Bi-level Planning Strategy for Distribution Network Based on Source-load Temporal Characteristics
Due to the temporal characteristics of distributed power sources and loads,the integration of distributed pow-er sources into power grid will introduce uncertain factors to the planning of distribution network.To address this issue,Latin hypercube sampling is employed to generate wind-solar-load temporal scenarios at first,and an AP-DTW-K-me-doids approach is proposed for scenario reduction.Second,a bi-level planning model is constructed,in whichthe total cost is taken as asingle objective at the upper-level,and the operational cost and voltage deviation form adual-objective-at the lower-level.The single-objective particle swarm optimization algorithm is used at the upper-level,while an im-proved multi-objective particle swarm optimization(MOPSO)algorithm is used at the lower-level.The improved algo-rithm incorporates a chaos quasi-oppositionstrategy to broaden the coverage of solutions,and a novel inertia weight de-cay mechanism is put forward to enhance the capability ofescapingfrom the local optima.Finally,starting from the en-tire temporal scenario and a single scenario with superposed inertia weights,the rationality of the proposed model and algorithm isvalidatedby an IEEE 33-node system.
location and capacity determinationbi-level planningmulti-objective particle swarm optimization(MOP-SO)algorithmscenario reduction
万方数据
维普
