基于鲁棒优化的网架重构机组恢复顺序优化

在当前以火电为主的系统中,大停电后机组恢复时间的长短对重构效果影响显著。针对机组启动准备时段和启动时段可能出现的时间延迟,以最小化重构期间的电量不足为优化目标,构建了计及火电机组启动时间不确定性的机组恢复顺序鲁棒优化模型。通过交叉粒子群算法与CPLEX优化求解相结合,可获得量化表征恢复效果和运行可靠性的机组恢复顺序,为调度人员应对可能出现的最严重机组恢复迟滞场景提供了更加全面的决策参考。对于调度人员自行拟定的机组恢复顺序,还可根据运行经验预估其成功实施的概率,通过CPLEX求解并筛选关键时步,为调度人员有的放矢地保证恢复效果提供量化指导。针对新英格兰10机39节点系统和某区域电网的仿真结果验证了该方法的有效性。

Optimization of Units' Restoration Sequence during Network Reconfiguration Process Based on Robust Optimization

Current power systems mainly consist of thermal units, the length of units' recovery time after blackout determines the effect of network reconfiguration significantly.In response to this problem, aiming at the possible delay when units prepare to start and start up.In this paper, the robust optimization model for units' restoration sequence is set up which considers uncertain units start time, and treat insufficient electricity as optimization goal during minimum reconfiguration period.Then, crossover particle swarm algorithm and CPLEX optimization solution are combined to solve this model.Based on this, a units' recovery sequence which can quantitative characterization effort of recovery and operation reliability can be received.Providing the reference for dispatcher to cope with the worst possible restore hysteretic scene.Further, for the proposed units' recovery sequence given by dispatcher, they can get the probability of successful implementation estimated by operating experience and obtained the key steps by CPLEX to ensure recovery effect.The outcome of its application on the New England 10-unit 39-bus power system and a region grid indicate the validity of the method proposed.