Degree Type

Dissertation

Date of Award

2017

Degree Name

Doctor of Philosophy

Department

Mechanical Engineering

Major

Human Computer Interaction; Mechanical Engineering

First Advisor

Theodore J. Heindel

Second Advisor

Judy M. Vance

Abstract

Multiphase flows are used in a wide variety of industries, from energy production to pharmaceutical manufacturing. However, because of the complexity of the flows and difficulty measuring them, it is challenging to characterize the phenomena inside a multiphase flow. To help overcome this challenge, researchers have used numerous types of noninvasive measurement techniques to record the phenomena that occur inside the flow. One technique that has shown much success is X-ray imaging. While capable of high spatial resolutions, X-ray imaging generally has poor temporal resolution.

This research improves the characterization of multiphase flows in three ways. First, an X-ray image intensifier is modified to use a high-speed camera to push the temporal limits of what is possible with current tube source X-ray imaging technology. Using this system, sample flows were imaged at 1000 frames per second without a reduction in spatial resolution. Next, the sensitivity of X-ray computed tomography (CT) measurements to changes in acquisition parameters is analyzed. While in theory CT measurements should be stable over a range of acquisition parameters, previous research has indicated otherwise. The analysis of this sensitivity shows that, while raw CT values are strongly affected by changes to acquisition parameters, if proper calibration techniques are used, acquisition parameters do not significantly influence the results for multiphase flow imaging. Finally, two algorithms are analyzed for their suitability to reconstruct an approximate tomographic slice from only two X-ray projections. These algorithms increase the spatial error in the measurement, as compared to traditional CT; however, they allow for very high temporal resolutions for 3D imaging. The only limit on the speed of this measurement technique is the image intensifier-camera setup, which was shown to be capable of imaging at a rate of at least 1000 FPS.

While advances in measurement techniques for multiphase flows are one part of improving multiphase flow characterization, the challenge extends beyond measurement techniques. For improved measurement techniques to be useful, the data must be accessible to scientists in a way that maximizes the comprehension of the phenomena. To this end, this work also presents a system for using the Microsoft Kinect sensor to provide natural, non-contact interaction with multiphase flow data. Furthermore, this system is constructed so that it is trivial to add natural, non-contact interaction to immersive visualization applications. Therefore, multiple visualization applications can be built that are optimized to specific types of data, but all leverage the same natural interaction. Finally, the research is concluded by proposing a system that integrates the improved X-ray measurements, with the Kinect interaction system, and a CAVE automatic virtual environment (CAVE) to present scientists with the multiphase flow measurements in an intuitive and inherently three-dimensional manner.

DOI

https://doi.org/10.31274/etd-180810-5199

Copyright Owner

Timothy Burkgren Morgan

Language

en

File Format

application/pdf

File Size

262 pages

Share

COinS