Personal Account: Stanford Geothermal Workshop
AltaRock Energy’s Matt Uddenberg accompanied a group of staff to this year’s Stanford Geothermal Conference, where AltaRock’s Susan Petty and Trenton Cladouhos gave a presentation on the Newberry EGS Demonstration. This was Matt’s first experience with the workshop, so we asked him to share his experience.
Why I Went
I was fortunate enough to have my first experience attending the Stanford Geothermal Workshop this year. My background is in geology; I also have some experience in finance and engineering. From my perspective I am interested in how operators or modelers characterize, simulate and optimize a geothermal field. Coming to Stanford I was looking to see what research was taking place that would allow operators to optimize their fields.
At the conference I saw many interesting solutions to an array of problems within the industry. I also saw that the new direction of geothermal research is most prominently Engineered Geothermal Systems (EGS). There were many talks on a variety of subjects but the most popular by far was coupled thermal-mechanical models trying to accurately predict fracture formation and propagation. Overall, the unofficial theme of this year’s conference seemed to be: how do we control the stimulation process so we can make high-functioning man-made reservoirs?
Stanford is a beautiful place. Having spent the winter in Bend, Oregon and Seattle, Washington, the sunlight that greeted me when I arrived in Palo Alto forced me to smile. Walking into the alumni center where the conference was held, I was pleased to see that it, too, was filled with sunlight.
Inside, it was immediately clear that this conference drew participants from around the world. There were researchers from Turkey, China, France, Germany, the Philippines, and more. As the event coordinator, Roland Horne, gave the welcome speech, looks of recognition darted around the room from one participant to another. Despite the audience’s global footprint, many of these people had met before. Geothermal is still a small industry, one where operators and researchers from around the world are colleagues, even across continents.
Despite the close connections of the industry, Roland Horne gave a historical perspective on how much smaller the conference had been when it first began. On an international scale, geothermal projects have certainly grown drastically in the past 30 years, though the United States has not kept pace with nations like Germany and Australia when it comes to developing environmentally and economically promising technologies such as EGS.
Day 1: Reservoirs and Cost Reduction
Walking around the conference and poking my head into different sections I saw that most of the research was focused on the overall concept of improving the efficiency of reservoir creation – thus lowering the cost of EGS as a form of energy production.
The first talk I saw was on rock mechanics. The presenters were exploring how small scale stress fields are affected by a propagating fracture going through a material with a set of defined criteria and modeled as a discreet element model.
Such research aims to help developers design a stimulation procedure with a predictable output, eg. a fracture with a certain aperture and orientation. Doing this would enable one to design a fracture system that best meet the needs of a given resource.
By the end of the day I had attended 12 talks on 12 different ways to model fracture propagation. Many overlapped or were closely related, asking similar questions and trying related methods to answer those questions. This is proof of why these conferences are so valuable: they’re an opportunity for researchers to both share and to receive, to be informed by the learning of others, and to grow from the community’s collective work.
Day 2: EGS Field Results
The next day I went to talks primarily focused on field results in conventional and EGS fields. The common presentation style of these talks was a slide or slides depicting a three dimensional space filled with different sized and colored dots.
These slides depicted visualizations of seismic data. During the stimulation process small micro-earthquakes (MEQs) are generated when natural fractures undergo shear failure, when fractures spontaneously close, or for a variety of other kinetic reasons. Using a micro-seismic array developers are able to monitor these events and determine at where they occur and how large an event they are.
In this area, I’d like to see more research connecting what developers are able to observe and what modelers were able to model. In a heterogeneous reservoir how could one ascertain rock properties away from the wellbore if all one can observe are micro-seismic events? What was the value of these intricate complex models if the inputs could not be known?
Day 3: Cost Modeling and Geochemistry
The third day I went to talks focused on cost modeling and geochemistry. These talks were the highlight for me.
Especially high on the list was Cornell University’s presentation on its new cost modeling software ‘Geophires’. They use simple but adequate models to model the evolution of a field dominated by porous media, fractures or both. They use these models to come up with an optimal plant size and technology and then use all of this information to determine a levelized cost of energy. Cornell’s solution to cost modeling is clear, accurate and elegant.
I also really enjoyed a series of talks that day about characterizing a geothermal hot spring by geological means and modeling known parameters to inform a better understanding of the system.
Final Thoughts and What Comes Next
Seeing these talks felt refreshing, and I especially appreciated how these groups used different forms of analysis and modeling data to create a defined characterization of a system. These characterizations were improved by using the output of one form of analysis to inform the structure of a model characterizing the system. This model would in turn show possible discrepancies in the data or analysis and then the analysis could be tweaked to better represent the system. Iterating this process would give one a clear understanding of a system as a whole.
What I would like to see next, building on what I experienced at the Stanford Geothermal Workshop:
- Tools that iterate between models, data and analysis to characterize a system. Currently, micro-seismic data is not being used by models to characterize the systems that are being developed.
- A way to validate the models currently being produced.
- Development of models that will inform the characterization of reservoirs.
As I left the Stanford alumni center on the last day, it was still sunny and there was a slight breeze. It had been a treat to see the work underway by researchers around the world. It was also both humbling and energizing to realize how much work still needs to be done, and I’m excited to be a part of it.