| 基于ECT投影数据的两相流流型辨识 |
| |
| 引用本文： | 颜华,刘艳红,张健. 基于ECT投影数据的两相流流型辨识[J]. 仪器仪表学报, 2006, 27(7): 702-705 |
| |
| 作者姓名： | 颜华 刘艳红 张健 |
| |
| 作者单位： | 沈阳工业大学信息科学与工程学院,沈阳,110023 |
| |
| 基金项目： | 辽宁省科委资金;辽宁省教育厅资助项目 |
| |
| 摘 要： | 提出一种基于电容层析成像（ECT）系统的两相流流型辨识新方法。ECT投影数据经特征参数提取前处理环节后馈人事先训练好的神经网络实现流型辩识。文中介绍了特征参数的提取、神经网络的构成以及训练方法。研究结果表明，该方法辨识精度高、辨识速度快，是两相流流型在线辨识的一种有效手段。
|
| 关 键 词： | 流型辨识 神经网络 特征提取 ECT投影数据 |
| 修稿时间： | 2005-04-01 |
Two-phase flow pattern identification based on ECT projection data |
| |
| Yan Hua,Liu Yanhong,Zhang Jian. Two-phase flow pattern identification based on ECT projection data[J]. Chinese Journal of Scientific Instrument, 2006, 27(7): 702-705 |
| |
| Authors: | Yan Hua Liu Yanhong Zhang Jian |
| |
| Abstract: | A new method of two-phase flow pattern identification based on Electrical Capacitance Tomography(ECT) system is proposed.The Characteristic parameters representing flow pattern are extracted from the ECT projection data,then fed into a pre-trained neural network to identify flow pattern. The characteristic parameter extraction,the neural network architecture and the network training are given.Research results show that the new method has good identification precision,fast identification speed,it is an effective tool for two-phase flow pattern on-line identification. |
| |
| Keywords: | flow pattern identification neural network characteristicextraction ECT projection data |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
