Study On The Soft-field Characteristic And Image Reconstruction Algorithm Of Electrical Resistance Tomography

Two-phase flow system is widely applied in industry. The study of two-phase flow parameter measurement technique is an important research field.Process Tomography technique is an attractive development orientation for the two-phase flow parameter measurement. It provides an effective parameter measurement mean, by which 2D/3D information of the two-phase flow system can be obtained. Electrical Resistance Tomography(ERT) technique, which is based on resistance sensor principle, has the advantages of low-cost, non-intrusive, fast-responding. As a novel measurement technique, ERT can play an important role in the on-line parameter measurement of the two-phase flow system. However, the development of ERT is preliminary, many problems should be solved before its practical applications.Two of the key problems of ERT technique are studied in this dissertation. The main contributions are listed as follows:1. The state of the art of Process Tomography technique is reviewed. The forward problem and inverse problem of ERT are analyzed. The existing problems are discussed. Meanwhile, it is emphasized that the soft-field characteristic and the low resolution of image reconstruction are the key problems that result in the difficulty of its practical applications.2. It is very difficult to analyze the characteristics of the 2D non-homogeneous fields or 3D fields by using analytical mathematics directly. A finite element simulation software is developed for ERT system. The main functions include the self-division of finite element, the setting of flow pattern and the solution of the forward problem, et al.. Also, it is the foundation of other research work in this dissertation.3. The sensitivity field of ERT sensor is studied detailedly by using the simulation software. Through the analysis of the soft-field characteristic, it is found that the sensitivity field distribution is affected by the location, the size and the conductivity of the disperse phase medium. Meanwhile, the soft-field characteristics of ERT systems in different numbers of electrodes are studied, and the main soft-field characteristics of ERT systems are found. The results show that the soft-field characteristics are greatly reflected in the influence of distribution of mediums and the difference in conductivity of two-phase mediums on the distributions of the sensitivity field. The effect in the vicinity of the exciting electrode and detecting electrode is greater than that of the center of the pipe. The larger the absolute value of sensitivity in the location of the disperse phase medium and the size of the disperse phase medium, andthe greater the difference of conductivity between the disperse phase and continuous phase medium, the greater the effect of the disperse phase medium on the sensitivity distribution. In addition, the distributions of ERT sensitivity fields in different numbers of electrodes are different, but the soft-field characteristics are similar. These results provide references for removal of the soft-field problem and improving image reconstruction accuracy.4. Many ERT reconstruction algorithms are analyzed and compared by the simulation experiment data. Their advantages and disadvantages are pointed out. A new ERT mathematical model is developed for image reconstruction. A new ERT image reconstruction algorithm is presented. It is a combination of Tikhonov regularization and Algebraic Reconstruction Technique(ART). It is found that the algorithm is stable and the influence of data fluctuation on image reconstruction is eliminated. Furthermore, the qualities of the images reconstructed by using the proposed algorithm are improved.