首页> 外文会议>The Third international conference on pumps and fans(3rd ICPF) >The study of predicting centrifugal fans' performance by wavelet networks
【24h】

The study of predicting centrifugal fans' performance by wavelet networks

机译:小波网络预测离心风机性能的研究

获取原文
获取原文并翻译 | 示例

摘要

Based on the wavelet transform theory,wavelet network is proposed as an alternative to feedforward neural networks for approximating arbitrary nonlinear functions.Since backpropagation neural netowrks have bene of particulr use in the research of predicting the performance of centrifugal fans, we introduce wavelet network into the same area.An algorithm of backpropagation type with conjugate gradient is devised for wavelet network training.Experimental results hsow that wavelet network is a great improvement upon the BP neural netwrok in constructing the model of predicting the performanc eof centrifugal fans.
机译:基于小波变换理论,提出了小波网络作为前馈神经网络的一种替代方案,用于近似任意非线性函数。设计了一种具有共轭梯度的反向传播类型算法进行小波网络训练。实验结果表明,小波网络是BP神经网络在构建离心风机性能预测模型中的重大改进。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号