基于抗衰减加权神经网络的网络设备损伤检测  

Network Equipment Damage Detection Method Based on Weighted Resistance Attenuation Neural Network Algorithm

作　　者：刘欢[1] 蔡红[1] 

机构地区：[1]上海行健职业学院,上海200072

出　　处：《科技通报》2015年第12期123-124,192,共3页Bulletin of Science and Technology

摘　　要：由于联网设备设计的特殊性,在网络状态下损伤时不易发现,还因为其功能的不同而处于不同的工作环境下,环境的差异对不同联网设备的影响不同,使得联网设备出现损伤的概率不同。提出了一种基于抗衰减加权神经网络算法的网络设备损伤检测方法。为了减少神经网络模型的衰减,引入自适应粒子群方法,对网络状态下联网设备进行分类处理,将其作为响应联网设备损伤的输入数据,之后采用抗衰减加权神经网络对网络状态下联网设备损伤进行检测,最后进行仿真实验。实验结果表明,利用本文算法进行网络设备损伤检测,可以极大地提高损伤检测的准确性和效率,从而满足实际需求。Due to the particularity of networking equipment design, in the condition of network when the injury is not easy to find, but also because of its function varies in different work environment, the environment of the differences on the influ-ence of different networking equipment. This paper proposes a weighted neural network algorithm based on resistance atten-uation network equipment damage detection method. In order to reduce the attenuation of neural network model, the intro-duction of the adaptive particle swarm method of classifying network state of networking equipment processing, as a re-sponse to a networking equipment damage of input data, using weighted resistance attenuation after neural network to test the networking equipment damage under the network condition, finally the simulation experiment, the experimental results show that the algorithm presented in this paper damage detection for network equipment, can greatly improve the accuracy and efficiency of damage detection, so as to meet the actual demand.

关 键 词：联网设备 网络状态 损伤检测 加权神经网络 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]