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%。

在模拟井实验的基础上，通过直接测量方法，对125mm垂直圆管中油水两相流流型进行了重新划分，给出 了垂直圆管中油水两相流流型的过渡边界，建立了用流量法和持水率法判别流型的标准，通过实验发现，小直径管中的流型划分结果不能成比例地应用到大直径管中，而且划分标准或界限也不能按比例放大或缩小。应用混沌分析方法，对油水两相流流动的持水率测量信号进行相空间重构处理，计算出不同流型下的分形维数和关联维数，实现了用混沌方法对两相流流型进行辨识。

Based on wavelet analysis, fractal theory and ECT (Electrical CapacitanceTomography) technique, a soft-sensing system of gas-solid fluidized bed is established, which can implement the on-line identificatin and the monitoring of the flow regime, and also can measure he voidage of the gas-solid fluidized bed. In this system, wavelet analysis and fractal theory are signals. of pressure flucturation; a 12-electrode ECT system is used to display the flow state and calculate the voidage of gas-solid fluidizd bed by suing a novel image reconstruction algorithm. Experimental results show that the soft-sensing system is effective. It can improve the accuracy of identification of flow regime and can measure voidage of gas-solid fluidized bed with an error less than 5%.