摘要
近年来,随着社会经济的发展,社会对电力资源的需求越来越大,我国的电网构架也在不断的升级,伴随而来的就是电流互感器饱和的现象也会提升,严重的影响电力系统安全稳定运行。本文基于神经网络下的机器学习建立了一种自动编译器,对电流互感器在饱和状态下产生的畸形电流进行一定的编译,从而使其二次电流信号能很好的跟随一次电流信号变化而变化,大大的提高了变电站的继电保护的可靠性。
In recent years,with the development of social economy,the demand for power resources in society is increasing,and the power grid architecture in China is also constantly upgrading.The phenomenon of current transformer saturation will also increase,seriously affecting the safe and stable operation of the power system.This paper establishes an automatic compiler based on machine learning under neural networks,which compiles the abnormal current generated by current transformers in saturation state to a certain extent,so that their secondary current signal can well follow the changes of the primary current signal,greatly improving the reliability of relay protection in substations.
作者
蔡飚
CAI Biao(SiFang Relay Protection Co.Ltd.,Wuhan 430011,China)
出处
《电气开关》
2023年第5期70-72,79,共4页
Electric Switchgear
关键词
神经网络
电流互感器
饱和性
可靠性
neural network
current transformer
saturation
reliability
