推荐星级:
  • 1
  • 2
  • 3
  • 4
  • 5

基于神经网络的引信电路故障诊断方法

更新时间:2020-10-26 12:48:38 大小:1M 上传用户:gsy幸运查看TA发布的资源 标签:神经网络引信电路 下载积分:1分 评价赚积分 (如何评价?) 打赏 收藏 评论(0) 举报

资料介绍

针对批量引信电路在生产和试验中的高可靠性、高效率的测试需求,提出了利用神经网络专家系统实现引信电路故障诊断的方法。通过介绍小口径计转数空炸引信的工作原理,分析了该引信电路常见的故障类型;按照故障类型从引信电路采集故障原始数据,经处理后作为神经网络训练的样本数据;建立了三层神经网络诊断模型,利用Matlab编程进行网络训练,给出了训练过程及误差曲线;使用样本数据验证了该神经网络专家系统的有效性与可靠性。配合自动测试系统,在测试过程中该方法能快速准确定位故障位置,显著提高批量引信电路测试效率及可靠性。

Aiming at the high reliability and high efficiency test requirement for producing and testing of the batch fuze circuits, the paper presented a fuze circuits fault diagnosis method based on neural network expert system. This paper first presented the working principle of the small caliber turns-counting proximity fuze to analysis the fault type of the fuze circuits, according to the fault type to collecting the raw data on fault diagnosis, after treatment, as a sample of the neural network training;A mathematical model was established by three layers neural network, network training with Matlab programming, and presenting the training process and error curve;Finally, the diagnosis result reliability of the neural network expert system was verified by the sample data. Combined with automatic test system, the method can fast and accurate to located the fault position, and the test efficiency of the batch fuze circuits is therefore highly improved.

部分文件列表

文件名 大小
基于神经网络的引信电路故障诊断方法.pdf 1M

【关注B站账户领20积分】

全部评论(0)

暂无评论

上传资源 上传优质资源有赏金

  • 打赏
  • 30日榜单

推荐下载