首页> 外文会议>International Conference on Fluid Power Transmission and Control(ICFP' 2005); 20050405-08; Hangzhou(CN) >Improving and Appraising on Training Algorithm of Neural Network in Soft Measurement of Dynamic Flow
【24h】

Improving and Appraising on Training Algorithm of Neural Network in Soft Measurement of Dynamic Flow

机译:动态流量软测量中神经网络训练算法的改进与评价

获取原文
获取原文并翻译 | 示例

摘要

To solve the measurement of dynamic flow in hydraulic servo systems by applying neural network is an important work in the field. The difficulty is the improving on the training algorithm. First of all, the performance of the training algorithm of neural network was sum-up and advanced. Then output weight optimization-hidden weight optimization, fusing and adaptive alternation algorithms were put forward respectively. Finally, their performances are validated and appraised through experimentations. The research work is very significative to the soft measurement technology of dynamic flow.
机译:应用神经网络解决液压伺服系统中动态流量的测量是该领域的一项重要工作。困难在于训练算法的改进。首先,总结并提高了神经网络训练算法的性能。然后分别提出了输出权重优化-隐藏权重优化,融合和自适应轮换算法。最后,通过实验对他们的表现进行了验证和评估。该研究工作对动态流量软测量技术具有重要意义。

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