Two-Phase Flow Water Fraction Measurement Using Curvelet Transform and Electrical Capacitance Tomography

To address the ill-posed nature of the inverse problem in electrical capacitance tomography (ECT), this study introduces a sparse reconstruction algorithm based on the curvelet transform. Firstly, the curvelet transform is applied as the sparse basis to decompose binary images of ideal medium distribution across multiple scales and angles. This approach improves the sparsity of ECT grayscale value vectors. Subsequently, the iterative thresholding shrinkage algorithm is employed for the sparse reconstruction of curvelet coefficients, and the optimal sensitivity pseudo-inverse matrix is calculated using the singular value decomposition algorithm. Finally, an ECT measurement platform is established, and five typical real medium phantoms are reconstructed through the static experiment to confirm the effectiveness of the proposed algorithm. Experimental results demonstrate that the algorithm proposed in this paper effectively suppresses artifacts and provides clearly boundaries in reconstructed images. Meanwhile, it outperforms traditional reconstruction algorithms in terms of image correlation coefficient and image relative error. Specifically, compared to the Landweber iteration algorithm, the proposed algorithm shows an increase of approximately 37. 9% in image correlation coefficient and a reduction of approximately 17. 6% in relative error. Post-processed images can be utilized for water fraction measurement, with an average relative error below 10%, potentially as low as 4. 32%. The algorithm introduced in this paper effectively enhances the quality of reconstructed images in ECT, meeting the industrial requirements for precise natural gas extraction. © 2024 Xi'an Jiaotong University. All rights reserved.