Current Research Interests
{Control of Complex Systems}
I am interested in complex systems. And, more specifically, controlling these systems.
​
​
Complex systems, like the economy or the human body, have properties that are both intellectually fascinating and frustrating for a control engineer: With such systems, following a traditional engineering approach to control design—which requires precise and complete mathematical modeling and analysis upfront—is impossibly demanding.
​
Instead, if we hope to establish working strategies for controlling or managing complex systems in practice, I believe we need to consider a new approach to control design. One that does not rely on our ability to fully understand and model the system at the onset. And here, I further believe, we can take inspiration from every-day human ability: The ability to expertly drive a car in the absence of a fixed model of the dynamics or drawn-out calculations, for example.
​
In the research I am newly pursuing, I hope to develop human-inspired control design strategies for complex systems.
This is an ambitious goal. But there is plenty of motivation, far beyond the intriguing academic aspects: Establishing a reliable route for controlling complex systems could advance manufacturing of pharmaceuticals—a trillion-dollar industry—, help us manage the economy as a whole, and even directly save lives.
​
​
​
Check out my Current Research Interests page for more on complex systems, human-inspired control, and the applications I am exploring. Or, reach out to me directly over email.
There are 56 million gallons of legacy nuclear waste currently stored in large tanks at the Hanford Site in the state of Washington. These tanks were not designed for the type of long-term storage they have been used for; unfortunately, some of the tanks have begun to, or have in the past, leaked waste into the surrounding soil.
​
Recognizing the serious threat to the environmental and human-health, more than 30 years ago the U.S. Department of Energy initiated what has become the nations largest cleanup effort—with tank waste retrieval and treatment expected to span to 2047 and cost more than $50 billion.
​
Separation-operations can aide in this cleanup.
At Georgia Tech, my PhD advisors, Professors Grover, Kawajiri, and Rousseau, and I explored the possibility of using crystallization as a method for separating non-radioactive salts from the Hanford Nuclear waste.
​
Nuclear waste at Hanford, it turns out, contains large amounts (up to 50%) of dissolved sodium-sulfate, -nitrate, and other salts. These components are innocuous by themselves. If they can be effectively separated, they may be disposed of by more conventional means—saving time, reducing cost, and enabling a more focused effort on containing the radioactive portions of the waste.
But the separation is challenging and requires careful control of the crystallization. This became the focus of my Doctoral Research and something that I, and my PhD advisors, spent multiple years and many hours thinking about and working on.
In 2015, our work towards this application was recognized by the Department of Energy's Office of Nuclear Energy and awarded an Innovation in Fuel Cycle Research Award in the category of Material Recovery and Waste Form Development. The operation remains under consideration for application to Hanford waste cleanup.
Doctoral Research | Application
{Separation by Crystallization to Aid in the Cleanup of Nuclear Waste}
Doctoral Research | General Framework
{A Framework for Understanding and Controlling Crystallization}
Working to establish a crystallization process to aide in nuclear waste cleanup placed an emphasis on a particular technical problem: Control over the size of crystals produced by batch cooling crystallization. ​The application also instilled a mentality into our research: We were interested in developing techniques that could be applied in practice.
​
In this vein, my Advisors and I devised a new framework for understanding batch cooling crystallization dynamics and controlling the mean size of crystals.
This framework represents a departure from the prevailing, population-model-based, paradigm for establishing control of crystallization. It is measurement-based, provides a different and intuitive perspective on the process dynamics and, most significantly, enables the development of feedback control strategies that can be applied in practice. These control strategies were demonstrated for real systems at the lab scale.
​
Check out my Thesis, Doctoral Research page, or contact me to learn more.
Welcome to the research page of Daniel Griffin.
​
​
This site describes my current research interests and organizes results
from my doctoral research.