摘要
电力信息网络安全态势评估是当今网络安全领域研究中的热门领域。但现有基于神经网络的网络安全态势评估方法效率较低,且容易陷入局部最优导致评估精度不高。提出一种改进人工蜂群优化神经网络的网络安全态势评估方法。首先,通过引入混沌序列改进人工蜂群算法提高蜂群的多样性,使其具备更强大的全局搜索能力。然后,利用改进的蜂群算法代替反向传播算法来优化神经网络的各权值参数。最后,新方法对真实的电力信息网络攻击实验进行了安全态势评估预测。与传统的评估方法相比,基于改进的人工蜂群和神经网络的安全评估方法提高了安全评估精度,加快了收敛速度。
The network security situation assessment(NSSA)of power information system receives much attention in the field of network security research.However,the current neural network-based security situation assessment method is inefficient and easy to fall into local optimum,which results in the low assessment accuracy.A new NSSA method based on neural network which is optimized by an improved artificial bee colony(IABC)algorithm was proposed.Firstly,the chaotic sequence is introduced to the artifical bee colony(ABC)algorithm to improve the diversity of bee colonies,which could make ABC algorithm jump out of local optimum more easily.Then,the IABC algorithm is used to optimize the weight parameters of the neural network instead of the back propagation algorithm.Finally,a security situation assessment for a real network attack of power information system was deployed with the new method.Compared with the traditional assessment method,the method based on IABC and neural network improves the accuracy of NSSA and accelerates the convergence.
作者
于海
李峰
霍英哲
尹晓华
YU Hai;LI Feng;HUO Ying-zhe;YIN Xiao-hua(State Grid Liaoning Electric Power Co.,Ltd.,Shenyang 110006,China)
出处
《科学技术与工程》
北大核心
2021年第9期3642-3648,共7页
Science Technology and Engineering
基金
国家自然科学基金(51437003)。
关键词
网络安全态势评估
神经网络
人工蜂群算法
混沌序列
入侵检测系统
network security situation assessment
neural network
artificial bee colony(ABC)algorithm
chaotic sequence
intrusion detection system
