基于电容层析成像和模糊模式识别的油气两相流流型辨识新方法的研究

基于电容层析成像和模糊模式识别技术别提出了一种油气两相流流型辨识的新方法。建立了12电极电容层析成像流型自动识别系统，该系统利用Tikhonov正则化原理并结合SIRT(Simultaneous Reconstruction Techniques)算法进行图像重建。Tikhonov正则化原理用于克服图像重建过程中的不适定问题，SIRT算法用于提高最终重建图像的质量。根据流型的随机和模糊特性，提出了一种根据管截面重建图像进行流型辨识的模糊流型判别方法。研究结果表明，提出的流型辨识新方法是有效的。对于层状流、核心流、环状流、均相流等流型，流型辨识的准确率高于95％，辨识一个流型所用的时间小于0.3秒。对于塞状流，流型辨识的准确率高于90％。

Study on Flow Pattern Identification of Gas-oil Two-phase Flow Based on Electrical CaPacitance Tomography and Fuzzy Pattern Recognition

On the basis of electrical capacitance tomography technique and fuzzy pattern recognition, a new method for flow pattern identification of gas-oil two-phase flow was proposed. A 12-electrode electrical capacitance tomography system for flow pattern identification was developed. To obtain the quantitative information of two-phase flow, a hybrid image reconstruction algorithm which combines Tikhonov regularization theory with SIRT (Simultaneous Reconstruction Techniques) algorithm was used. The use of Tikhonov regularization theory is to solve the ill-posed problem of image reconstruction and the use of SIRT algorithm is to improve the quality of the final reconstructed image. Based on the stochastic and fuzzy characteristics of flow pattern, a new fuzzy flow identification method was presented. Experimental results show that the developed measurement system is successful. The accuracy of flow pattern identification of stratified flow, annular flow, core flow and homogeneous flow is more than 95% and the speed of flow pattern identification is less than 0.3s. For plug flow, its accuracy is more than 90%.