摘要
对电力设备远程通信控制信号的入侵检测,能够更好的保障电网的稳定运行。远程通信控制信号的入侵检测,需要将全部的控制信号样本划分为多个独立的训练样本集和测试集,组建检测模型,完成控制信号的入侵检测。传统方法利用建立信号聚类目标函数,组建入侵检测模型,但忽略了对控制信号样本的划分,导致检测结果精度低。提出改进最小闭包球向量机的电力设备的远程通信控制信号入侵检测方法。首先对采集的控制信号进行归一化处理,将全部的控制信号样本划分为多个独立的训练样本集和测试集,对各个控制信号样本进行训练,然后组建电力设备的远程通信控制信号入侵检测模型,计算出标准正态分布的Gaussian径向基函数,完成对电力设备的远程通信控制信号入侵检测。实验仿真证明,所提方法检测精度较高,可以为保障智能电网的稳定运行提供科学的依据。
An intrusion detection method for control signal of remote communication in power equipment is proposed based on modified vector machine with minimum enclosing sphere.Firstly,the collected control signal is normalized,and the samples of all control signals are divided into many independent training sample sets and test sets.Each sample of control signal is trained.Then,an intrusion detection model is built,and the Gaussian radial basis function with standard normal distribution is worked out.Finally,the intrusion detection for control signals of remote communication in power equipment is completed.Simulation proves that the proposed method has high detection precision.It can provide basis for ensuring stable operation of smart grid.
出处
《计算机仿真》
北大核心
2017年第3期377-380,共4页
Computer Simulation
关键词
电力设备
控制信号
入侵检测
Power equipment
Control signal
Intrusion detection
