基于BP神经网络的电网安全性预警系统研发

对紧急情况下的电力系统做到快速预警且提供相应的应急预案,是一个急需解决的重要课题。从电力系统防灾容灾预警机制出发,针对电网故障的时空分布特性,采用BP神经网络对其安全隐患进行识别,构建了基于季节气候、地形地貌及自然灾害等影响电力系统安全状态的预警系统,并通过吉林电网常见故障对该系统的实用性进行了有效验证。

Development of a BP Neural Network Based Security Warning System for Electric Power Grid

It is an important issue to make a quick security-warning to electric power system and provide corresponding contingency program against some emergency cases. Based on the disaster prevention ＆ recover mechanism of electric power system, and taking the space-time distribution charaters into account, this paper develops a BP-neuraI network based security-warning system for electrc power grid. The warning system can identify the hidden troubles of the power system security according to the influencing factors such as season climate, geography, natural disasters, etc. The validity of the developed warning system has been proved with the faults oc- curred in Jilin Power Grid.