With renewable energy on the rise and the demand for more and more sustainable sources increases, it is very important to consider all the different ways in which we can generate energy. In the US in 2017, only 20% of the energy consumed came from renewable or nuclear sources, and only 2% of that amount was from geothermal energy but there is much greater potential.
My research aimed to tackle some of the questions surrounding geothermal energy production in an economically feasible way. The overarching problem is how fractures affect production. To produce geothermal energy, you pump some sort of liquid (typically water) through a well that runs deep underground and the hot rock heats the water, creating steam that can spin turbines. However, it is very difficult to measure rock geometry in total and the presence of fractures or discontinuities in the media could be affecting the production in ways we do not see directly. Additionally, injecting water at high pressures could be opening up more fractures and further affecting the production rates.
To understand the entirety of the problem, I first spent some time familiarizing myself with the heat and mass transfer physics as well as the existing information in geothermal energy. My primary source for the mathematical aspects was the textbook Numerical Heat Transfer and Fluid Flow by Suhas V. Patankar. To learn more about geothermal production as a whole, I made use of various studies such as a paper about Thermal Drawdown-Induced Flow Channeling in Fractured Geothermal Reservoirs and some Department of Energy publications.
I investigated the effect of fractures on geothermal energy production using the program TOUGH2 (Transport of Unsaturated Groundwater and Heat), a piece of software that performs calculations over a mesh of geological materials to investigate the processes. I also used toughio to write my meshes and export my files to visualizations. After spending some time working on simple problems and familiarizing myself with the programs, I began working on my main problem.
By starting with a base mesh of shale and a well with an injection point, I was able to maintain all of the same parameters for every variation of my study besides the fracture geometry. I ran multiple iterations of multiple variations, and am still moving forward with this process to refine my results. My three main meshes were one with no fractures, one with only orthogonal fractures, and one with non-orthogonal fractures. All other properties are fixed and remain throughout the three main meshes.
As the programs and much of the physics are very new to me, this is still an ongoing project and the results have not yet been repeatable enough or reasonable enough to firmly say anything about the effect of fractures on geothermal energy production. As I continue to investigate the problem and iterate on the work I have done, I am converging upon something more concrete, but the problem of geothermal energy and fractures is complex and in order to be thorough, very time-consuming. Nevertheless, it is an interesting problem and it is clear that fractures do affect the production of geothermal energy in some capacity.
Research is hard. And virtual research is even harder, especially if you are learning entirely new concepts and software. This was my first research project and my first longterm career-oriented position, so I had no idea what to expect, but it was definitely more challenging and less straightforward than I imagined.
In the beginning, I was mostly just reading and working out the math to the numerical model and the types of problems I was going to be working on. This part of the project felt very similar to taking a challenging class to me, which was in some ways strange but also comforting. It was strange in the sense that this feels like school, but it is a job and comforting because it showed me that what we do in class really does set us up for further work. There were parts of this that were slightly challenging just because certain concepts were new to me, but for the most part my school bred study habits and the internet allowed me to understand and digest the information I was given. However, at times I felt a little lost because I didn’t understand what I would need to use for the next part of my project directly or if it was just good background information.
When I started working on the program, the aspects of the physics that I would need became much clearer to me, which was nice. But the program itself took me forever to learn. You have probably never heard of TOUGH2 -- I hadn’t -- but it is a very powerful tool used in energy geosciences, but it is not very user friendly. There is no user interface, the input files must be perfectly formatted, and the outputs are thousands of lines of numbers. And for a while, it made absolutely no sense to me. My simulations wouldn’t run, and I would spend hours digging through my files just to find the problem was one extra space. I also spent day after day reading through the user manual to figure out what parameters go where and what information I was missing. It was frustrating to say the least. But with the help of my host and another student, I slowly started to figure out the program and became increasingly comfortable with the formatting. Even as I figured out the inputs and ran many successful simulations, I couldn’t surmount the output files. I exported some of the data to excel files, but even that was a messy and inefficient process. Luckily another researcher at LBNL has an open-source program for exporting the files so that they can be visualized. But I had to learn a whole other piece of software, which took a while and I had some problems, but everyone was very supportive and quick to respond to my questions.
It continued on like this, with every problem solved there were more new problems that I had only just begun. And it was frustrating, especially because I didn’t have anyone to ask and get an immediate answer due to the virtual nature of our current world. But I learned a lot because of it, both about the programs and the problems I was working on as well as the research process and virtual work. My host said to me at one point that it is normal for research to take much longer than you expect, and that was very heartening. I really enjoyed working on this project, and I loved how individual and independent it was, even if it was at times frustrating. Where I find time, I will continue to work on my research and hopefully find some more solidified results, and I am very grateful to have had the opportunity to do this work this summer.