【作者】 冀海峰；

【导师】 李海青； 黄志尧；

【作者基本信息】 浙江大学 ， 控制科学与工程， 2002， 博士

【摘要】 两相流系统是一个复杂的非线性动态系统，相间存在着界面效应和相对速度，致使两相流参数检测的难度较大，在国际上，也是一个急待发展的探索研究领域。在两相流系统中，流型是一个重要的检测参数，它不但影响两相流的流动特性和传热传质等性能，而且影响两相流系统其它参数的准确测量，因此两相流的流型辨识是该领域中的一个重要研究方向。 本文基于小波分析技术，辅以电容层析成像技术、多传感器信息融合技术以及模糊处理技术，对气固流化床、气液两相流和油田两相流系统的流型进行了分析和辨识，并在自行设计和建立的装置以及大庆油田测井研究所模拟井装置上进行了实验验证，得到了一些有益的结论： ● 提出了一种将小波分析技术应用于气固流化床流型分析与辨识的新方法，并构造了相应的特征值。首先利用所采集到的压力波动信号的小波分解，对气固流化床流型的转变进行了分析，发现在不同流型下，经小波分解后的信号有着不同的表现形式。在此基础上，构造出特征值——尺度能量百分比，并建立了气固流化床从固定床到鼓泡床的判别准则，经实验验证所提出的方法是有效的，取得了良好的流型辨识效果。 ● 将电容层析成像系统成功地应用于气固流化床的流型(固定床、鼓泡床和湍动床)辨识中，并提出了一种新型的电容层析成像图像重建算法：利用Tikhonov正则化原理和ART算法相结合的组合型图像重建算法来实现整体空隙率的测量。此算法通过Tikhonov正则化克服了图像重建过程中的不适定性，从而保证了初始图像的准确性和有效性，并利用ART算法的收敛性，改善了图像的重建质量，从而提高了由图像重建得到的空隙率计算的准确度，为将电容层析成像系统应用于气固流化床的流型辨识中提供了更可靠的依据。 ● 为了进一步提高气固流化床流型辨识的准确率，本论文引入了多传感器信息融合技术。采用微差压和电容层析成像两种传感器，在利用压力波动信号的小波分解进行流型辨识以及利用空隙率波动信号进行流型辨识的基础上，对每个传感器的辨识结果利用D-S证据理论，在 摘 要 决策级上进行了合理的信息融合处理，得到最终的辨识结果。初步实 验表明，所采用的信息融合方法对提高流型辨识的准确率是有效的， 使气固流化床中固定床、鼓泡床和湍动床的辨识准确率分别提高到 95％、90％和 85％。．对于水平管气水两相流系统，本文建立了流型辨识的软测量模型：利 用简单易测的差压波动信号作为辅助变量，结合小波分析技术和模糊 模式识别方法，对主导变量——流型进行了辨识，提出了不同管径下 的气水两相流流型的综合辨识方法。经实验验证，所提出的方法是有 效的。．本文初步尝试了将小波分析技术应用于油田两相流系统中。对油田中 大管径竖直上升管气水两相流进行了初步分析，结果表明，不同流型 下，大管径竖直上升管的差压波动信号的小波分解在时间－尺度上的分 布存在着较大的差异，可以用来进行流型的分析和辨识。更多还原

【Abstract】 The two-phase flow system is a complex, nonlinear and dynamic system. There are interface effect and relative velocity between the two phases, so the measurement of two-phase flow is more difficult. In the parameter measurement technique of the two-phase flow, the identification of the flow regimes (patterns) is one of most important research fields. Heat transfer, mass transfer and the measurement accuracy of other flow parameters depend on the flow regimes.In this thesis, the author applies the wavelet analysis, ECT, multisensor information fusion and fuzzy technology to the identification of the flow regimes of gas-solid fluidized bed and gas-liquid two-phase flow. The experimental results on the self-designed equipment of gas-solid fluidized bed and gas-liquid two-phase flow show that the presented methods are effective. The main works are written as the following: A new method of the application of wavelet analysis to the identification of the flow regimes of gas-solid fluidized bed is proposed, and the eigenvalues used in identification of flow regimes are constructed. In different flow regimes of gas-solid fluidized bed, the decomposed signals of the pressure fluctuation signal have obvious differences. The eigenvalues of ’scale energy percent’ are obtained to identify the flow regimes of fixed bed and bubbling bed of gas-solid fluidized bed, and the rules of identification are also established. The experimental results show that the method is effective. The electrical capacitance tomography(ECT for short) system is successfully applied to the identification of flow regimes of gas-solid fluidized bed by the voidage measurement of gas-solid fluidized bed, and a new combinatorial image reconstruction algorithm which combines Tikhonov regularization theory with ART (Algebraic Reconstruction Techniques) algorithm is presented to implement the global voidage measurement of gas-solid fluidized bed. The use of Tikhonov regularization theory is to solve the ill-posed problem of image reconstruction and the use of ART algorithm is to improve the quality of the final reconstructed image that determines the precision of the voidage measurement. For improving the accuracy of identification of regimes of gas-solid fluidized bed, the multisensor information fusion technique is introduced. In the dissertation, the pressure transducer and the ECT transducer are applied to provide the different signals-pressure fluctuation signal and voltage fluctuation signal. Based on the results of the identification of flow regimes using wavelet decomposition of pressure fluctuation signal and voidage fluctuation signal, using the D-S evidence theory of information fusion, the author identifies the flow regimes of fixed bed, bubbling bed and turbulent bed successfully, and the results show that the method is effective. The accuracies of identification of fixed bed, bubbling bed and turbulent bed have improved to 95%, 90% and 85% respectively. In gas-liquid two phase flow system, the model of soft-sensoring technique is constructed: regarding the differential pressure fluctuation signal as the secondary variable and combining the wavelet analysis and fuzzy identification technique, the primary variable of flow regimes of gas-water two-phase flow is measured. An identification method applied to the gas-water two-phase flow with different pipe diameter is presented. The results show that the method is effective. The wavelet analysis is used in oil-flied two-phase flow system primarily. In the large-diameter upright pipe of Institute of Oil Logging of Daqing Oil Field, the wavelet analysis of differential pressure fluctuation signals along the pipe shows that the wavelet analysis can be used to identify the regime of the upright gas-water two phase flow and the deep research will continue.更多还原