引用本文:邓小清, 王伦浪, 王刚, 蒲国林.一种基于BPNN的智能巡检异常预测模型的研究[J].西南大学学报(自然科学版),2019,41(10):142~148
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 81次   下载 122 本文二维码信息
码上扫一扫!
分享到: 微信 更多
一种基于BPNN的智能巡检异常预测模型的研究
邓小清, 王伦浪, 王刚, 蒲国林1,2
1. 四川文理学院 智能制造学院, 四川 达州 635000;2. 安康学院 电子与信息工程学院, 陕西 安康 725000
摘要:
在传统网络巡检方法中,网络异常发现主要基于单一参数进行阈值触发,误报率较高,效率低.为了高效准确地发现网络异常,提出了一种基于BPNN的网络异常预测模型.首先对采集系统采集的数据进行特征提取和初始化处理;然后,将初始化后的数据作为神经网络样本进行训练,根据误差阈值调整网络参数,确定网络结构;最后,在Matlab环境下进行仿真实验,将提出的BP神经网络模型用于网络异常预测,结果表明本文提出的方法对网络异常预测有较高的预测率.
关键词:  BPNN  智能巡检  异常预测
DOI:10.13718/j.cnki.xdzk.2019.10.019
分类号:TP183
基金项目:国家自然科学基金项目(61152003);四川省教育厅重点项目(16ZA03532016);四川文理学院重点项目(2016KZ002Z).
A Study on a BPNN-Based Anomaly Prediction Model for Intelligent Inspection
DENG Xiao-qing, WANG Lun-lang, WANG Gang, PU Guo-lin1,2
1. School of Intelligent Manufacturing, Sichuan University of Arts and Science, Dazhou Sichuan 635000, China;2. School of Electronic and Information Engineering, Ankang University, Ankang Shaanxi 725000, China
Abstract:
In traditional network inspection methods, the network anomaly discovery in intelligent patrol inspection is triggered by a single threshold based on performance parameters, with high false positive rate and low efficiency. In order to predict network anomalies effectively and accurately, we propose a network anomaly prediction model based on BPNN (BP neural network)in this paper. The collected data are extracted, preprocessed and trained as neural network samples, and then the network parameters are adjusted according to the error threshold to determine the network structure. A simulation experiment is carried out in Matlab environment, and the proposed BP neural network model is used for network anomaly prediction. The results show that the method proposed in this paper has satisfactory prediction rate for network anomaly prediction.
Key words:  BPNN(Back Propagation Neural Network)  intelligent inspection  anomaly prediction
手机扫一扫看