摘要
电网安全数据子集中缺少处理目标函数的松弛变量,导致抽检数据处理线程吞吐量过高,为此,提出支持向量机的电网安全抽检数据分析方法。根据支持向量机提取电网安全抽检数据特征,计算特征子集上的信息维度;利用该信息维度设计支持向量机数据分析函数,采用敏感函数训练损失的电网安全数据子集,处理目标函数松弛变量,构建电网安全抽检数据分析模型,完成电网安全抽检数据分析。实验结果表明,该方法的抽检数据处理线程吞吐量最高为6000条/s,说明电网安全抽检数据分析效果较好。
The lack of slack variables to process objective function in the power grid security data subset leads to the high throughput of the sampling data processing thread. Therefore, this paper proposes a power grid security sampling data analysis method based on Support Vector Machine. According to Support Vector Machine, the features of power grid security sampling data are extracted, and the information dimension on the feature subset is calculated. This information dimension is used to design the Support Vector Machine data analysis function, and the lost power grid security data subset is trained by the sensitive function. Besides, the slack variables of the objective function is processed to construct the power grid security sampling data analysis model, and the power grid security sampling data analysis is completed. The experiment results show that the maximum throughput of sampling data processing thread of this method is 6000/s, which shows that the effect of power grid security sampling data analysis is good.
作者
骆星智
赵钰
孙磊
宫杨非
LUO Xing-zhi;ZHAO Yu;SUN Lei;GONG Yang-fei(State Grid Anhui Electric Power Co.,Ltd.,Hefei 230041,China)
出处
《信息技术》
2023年第1期126-130,136,共6页
Information Technology
关键词
支持向量机
电网安全
抽检数据
敏感函数训练
数据挖掘
Support Vector Machine
power grid security
sampling data
sensitivity function training
data mining
