论文部分内容阅读
针对单一因素难以有效对电网覆冰事故进行预警的现状,该文分析了导致电网覆冰的多个影响因素,如温度、相对湿度、降水量、风速、海拔高度等。以这些影响因素为条件属性、以电网覆冰事故情况为决策属性,在收集相关数据的基础上,建立了初始决策表。针对应急预警系统中信息不完备与信息冗余并存的特点,引入粗糙集模型对初始决策表进行了不完备信息处理、属性离散、属性约简以及预警规则的提取。最后得出不同影响因素取值组合条件下的预警规则,为电网覆冰事故预警提供参考。
Aiming at the current situation that a single factor can not effectively pre-warning the icing accident, several influencing factors such as temperature, relative humidity, precipitation, wind speed, altitude and so on are analyzed. Taking these influencing factors as the condition attributes and the grid icing accidents as decision attributes, the initial decision table was established based on the collection of relevant data. Aiming at the characteristics of incomplete information and redundant information coexisting in emergency early warning system, the rough set model is introduced to incomplete information processing, attribute discretization, attribute reduction and early warning rules extraction. Finally, we get the pre-warning rules under the combination of different factors, which provide a reference for early warning of icing accident in power grid.