摘要
为解决传统方法下电力供应链分析识别准确率低、泛化能力和鲁棒性差的情况,提出一种利用深度学习中胶囊网络算法来构建电力供应链风险评估模型。为了验证该算法在电力供应链风险中的识别效果,通过与传统方法进行对比,结果显示提出的算法在准确度、泛化能力以及鲁棒性等方面都取得了可竞争的实验效果。
In order to solve the problems of low accuracy,generalization ability and poor robustness of power supply chain analysis and recognition under traditional methods,a capsule network algorithm is proposed by deep learning to build a power supply chain risk assessment model.In order to verify the identification effect of the algorithm in the risk of power supply chain,comparison is made with traditional methods.The results show that the algorithm proposed has achieved competitive experiment results in terms of accuracy,generalization ability,and robustness.
作者
冯曙明
胡天牧
杨永成
潘晨溦
高正平
沈键
FENG Shuming;HU Tianmu;YANG Yongcheng;PAN Chenwei;GAO Zhengping;SHEN Jian(Jiangsu Electric Power Information Technology Co.,Ltd.,Nanjing 210024,China;State Grid Jiangsu Electric Power Co.,Ltd.,Nanjing 210024,China;Material Branch of State Grid Jiangsu Electric Power Co.,Ltd.,Nanjing 210024,China)
出处
《微型电脑应用》
2022年第8期32-34,共3页
Microcomputer Applications
基金
国网江苏省电力有限公司项目(J2020063)。
关键词
深度学习
机器学习
胶囊网络
电力系统
风险评估
deep learning
machine learning
capsule network
power system
risk assessment
