首页> 外文会议>15th world conference on non-destructive testing >Flow Pattern Identification based on Fuzzy Neural Network Using Multi-Electrode
【24h】

Flow Pattern Identification based on Fuzzy Neural Network Using Multi-Electrode

机译:基于多电极模糊神经网络的流型识别

获取原文

摘要

Investigation and control of flow phenomena in two-phase flow requires adetailed knowledge on the flow regimes and a number of phase flowproperties. Electrical capacitance sensor is shown here to be a robust toolfor this purpose, measured capacitance which reflect flow distribution areprocessed with fuzzy method. The inhomogeneity of sensors' sensitivitydistribution and medium distribution are fully considered, a novel methodfor flow pattern identification based on fuzzy neural network is presented.By self-organizing learning of Kohonen network and supervision learning ofBP neural network , The trained neural network has been applied toexperimental data for flow pattern identification. Through the casesinvestigated, experimental results show that it is feasible for flow patternidentification of core, annular and stratified flow.
机译:研究和控制两相流中的流现象需要 有关流态和许多相流的详细知识 特性。此处显示的电容传感器是一种强大的工具 为此,测得的反映流量分布的电容为 用模糊方法处理。传感器灵敏度的不均匀性 充分考虑了分布和介质分布,这是一种新方法 提出了一种基于模糊神经网络的流型识别方法。 通过Kohonen网络的自组织学习和监督学习 BP神经网络,经过训练的神经网络已应用于BP神经网络。 用于流型识别的实验数据。通过案例 经研究,实验结果表明该流型是可行的 识别核心,环形和分层流。

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