14 December 2020 in IEEE Transactions on Instrumentation and Measurement
The sensitivity distribution of the Electromagnetic Tomography (EMT) system based on Tunneling Magneto Resistance (TMR) is quite different from the traditional coil measurement EMT system due to the change of detection sensor. The sensitivity of the TMRs-EMT system is higher only near the TMR sensors, but lower at other locations in the region of interest, which results in more serious ill-conditioned problem of image reconstruction. Focusing on the ill-conditioned problem, the Landweber algorithm is optimized by using an appropriate weight and an appropriate relaxation strategy in this paper. By verifying the consistency of variation of spectral radius of iterative matrix and variation of condition number for Landweber method, a new weight matrix is introduced for the general Landweber method to reduce the condition number. On this basis, an optimal relaxation factor is determined based on the principle of minimizing the spectral radius of the newly derived iterative matrix. By comparing the imaging results of the new method with the Landweber with a relaxation factor of 1, it can be seen that the average correlation coefficient with the new approach for the five models used in this paper is 0.8466, which is much higher than 0.7633 of the Landweber with a relaxation factor of 1. Meanwhile, the proposed method has a faster imaging speed. The convergence analysis of the two methods shows that the new relaxation factor determination strategy can overcome the semi-convergence problem of Landweber.
27 July 2022 in Measurement Science and Technology
Electromagnetic tomography and electrical resistance tomography (EMT/ERT) dual-modality sensors can combine the advantages of single-modality sensors and detect the distribution of a three-phase medium in a gas–liquid–solid three-phase fluidized bed containing magnetic catalyst. However, the detailed design and optimization of the structure of the EMT/ERT dual-modality sensor arranged in the same cross-section are lacking. In this paper, this problem is solved from the perspective of mutual interference with both EMT and ERT. In numerical simulations, the boundary measurement data and the quality of reconstructed images are compared between single-modality and dual-modality. It is concluded that the EMT has no interference with the ERT, whereas the EMT is interfered by the ERT with the increase of duty ratio. In addition, it is determined that the position of the center of the coil facing the center of the electrodes and the electrode duty ratio of 0.5 are the optimal dual-modality structure. A dual-modality sensor experiment is carried out, and the optimal structure of dual-modality sensor is further validated.
18 May 2022 in IEEE Transactions on Instrumentation and Measurement
The electrical resistance tomography and electromagnetic tomography (ERT-EMT) dual-modality system can be used to reconstruct the image of cross-sectional three-phase distribution for gas–liquid–solid fluidized bed by fusing the reconstructed images of ERT and EMT. However, due to the small size of solid particles, the complexity of fluidization process, and the inconsistency of media distribution recognized by ERT and EMT, the existing fusion algorithms of dual-modality electrical tomography, which either divide the media distribution into binary distribution or use the consistency of the media distribution to be recognized to improve the image quality, are not applicable. Aiming at the problem, a new image fusion method based on guided image filtering (GIF) and image statistics is proposed in this article. First, the GIF method is used to decompose the images of ERT and EMT into base layers containing large-scale variations in intensity and detail layers with small-scale details. Then, the weights corresponding to the base layers and detail layers are calculated with the image statistics method and the GIF method, respectively. Through the process, the useful information in the reconstructed images of ERT and EMT is fully utilized. The simulation and experimental results demonstrate that the proposed fusion method can accurately reconstruct the gas–liquid–solid distribution.
4 August 2022 in IEEE Transactions on Instrumentation and Measurement
For the gas–liquid–solid fluidized beds with magnetic catalyst, the solid fraction can be estimated through the reconstructed permeability distribution using the electromagnetic tomography (EMT) based on the tunneling magneto resistance (TMR) sensor. Considering that the existence of mixed phase and obscure phase interface leads to the inaccurate estimation of phase fraction, a new approach to estimating the solid fraction is proposed. Firstly, the equivalent magnetic circuit analysis method is introduced to abstract the magnetic field distribution, and the relationship between the boundary measurement data and the equivalent permeability of the cross section is derived. Then, the effective medium model is employed to estimate the solid fraction of the cross section based on the equivalent permeability. The solid fraction estimation method based on the above proposed method is more accurate than that based on the gray level of reconstructed images, especially when the permeability distribution is complex. When the true solid fraction is in the range of 0%–100%, the average of the absolute error of simulation and experiment is 2.5% and 4.8%, respectively.
3 December 2024 in Measurement Science and Technology
Imaging plays a vital role in enabling the visualization and analysis of objects and phenomena across various scientific disciplines and industrial sectors, spanning a wide range of length and time scales. This roadmap presents a critical overview of 13 industrial imaging techniques, which are organized into three thematic categories according to their applicability to either solid, fluid, or both solid and fluid targets. The objectives of this roadmap are to highlight challenges and provide perspectives for next-generation imaging systems, which can serve as a guide to researchers and funding agencies in identifying new prospects. It has been found that the common challenges of imaging techniques have remained fundamentally unchanged over the years, including improving coverage, speed, resolution, accuracy, and robustness; however, there is an increasing reliance on data-driven or artificial intelligence (AI) approaches. Addressing these challenges necessitates easy access to high-performance computing resources. Notably, the trustworthiness and traceability of AI approaches should be enhanced through the sharing of benchmarking data, balancing with physics-based techniques, and the adoption of more explainable AI.
28 February 2023 in IEEE Transactions on Instrumentation and Measurement
Electromagnetic tomography based on tunneling magnetoresistance (TMR-EMT) is a new type of electrical tomography technology. In contrast to the traditional EMT that uses the magnetic flux obtained from the coil to reconstruct the electrical conductivity distribution in the region of interest (ROI), the TMR-EMT reconstructs the magnetic permeability distribution in the ROI by the boundary magnetic induction intensity in the direction of the sensitive axis of the uniaxial TMR (Uni-TMR). However, the Uni-TMR measurements are influenced by the direction of boundary magnetic induction intensity at the Uni-TMR location; when the direction is orthogonal to the sensitive axis direction, the Uni-TMR cannot measure the boundary magnetic induction intensity, and such locations are named TMR-EMT blind spots in this article. The blind spots will exacerbate the ill-condition of inverse problem solving of TMR-EMT and even lead to failure of imaging. Aiming at the problem of TMR-EMT blind spots, this article proposes a solution to measure the boundary magnetic induction intensity using biaxial TMR (Bi-TMR). The Bi-TMR consists of two Uni-TMRs, whose sensitive axis directions are orthogonal to each other. The Bi-TMR measurement is obtained by squaring the sum of the squares of these two Uni-TMR measurements, which is unaffected by the boundary magnetic induction intensity direction. Simulation and experimental results show that this solution does not have the blind spots and can obtain complete boundary magnetic induction intensity in plane.