New Electrical Tomography Techniques And Their Applications

Electrical Tomography (ET) techniques have important research significances and broad application prospects in the research field of two-phase/multiphase flow measurement due to the advantages of rapid response, low cost, good safety and non-radiation. Up to date, the ET techniques are still not mature, although significant research progresses have been made. On the one hand, the conventional Electrical Resistance Tomography (ERT) systems are on the basis of contact conductivity detection technique, the electrodes are directly in contact with the conductive fluid, which may result in polarization effect and electrochemical erosion effect of the electrodes. On the other hand, dual-modality/multi-modality tomography techniques should be investigated in practical industry applications, because a single-modality tomography system can not meet the measurement requirements of some complex processes, e.g., gas-oil-water multiphase flows.Capacitively Coupled Contactless Conductivity Detection (C4D) technique can avoid the polarization effect and the electrochemical erosion effect of the electrodes, and hence, effectively eliminate the unpredictable measurement errors caused by contaminated electrodes. Therefore, the C4D technique not only provides an effective approach for overcoming the drawbacks of the conventional ERT systems caused by direct contact between the electrodes and the measured fluid, but also provides a new approach for developing the ET techniques.The aim of this dissertation is to study new ET techniques, which are on the basis of the C4D principle. On the one hand, a new C4D-based ERT technique termed Capacitively Coupled Electrical Resistance Tomography (CCERT) is presented and investigated. On the other hand, based on the C4D principle, a new integrated dual-modality Electrical Capacitance Tomography/Electrical Resistance Tomography (ECT/ERT) technique is proposed and investigated.The main research work and contributions in this dissertation are listed as follows:(1) Based on the C4D principle, the CCERT technique is presented and the feasibility of the CCERT technique is verified. A mathematical model of the CCERT sensor is established and the typical sensitivity distribution of the CCERT sensor is calculated and analyzed on the basis of Finite Element Method (FEM). The phase sensitive demodulation (PSD) method is applied to obtain the resistance measurements of the conductive fluid. A CCERT prototype is developed and the effectiveness of the CCERT prototype is demonstrated.(2) To overcome the ill-posedness of image reconstruction of CCERT, on the one hand, combining Tikhonov regularization method with Simultaneous Iterative Reconstruction Technique (SIRT), a new hybrid image reconstruction algorithm (the first new image reconstruction algorithm) for CCERT is proposed. On the other hand, another image reconstruction algorithm (the second new image reconstruction algorithm) for CCERT is proposed with the combination of Levenberg-Marquardt (L-M) method and Simultaneous Algebraic Reconstruction Technique (SART). Image reconstruction experiments are carried out with a 12-electrode CCERT prototype. The experimental results show that the images reconstructed by the proposed image reconstruction algorithms are satisfactory and are in accord with the actual distributions of two-phase flows. The research work also indicates that the proposed image reconstruction algorithms are more suitable for image reconstruction of CCERT, comparing with the conventional image reconstruction algorithms of ECT and ERT. Compared with the first new image reconstruction algorithm, the second new image reconstruction algorithm can automatically determine the regularization parameter. When the measurement errors are small, it is more approprite to adopt the image recontrustion algorithm based on L-M and SART, which may have better reconstruction performance under this circumstance. When the measurmenet errors are large, it is more approprite to use the image reconstruction algorithm based on Tikhonov regularization and SIRT, which has better stability.(3) Flow pattern identification of gas-liquid two-phase flow using CCERT is studied. A new flow pattern identification method for gas-liquid two-phase flow, which is on the basis of CCERT and the k-nearest neighbor (k-NN) algorithm, is proposed. With the 12-electrode CCERT prototype, flow pattern identification experiments are carried out. The experimental results show that the identification accuracies of three typical gas-liquid two-phase flows, such as the homogeneous flow, the stratified flow, and the annular flow, are 90.6%,91.3% and 86.1%, respectively.(4) Based on the C4D principle, a new dual-modality ECT/ERT technique is proposed. The equivalent circuit model of the dual-modality ECT/ERT technique is analyzed and investigated. A new integrated dual-modality ECT/ERT system, which is on the basis of the digital phase sensitive demodulation (DPSD) method, is developed. Image reconstruction of the new dual-modality ECT/ERT system is implemented by the linear back-projection algorithm. The experimental results demonstrate the feasibility of the proposed dual-modality ECT/ERT technique and the effectiveness of the developed dual-modality ECT/ERT system.(5) To automatically implement the modality switching between two different modalities (ECT modality and ERT modality), a modality recognition method, which is on the basis of the k-NN algorithm, is proposed. The research results indicate that the proposed modality recognition method is effective and can determine the correct modality.