摘要
为了提高变电设备运行的安全性与可靠性,并使报警系统能够迅速对变电设备的异常运行状态作出报警反应,开展了基于决策树的变电设备带电感应报警方法的研究.首先,采集与预处理变电设备带电感应数据;然后,提取与变电设备运行状态和故障模式相关的特征向量,构建决策树模型实时监测变电设备的运行模式和运行状态;最后,设定报警阈值,实现对变电设备故障的实时报警.实验测试结果表明应用所提方法后,报警反应速度加快,平均报警时间均在 2 s内.
Abstract
In order to improve safety and reliability of substation equipment operation and enable prompt response of alarm system to to abnormal operating conditions,a decision tree-based live induction alarm method for substation equipment was studied.First the live induction data of substation equipment were collected and preprocess.Second by extracting fea-ture vectors related to operational status and fault modes of substation equipment,a decision tree model was constructed to monitor the operational mode and status of substation equipment in real-time.Finally the alarm threshold was deter-mined to achieve real-time alarm for substation equipment faults.The experimental test results showed that the application of the proposed method achieved faster alarm response with average alarm time of less than 2 seconds.
NETL
NSTL
万方数据