Investigation Into System Optimization And Image Reconstruction Algorithm Of Multi-Phase Flow CT

Because of the widely existence of multi-phase flow in industrial process, multi-phase flow detection technology starts to draw more attention. Process tomography (PT) technology includes electronic tomography, X-ray tomography, ultrasound tomography etc, in which CT is widely accepted because of its advantage of high resolution. CT system is suitable to obtain global cross-sectional information in industrial field as well as in medical field.After a brief introduction of the basic theory of multi-phase flow detection technology and CT system, optimization and reconstruction algorithm are investigated. At last, practical experiment is conducted, whose result is analyzed and images are given.1) System SimulationGeant4 is a kind of high-energy particle transport simulation software toolkits based on Monte Carlo method, which is developed by European Organization for Nuclear Research (CERN). With the help of Geant4, the simulation model of Multi-Phase CT system is established and the physical process in the system is successfully analyzed. Based on optimization parameters, SPR (Scattering Photon Rate) and the quality of reconstruction image, the multiphase CT system is optimized and reconstructed. Experimental results show that the optimized CT system can restrain the scattering photon, and improve the quality of reconstructed image markedly.2) Image Reconstruction AlgorithmInvestigation is conducted into image reconstruction algorithm, such as Projection Algorithm and Iterative Reconstruction Algorithm. Data from simulation model are used for image reconstruction. Then some improvements are made from two different stand of view.The first one is Data Inverstion. Due to some inevitable reasons, some parts of the signal are stronger than others. Theoretically each serie of data possesses information about the whole object that we want to analyze. So comes a different idea for image reconstruction, deducing a series of data from the existing ones. Another improvement is based on the mathematical model of CT system. Some new methods for solving ill-posed equations are applied to the traditional ART algorithm.3) Practical ExperimentIn our group, a practical CT system is established. And several experiments are conducted with it. Then analysis is given to further our research.