基于分布估计和智能存储的复杂电网可靠性评估方法

发布时间：2018-06-24 14:08

本文选题：分布估计算法 + 智能存储　；参考：《中国电机工程学报》2017年19期

【摘要】：为了加快复杂电网可靠性评估的计算效率,该文提出一种基于分布估计与智能存储的大电网可靠性评估方法。该方法在抽样阶段与状态评估阶段对传统的可靠性评估方法进行综合改进。在抽样阶段,通过改进分布估计算法,提高种群的优良性,快速实现对状态空间的约减,在约减后余下高失负荷密度的状态空间中随机抽样,达到减小方差的目的。在状态评估阶段,引入基于双十字链表的智能存储技术,在加快状态搜索匹配速度的同时,避免对同一状态序列重复进行最优潮流计算,进一步地提高计算效率。最后在典型IEEE-79节点的电力系统中对该方法进行测试,结果表明,该方法相比于重要抽样等方法,不仅具有更高的精度,而且计算速度更快,算例仿真验证该方法的有效性与可行性。
[Abstract]:In order to speed up the computational efficiency of reliability assessment for complex power systems, a new reliability evaluation method for large power networks based on distributed estimation and intelligent storage is proposed. This method improves the traditional reliability evaluation method in sampling stage and state evaluation stage. In the sampling stage, by improving the distribution estimation algorithm, the population quality is improved, and the reduction of the state space is realized quickly, and the random sampling in the state space with the remaining high loss of load density after the reduction is carried out, so as to reduce the variance. In the stage of state evaluation, the intelligent storage technology based on double cross linked list is introduced to accelerate the speed of state search and matching, and to avoid repeated optimal power flow calculation for the same state sequence, thus further improving the computational efficiency. Finally, the method is tested in a typical IEEE-79 power system. The results show that compared with the important sampling method, this method not only has higher accuracy, but also has faster calculation speed. The effectiveness and feasibility of the method are verified by an example.
【作者单位】：新能源电力系统国家重点实验室(华北电力大学);
【分类号】：TM732

【相似文献】