基于神经网络的电磁干扰的预测

杨天鹏; 马齐爽; 谢清明

基于神经网络的电磁干扰的预测

1.
北京航空航天大学自动化科学与电气工程学院, 北京 100191
2. 中国航天科工集团公司飞航动力装置研究所, 北京 100074

详细信息

中图分类号: TN911.73
计量
- 文章访问数: 1710
- HTML全文浏览量: 214
- PDF下载量: 616
- 被引次数: 0
出版历程
- 收稿日期: 2012-12-10
- 修回日期: 2013-05-07
- 网络出版日期: 2013-05-31

Prediction of electromagnetic interference based on neural network

1.
School of Automation Science and Electrical Engineering, Beijing University of Aeronautics and Astronautics, Beijing 100191, China
2. China Aerospace Science and Industry Group, Aviation Aerospace Power Device Research Institute, Beijing 100074, China

摘要

摘要: 提出了一种应用神经网络预测电磁干扰的方法.针对遗传算法总体搜索能力较强但容易陷入局部最优,而模拟退火算法具有较强的局部搜索能力,又能避免搜索陷入局部最优解的特点,将模拟退火算法与遗传算法相结合,优化多层前馈(BP, Back Propagation)神经网络,获取最优的权值和阈值,并采用模拟退火的思想确定隐含层神经元的个数,进而建立基于神经网络的电磁干扰预测模型.以双平行导线间的电磁干扰问题为实例,明确干扰要素,建立训练样本和测试样本,对比期望输出和预测输出之间的误差,结果表明该方法可以准确有效地进行电磁干扰预测.
- 神经网络 /
- 模拟退火算法 /
- 遗传算法 /
- 电磁干扰
Abstract: A method to predict the electromagnetic interference using neural network was proposed. Genetic algorithm has the strong overall search ability but easy to fall into local optimum, and simulated annealing algorithm has the partial search ability, avoiding the search into local optimal solution. By using the simulated annealing algorithm and genetic algorithm combining, the back propagation (BP) neural network weights and thresholds were optimized, and the number of hidden layer neurons was determined by the simulated annealing ideas. Then, the neural network-based predictive models of electromagnetic interference was established. With the two parallel leads to electromagnetic interference matter as predicted instance, interference factors were identified, and the training and test samples were established. In contrast to the error between the expected output and the predicted output, the results show that the method can accurately predict the electromagnetic interference effectively.
- neural network /
- simulated annealing /
- genetic algorithm /
- electromagnetic interference

HTML全文

参考文献(9)

[1]	苏东林,雷军,刘焱,等.一种大型复杂电子信息系统电磁兼容顶层量化设计新方法[J].遥测遥控,2008,29(4):1-8 Su Donglin,Lei Jun,Liu Yan,et al.A novel method of top-level EMC design technology for large and complex electronic information systems[J].Journal of Telemetry,Tracking and Command,2008,29(4):1-8(in Chinese)
[2]	史峰,王小川,郁磊.MATLAB神经网络30个案例分析[M].北京:北京航空航天大学出版社,2010:11-18 Shi Feng,Wang Xiaochuan,Yu Lei.MATLAB neural network analysis of 30 cases[M].Beijing:Beijing University of Aeronautica and Astronautics Press,2010:11-18 (in Chinese)
[3]	张德丰.MATLAB神经网络应用设计[M].北京:机械工业出版社,2009:46-53 Zhang Defeng.MATLAB neural network design[M].Beijing:China Machine Press,2009:46-53 (in Chinese)
[4]	李海民,吴成柯.基于BP网络的遗传算法[J].模式识别与人工智能,1999(6):131-135 Li Haimin,Wu Chengke.Genetic algorithm based on BP network[J].Journal of Pattern Recognition and Artificial Intelligence,1999(6):131-135(in Chinese)
[5]	李敏强,徐博艺,寇纪淞.遗传算法与神经网络的结合[J].系统工程理论与实践,1999(2):65-69 Li Minqiang,Xu Boyi,Kou Jisong.The combination of genetic algorithms and neural networks[J].Journal of Systems Engineeri-ng Theory and Practice,1999(2):65-69 (in Chinese)
[6]	张晓绩,方浩,戴冠中.遗传算法的编码机制研究[J].信息与控制,1997,26(2):134-139 Zhang Xiaoji,Fang Hao,Dai Guanzhong.Genetic algorithm enco-ding mechanism[J].Journal of Information and Control,1997,26(2):134-139(in Chinese)
[7]	吴斌,涂序彦.快速遗传算法研究[J].电子科技大学学报,1999(1):49-53 Wu Bin,Tu Xuyan.The quick genetic algorithm research[J].Journal of University of Electronic Science and Technology,1999(1):49-53(in Chinese)
[8]	李永明,祝言菊,李旭.电磁兼容的人工神经网络预测技术分析[J].重庆大学学报,2008,31(11):1313-1322 Li Yongming,Zhu Yanju,Li Xu.Artificial neural networks-based prediction of electromagnetic compatibility problems[J].Journal of Chongqing University,2008,31(11):1313-1322(in Chinese)[9] Macqueen C N,Irving M R.An algorithm for the allocation of distribution system demand andenergyloss[J].IEEE Transacti-ons on Power System,1996,11(2):339-342
[9]	Kalyanmoy D,Amrit P,Sameer A,et al.A fast and elitist multiobjective genetic algorithm[J].IEEE Transactions on Evolutionary Computation,2002,6(2):182-193