异步联网下云南电网高频切机优化配置方案
Optimal Configuration Scheme of Over-frequency Generator Tripping for Yunnan Power Grid of China with Asynchronous Interconnection
-
摘要: 中国云南电网与中国南方电网主网实现异步联网后,外送直流线路故障引发的大量盈余功率使得系统频率稳定问题愈加突出。针对直流闭锁故障,推导了各发电机组对故障盈余功率的贡献因子,并提出了综合考虑故障严重程度和概率的切机优先度指标。随后讨论了高频切机首轮频率和末轮频率的配置原则,并采用动态交叉排序的方法满足被切机组的分散性要求。为提高仿真效率,先通过调节功率估计高频切机的近似解,再利用稳态频率约束修正该近似解,从而完成对每轮切除机组的优化配置。最后,通过应用聚类方法对生成的典型故障进行校验,结果表明所提方法不仅能提高暂态过程中最低频率并能有效降低最高频率,从而更利于系统稳定并规避低频减载动作风险,同时在保证稳态频率处于允许范围内的条件下有效减少了整体切机量。Abstract: After the Yunnan power grid of China and the China southern main power grid are interconnected asynchronously, the large amount of surplus power caused by the fault of sending DC lines makes the problem of system frequency stability more severe. Aiming at the DC blocking faults, the contribution factor of each generator unit to the surplus power caused by the fault is derived, and a tripping priority index with fault severity and probability taken into account is proposed. Furthermore, the configuration principles of the first-round and the last-round frequency settings for over-frequency generator tripping are discussed,and the dynamic cross-sequencing method is leveraged to meet the dispersion requirements of the tripped units. In order to improve the simulation efficiency, the approximate solution of over-frequency generator tripping is estimated by adjusting power first, and then the solution is modified by using the steady-state frequency constraint to complete the optimal configuration of the tripped units at each round. Finally, the generated typical faults are verified by using the clustering method. The simulation results demonstrate that the proposed method not only increases the minimum frequency but also reduces the maximum frequency during the transient process, which is more beneficial to the system stability and avoids the risk of under-frequency load shedding. At the same time,the overall tripping capacity is effectively reduced on the premise of ensuring that the steady state frequency is within the allowable range.