Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


dissertation_Shah.pdf (7.85 MB)

ETD Abstract Container

Abstract Header

Electrical Capacitance Volume Tomography (ECVT) Based Imaging and Velocimetry for Two-phase Flow Measurements

Chowdhury, Shah Mahmud Hasan
http://orcid.org/0000-0002-5203-6385
http://rave.ohiolink.edu/etdc/view?acc_num=osu1627023824224726

Abstract Details

2021, Doctor of Philosophy, Ohio State University, Electrical and Computer Engineering.
Electrical Capacitance Volume Tomography (ECVT) refers to three-dimensional (3D) imaging of flow media exhibiting two or more material phases in a region of interest (RoI) based on electric permittivity variations. Such multiphase flows are commonly found in industrial settings such as gas-oil-water flow in oil pipelines, gas-solid flows in various chemical processes etc. An ECVT sensor is comprised of metal electrodes flush mounted on an insulating pipe wall surrounding the RoI. They are arranged in a multi-layer pattern being able to capture the 3D variation in permittivity, i.e. both the cross-sectional (xy) and the axial (z) variation. Because of the wall, there exists only capacitive coupling between the electrodes and the RoI, which makes the modality non-intrusive in nature. Other advantages include conformal sensor shape, cheaper electronics due to low-frequency operation, and fast acquisition rate suitable for capturing fast moving flows. The mutual capacitance measured among the electrodes is used, with the aid of an appropriate image reconstruction algorithm, to reconstruct a 3D image of the permittivity distribution corresponding to the actual material distribution in the RoI. A limitation of ECVT is the poor image resolution compared to other imaging modalities, e.g. X-ray, which originates from the ill-posed nature of the inverse problem associated with ECVT. Flow velocimetry has been a topic of interest for decades. A lot of information about a flow can be derived if the velocity profile can be determined. Although ECVT can perform flow imaging, there has not been a convenient way of determining the velocity profile. Previous efforts include cross-correlating two successive images, which is computationally intensive and not robust as cross-correlation works in very simple cases only. Moreover, errors incurred in image reconstruction are compounded with cross-correlation which makes the situation worse. In this regard, a different velocimetry method is documented in this dissertation which is free of cross-correlation. The method exploits a mapping between the moving flow and the temporal change in capacitance. It formulates a new forward problem, which can be solved using the conventional image reconstruction methods used for imaging. This method overcomes the limitations with the previous cross-correlation based approach, however, it has its own shortcoming. It is more challenging than imaging as it deals with three unknowns, i.e. the velocity components in three axial directions, as opposed to only one unknown for imaging which is the permittivity. The number of known variables is, however, only one for both problems which is the capacitance. This difficulty often degrade the performance of velocimetry as compared to imaging in similar cases, which is documented in terms of simulation results. In addition to that, experimental results are included with various data conditioning methods such as data smoothing, outlier removal etc. Another contribution documented in this dissertation is the electronic scanning for ECVT, which aims improving the image resolution. For electronic scanning, a high electrode density sensor is employed as compared to a conventional sensor. Then, the electrode segments are connected and reconfigured dynamically to mimic physical rotation and displacement of the sensor on its axis. It is shown that electronic scanning is capable of increasing the resolution over conventional ECVT, however, at the cost of additional acquisition time because of the scanning. In this regard, a number of scanning strategies are described featuring different synthetic electrode shapes, and the optimum one is pointed out considering different acquisition times. Also, the strategies are implemented in SPICE to evaluate their feasibility in circuit aspects, e.g. signal to noise ratio (SNR), as compared to a conventional ECVT sensor. The conclusions derived from this analysis would serve future hardware implementation and testing of electronic scanning with adaptive ECVT sensors. Lastly, a study is included for volume fraction estimation of a two-phase flow based on ECVT capacitance data. The estimated volume fraction is intended to be used as a stopping criterion for an iterative image reconstruction method used throughout this dissertation.
Fernando Teixeira, PhD (Advisor)
Robert Burkholder, PhD (Committee Member)
Bradley Clymer, PhD (Committee Member)
112 p.
Electrical Engineering; Electromagnetics
Electrical capacitance volume tomography, ECVT, multiphase flow, two-phase flow, flow velocimetry, electronic scanning, fluidized bed

Recommended Citations

Citations

  • Chowdhury, S. M. H. (2021). Electrical Capacitance Volume Tomography (ECVT) Based Imaging and Velocimetry for Two-phase Flow Measurements [Doctoral dissertation, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1627023824224726

    APA Style (7th edition)

  • Chowdhury, Shah Mahmud Hasan. Electrical Capacitance Volume Tomography (ECVT) Based Imaging and Velocimetry for Two-phase Flow Measurements. 2021. Ohio State University, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1627023824224726.

    MLA Style (8th edition)

  • Chowdhury, Shah Mahmud Hasan. "Electrical Capacitance Volume Tomography (ECVT) Based Imaging and Velocimetry for Two-phase Flow Measurements." Doctoral dissertation, Ohio State University, 2021. http://rave.ohiolink.edu/etdc/view?acc_num=osu1627023824224726

    Chicago Manual of Style (17th edition)

osu1627023824224726
127
© 2021, all rights reserved.
This open access ETD is published by The Ohio State University and OhioLINK.