基于声发射的气固两相流流型识别

发布时间：2019-05-08 22:08

【摘要】：气固流化床中流型的识别是两相流相关参数检测的重要内容。利用声发射检测技术对气固流化床进行数据采样,结合数理统计、小波算法以及信息熵等方法提取气固流化床中不同流型的特征参数。由实验结果可知,此方法提取出的特征参数真实反映了典型气固流型下的动力学特征,在流化床中,鼓泡床、湍动床和快速流化床在时域、频域有着显著的区别。将小波能量、数理统计的结果以及小波包分解后的信息熵导入支持向量机(SVM)中,进行气固流型识别,正确率85%。因此,声发射检测技术与支持向量机相结合的方法可用于气固两相流流型的识别,具有一定的可行性。
[Abstract]:The identification of flow patterns in a gas-solid fluidized bed is an important part of the measurement of correlation parameters of two-phase flow. The acoustic emission detection technique is used to sample the data of gas-solid fluidized bed. The characteristic parameters of different flow patterns in gas-solid fluidized bed are extracted by means of mathematical statistics wavelet algorithm and information entropy. The experimental results show that the characteristic parameters extracted by this method reflect the dynamic characteristics of a typical gas-solid flow model. In fluidized bed, bubbling bed, turbulent bed and fast fluidized bed have significant differences in time domain and frequency domain. The wavelet energy, the result of mathematical statistics and the information entropy of wavelet packet decomposition are introduced into support vector machine (SVM), and the correct rate of gas-solid flow pattern recognition is 85%. Therefore, the combination of acoustic emission detection technology and support vector machine can be used to identify gas-solid two-phase flow patterns, and it is feasible to some extent.
【作者单位】：内蒙古科技大学信息工程学院;
【基金】：内蒙古自治区自然科学基金资助项目(2014MS0609) 国家自然科学基金资助项目(61463041,61362023,61463042)
【分类号】：O359

【相似文献】