电网故障最优诊断的查询方法

发布时间：2018-02-27 03:36

本文关键词： 基于模型的诊断电网故障诊断最优诊断因果关系电气信息告警信息贝叶斯理论　出处：《高电压技术》2017年04期 　论文类型：期刊论文

【摘要】：针对传统专家系统在电网故障诊断应用中的局限性,提出一种基于模型诊断的电网故障最优诊断的查询方法。该方法按照基于因果关系的诊断思想,得到预设故障输出对应的预备候选诊断,然后根据故障后的电气信息从匹配的预设故障输出中确定候选诊断,最后,运用贝叶斯理论计算候选诊断的故障概率,并查询最大概率的候选诊断作为电网故障的最优诊断。该方法通过离线获得预备候选诊断,在线确认候选诊断的手段,缩减了诊断的时间,在利用贝叶斯定理处理诊断的不确定性时,将告警信息引入到模型诊断逻辑框架内计算元件的实际故障概率,提高了诊断的准确性。算例分析结果验证了所提方法的有效性和可行性。
[Abstract]:In view of the limitation of traditional expert system in the application of power network fault diagnosis, a query method of optimal fault diagnosis based on model diagnosis is proposed, which is based on the idea of causality diagnosis. The candidate diagnosis corresponding to the preset fault output is obtained, and then the candidate diagnosis is determined from the matched preset fault output according to the electrical information after the fault. Finally, the fault probability of the candidate diagnosis is calculated by using Bayesian theory. The candidate diagnosis of maximum probability is used as the optimal diagnosis of power network fault. This method can obtain the candidate diagnosis offline, confirm the candidate diagnosis online, and reduce the time of diagnosis. When Bayesian theorem is used to deal with the uncertainty of diagnosis, the alarm information is introduced into the logic framework of model diagnosis to calculate the actual failure probability of components. The accuracy of diagnosis is improved and the effectiveness and feasibility of the proposed method are verified by an example.
【作者单位】：东北大学信息科学与工程学院;
【基金】：国家自然科学基金(61433004) 中央高校基本科研业务费专项资金(N130604001)~~
【分类号】：TM711

【相似文献】