As part of the Moore-Sloan Data Science Initiative’s ongoing commitment to promoting diversity, we are highlighting the work of 5 exceptional women in the field of data science. The first profile in our series is on Katy Huff, a Berkeley Institute for Data Science, Moore/Sloan fellow.
Nuclear power is a highly controversial topic within the debate on clean energy. 20% of the United States total energy usage is derived from nuclear reactions, and in one year, a nuclear reactor can generate the equivalent energy of 13.7 million barrels of oil, or 3.4 million tons of coal. It is a highly efficient, and promising energy source, albeit with several disastrous incidents. In the past 50 years, several catastrophic instances have tarnished the reputation of nuclear power.
To tackle these pressing issues in the field of nuclear power, Kathryn Huff is using computer modeling and simulation to eliminate human error in nuclear reactions, and find safer ways to continue the practice of nuclear energy. Kathryn was drawn to the field of nuclear energy out of a concern for the environment, and she sees nuclear fission as an essential part of a sustainable energy future for the world. As a Berkeley Institute for Data Science, Moore/Sloan fellow, Kathryn’s work focuses on the intersection of nuclear engineering and data science. Specifically, she works on Python’s application in the field of nuclear science, and uses Python’s open-sourced nature to collaborate and share her work with other nuclear scientists.
One of Kathryn’s main projects, PyRK, is a python modeling package that maximizes the amount of energy harnessed from nuclear reactions, while ensuring safety. Python allows a high number of variables to be factored into an equation using a relatively small amount of code, and with nuclear reactor equations, the number of variables increases as each reaction progresses. The power created from neutrons during the initial nuclear fission go back and contribute more power to the original kinetic reaction, creating a feedback equation. PyRK models this equation, and monitors possibly disruptive variables in the process. PyRK also allows scientists to monitor the start ups and shut downs of nuclear reactors, the procedures prone to the highest number of accidents. Similarly, nuclear reactors are in constant need of coolant introduced to the system to prevent overheating, and PyRK allows scientists to accurately calibrate how much coolant is needed.
The same concepts in PyRK can be used for other feedback equations, in both the sciences, and other fields. The underlying theme of Kathryn’s work is to one day eliminate nuclear disasters through improved modeling solutions, and use Python to create a safer and more reliable energy solution for the world. There is a huge demand for a reliable clean energy sources that is both safe, and efficient, and PyRK is one step closer towards Nuclear energy becoming a leading clean energy solution.
By Jack Lowery