into spanning tree closed cycles. We comprehensive study computational performance various formulations, settings include challenging applications such multi-period LOPF with storage generation capacity expansion. show solves up 7 times faster than angle commercial linear programming solver, while another cycle-base 20 faster, an average speed-up factor 3 for standard networks considered here. If capacities are also optimized, rises 12, reaching 213 particular instance. The is largest many decentral generators throughout network, which highly relevant given rise distributed renewable challenge operation planning networks.