基于电容层析成像技术重构图像的两相流流型识别  

Two-phase Flow Pattern Recognition Based on Electrical Capacitance Tomography Reconstructed Images

作　　者：王小鑫 王博 陈阳正 胡红利[3] WANG Xiao-xin;WANG Bo;CHEN Yang-zheng;HU Hong-li(Shaanxi Key Laboratory of Photoelectric Sensing Logging,Xi’an Shiyou University,Xi’an,Shaanxi 710065,China;Geophysical Exploration Technology and Equipment R&D Center,Xi’an Research Institute of China Coal Technology&Engineering Group Corp,Xi’an,Shaanxi 710077,China;State Key Laboratory of Electrical Insulation&Power Equipment,Xi’an Jiaotong University,Xi an,Shaanxi 710049,China)

机构地区：[1]西安石油大学陕西省油气井测控技术重点实验室,陕西西安710065 [2]中煤科工集团西安研究院有限公司地球物理勘探技术与装备研发中心,陕西西安710077 [3]西安交通大学电力设备电气绝缘国家重点实验室,陕西西安710049

出　　处：《计量学报》2020年第8期942-946,共5页Acta Metrologica Sinica

基　　金：国家自然科学基金(41874158);陕西省自然科学基础研究计划项目(2019JQ-822)。

摘　　要：针对工业多相流流型识别的需求,在过程层析成像技术的研究基础上,将重构图像信息即像素分布进行简单有效地分析处理,利用二维最大熵阈值分割技术及遗传算法优化的神经网络分类器对重构的图像进行处理以实现流型识别。将该方法在气固(空气/煤粉)两相流气力输送平台上对3种典型气固流型进行验证,实验结果表明:50组测试样本中识别正确率为94.7%。To realize the industrial multiphase flow pattern recognition,based on the research of process tomography,the reconstructed image information is analyzed and processed simply and efficiently,the two-dimensional maximum entropy threshold segmentation technique and the genetic algorithm optimized neural network classifier are used to realize the flow pattern identification.Three typical gas-solid flow patterns were verified on the two-phase flow pneumatic transport platform by the mentioned method.The experimental results showed that the recognition accuracy is 94.7%.

关 键 词：计量学 两相流 流型识别 电容层析成像 二维最大熵 阈值分割 神经网络 

分 类 号：TB937[一般工业技术—计量学]