摘要
由于传统监测系统在配电网运行节点布置上过于分散,导致其不能高频率获取信息数据且存在监测结果误差,因此基于神经网络设计配电运行的智能监测系统。在硬件设计上,设计和规划网络连接总体架构,促使设备之间能够完成实时的数据接收和转化;在软件设计上,分段部署神经元监测节点,采集实时配电网运行信息数据,建立神经网络线性矩阵,转化数据类别完成智能监测。以持续运行的变压器热点温度变化值作为测试样本,在不同的运行负载条件下,该系统监测到的热点值与实际值基本一致,绝对误差能够控制在0~3℃,可以在设备达到热点极值时发出预警信息,延长设备的使用时间,具有较好的实际应用效果。
Because the traditional monitoring system is too scattered in the distribution network operation node layout,it can not obtain high-frequency information and data and there are errors in the monitoring results.Therefore,the intelligent monitoring system of the distribution operation is designed based on the neural network.In hardware design,design and plan the overall network connection architecture to promote real-time data reception and conversion between devices;in software design,neuron monitoring nodes are deployed in sections to collect real-time distribution network operating information and data,and build nerves Network linear matrix,transform data categories to complete intelligent monitoring.Taking the continuously running transformer hot spot temperature change value as the test sample,under different operating load conditions,the hot spot value monitored by the system is basically the same as the actual value,and the absolute error can be controlled within 0 ~ 3 ℃,which can reach the hot spot extreme in the equipment.It sends out early warning information when on duty,prolongs the use time of the equipment,and has a better practical application effect.
作者
王涛
WANG Tao(CNOOC(China)Co.,Ltd.,Tianjin Branch,Tianjin 300450,China)
出处
《信息与电脑》
2021年第24期182-185,共4页
Information & Computer
关键词
神经网络
配电运行
智能监测
neural network
power distribution operation
intelligent monitoring
