【作者】 王晓萍；

【导师】 童勤业； 张卫东；

【作者基本信息】 浙江大学 ， 测试计量技术与仪器， 2004， 博士

【摘要】 气固流化床是化工、能源等生产过程中的重要装置，在流化床建模、设计、开发和运行中需要获取流化床的流型和流型转换的信息。这对提高流化床的性能和传热、传质效率至关重要，流型参数的检测与识别一直是气固两相流参数监测的重要研究内容。 气固流化床压力脉动时间序列隐含着包括流型和过渡流型在内的两相流系统的动态行为信息，由于气固流化床在很宽的操作系统范围内是一个复杂的非线性时变系统，因此，本文采用现代非线性信息处理技术来进行流化床压力脉动信号的特征提取与流型识别方法的研究。以期能从新的角度更深入地认识流态化规律，探索流化床流型识别的新技术与新方法。 本文在大量实验数据基础上，将现代非线性信息处理方法包括希尔伯特黄变换(Hilbert-Huang Transform简写HHT)，互信息和信息传输以及模糊信息融合理论，首次应用到气固流化床压力脉冲时间序列的分析中，从理论和实验两个方面，系统地探讨了流化床流型的非线性识别原理，提出和实现了适合于工业现场和实验研究的三种流型识别方法。 首先采用非平稳、非线性信号处理的HHT变换，从压力脉动信号中提取一族内禀模态函数(IMFs)，建立时间—频率—能量三维希尔伯特黄谱图，从中发现了流型的转换与能量分布之间的联系；并进一步在实验研究中，计算得到了压力脉动信号不同阶次IMF能量转移与流型的定量关系。据此，提出了基于IMFs能量转移和不同频段能量分布的流型识别方法，该方法分析速度很快，具有十分良好的工程应用前景。 在讨论了相空间重构技术、复杂性参数算法和互信息理论基础上，计算了流化床系统内不同测点之间两两压力脉动信号之间的互信息传输时间序列，建立了以复杂性测变参数为其元素的信息传输矩阵(Information Transform Matrix简写ITM)，进一步定义了基于系统信息传输的信息传输平均值MITM(Mean of ITM)参数来表征气固流化床在不同流化状态时信息传输量的大小。 为了研究和识别流化床系统内的过渡流型，本文依据过渡流型的模糊性，在实验数据统计基础上应用模糊集理论，建立了不同过渡流型相对于确定性流型的隶属度函数。首次将过渡流型进行了量化，并采用模糊信息融合技术对多个特征参数和多个传感器信息进行特征层和决策层融合，提高了流型识别的准确率。从理论上和技术上为气固流化床流型和过渡流型的识别提供了新方法。 最后为使研究的流型识别方法能在实验室和工业现场得到应用，本文提出了流化床流型检测与识别系统的硬件和软件设计方案，并完成了三种非线性流型识别方法的应用软件包的编程与设计，为研究成果的应用打下了基础。更多还原

【Abstract】 Gas-solid fluidized beds have been assumed as important equipment in the production process in chemical and energy industry. Information of flow regimes and their transition is needed in the model building, design, research and development due to the importance of improving the performance of fluidized beds and increasing the heat transfer rate and mass transfer rate. The detection or identification of the parameters characterizing different flow regimes has long been a significant topic in the parameter measurement of two-phase system.Pressure fluctuation time series in gas-solid fluidized beds contain a lot of dynamic information, such as the information of flow regimes and transition states. Since gas-solid fluidized beds are a complex nonlinear time-varying system under wide operation, nonlinear theory is adopted and implemented to extract the characters of pressure fluctuation time series in gas-solid fluidized beds and find the new way to flow regime identification. Some new technologies and new ideas about the flow regime identification are discussed in this paper so as to cognize the fluidization from a new viewpoint in depth.Based on a large amount of experimental data, nonlinear analysis, such as Hilbert-Huang Transform (HTT), mutual information and fuzzy information fusion theory, are used in the time series analysis of pressure fluctuation in gas-solid fluidized beds for the first time. Application of nonlinear theory to the flow regime identification is discussed systematically from the aspects of theory and experiment. Finally, three kinds of regime-identification methods are proposed and proved to be suitable for using in practical industrial fields and experimental studies.In this study for the first time, the Hilbert-Huang Transform (HHT) is applied to analysis of the pressure-fluctuation in gas-fluidized beds. By using this new nonlinear and non-stationary signal-processing method, the Intrinsic Mode Functions (IMFs) are extracted from the pressure fluctuation signals and the time-frequency-energy distribution called Hilbert spectrum is established. From the spectrum the relationship between the flow transition and the energy distribution can be found. With the following study, the corresponding relations of the energy-transmission between different orders of Intrinsic Mode Functions (IMFs) under different flow states are obtained. Based on it, a new method, with which the energy-transmission between IMFs of pressure fluctuations andthe energy distribution in different IMFs can be used to detect the flow regime identification, is formed in this study. Because of the speed of analyzing by the new method, the prospect of its application to engineering will be fine.After the discussion of phased-space reconstruct technology, complexity parameter algorithm and mutual information theory, the mutual information of different time serials from different sensors at different regimes in gas-solid fluidized beds are calculated. The fluctuate complexity of those mutual time information are got, and then the Information Transform Matrix (ITM) is obtained. Further more, the Mean of ITM (MITM) is applied to defining the identification of different fluidized regimes.According to the previous study, transition states between different flow regimes in gas-solid fluidized beds show with obvious fuzzy characteristics. In this paper, fuzzy theory is implemented to identify the transition states in gas-solid fluidized beds. Membership functions are established to represent the membership grade of transition states for deterministic flow regimes. Transition states are quantified with fuzzy language and applying the fuzzy information fusion theory, multi parameters delivered from separated sensors are fused at characteristic level and multi decisions of each sensor are fused at the decision level. The experimental results indicate that the identification rate of flow regimes is improved. The fuzzy information fusion technology provides a new way to identify the transition states.Finally, for ap更多还原