现有电力系统警报处理方法完全没有利用或没有充分利用警报的时序特性。在人工智能研究领域发展起来的时序约束网络是一种能够清晰而直观地表示时序逻辑的有向无环图,适于解决计及警报信息时序特性的警报处理问题。在此背景下,构建了充分利用时序信息的警报处理解析模型。该模型不仅能够分析导致警报产生的具体事件以及该事件所在的时间区间,而且可以识别异常或遗漏的警报。基于时序约束网络中时间点和时间距离约束等相关概念,研究了事件发生时间的不确定性问题,避免了现有的计及时序特性的模型必须对时间精确定义的缺点。最后,用实际电力系统的警报处理案例对所提出的模型和方法进行了说明。 Existing power system alarm handling methods either do not utilize or make full use of the alarm’s timing features. The Timing Constraint Network, developed in the field of artificial intelligence, is a directed acyclic graph that can clearly and directly represent temporal logic and is suitable for solving the problem of alarm processing which accounts for the timing characteristics of alarm information. Under this background, an alarm processing and analysis model that takes full advantage of timing information is constructed. The model not only analyzes the specific event that caused the alert, but also the time interval of the event, and can identify abnormal or missing alerts. Based on the related concepts of time point and time distance constraints in the timing constrained network, the problem of the uncertainty of the time of the event is studied and the existing shortcomings of the time-based model must be avoided. Finally, the proposed model and method are illustrated by the actual power system alarm handling case.