基于优化支持向量回归的工业互联网安全态势预测方法  

A Method of Security Situation Prediction for Industrial Internet Based on Optimized Support Vector Regression

作　　者：胡向东[1] 吕高飞 白银 HU Xiang-dong;LÜGao-fei;BAI Yin(School of Automation,Chongqing University of Posts and Telecommunications,Chongqing 400065,China;School of Advanced Manufacturing Engineering,Chongqing University of Posts and Telecommunications,Chongqing 400065,China)

机构地区：[1]重庆邮电大学自动化学院,重庆400065 [2]重庆邮电大学先进制造工程学院,重庆400065

出　　处：《电子学报》2023年第2期446-454,共9页Acta Electronica Sinica

基　　金：教育部-中国移动科研基金(No.MCM20150202,No.MCM20180404);重庆市高校创新研究群体(No.CXQT20016)。

摘　　要：作为支撑智能制造等的新型工业基础设施,工业互联网的安全态势预测是一个关键性需求和应用新挑战.本文提出一种基于优化支持向量回归的工业互联网安全态势预测方法,即利用差分进化算法和自适应参数调整策略克服灰狼优化算法计算速度慢、优化精度低的缺点;再利用改进的灰狼优化算法优化支持向量回归参数;最后,利用最优化参数组合建立支持向量回归预测模型,实现工业互联网环境下的安全态势预测.仿真实验结果表明,在容许偏差为0.05或0.1时,本文方法的预测准确率分别为90%和100%,预测结果的绝对误差均小于0.07,相比于对比方法有更高的预测准确率和预测精度.The Industrial Internet is an emerging modern infrastructure for supporting smart manufacturing.Accurate security situation prediction of industrial Internet is nowadays still a key demand and challenge for industrial application.To this aim,a novel method of security situation prediction for industrial Internet based on optimized support vector regression is proposed in this paper.The proposed method is a three-step procedure:in the first step,an improved gray wolf optimizer algorithm,based on differential evolution and adaptive parameter adjustment strategy,with high calculation speed and optimization accuracy is proposed;then,the optimized parameters of support vector regression are obtained;after that,accurate security situation prediction model for industrial Internet is established.The simulation results show that the prediction accuracy rate of the proposed method are 90%and 100%when the allowable deviations are 0.05 or 0.1,respectively,and the corresponding absolute errors are less than 0.07,and thus the proposed method can enhance the accuracy rate and precision of prediction,in contrast to the existing methods.

关 键 词：工业互联网 安全态势预测 支持向量回归 灰狼优化算法 差分进化算法 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术] TN918.91[自动化与计算机技术—计算机科学与技术]