860
What Can Electrochemistry Learn from Chess? Using Data Science to Speed up Optimization of Electrochemical Models

Monday, 1 October 2018: 08:40
Universal 3 (Expo Center)
N. Dawson-Elli (University of Washington, Seattle), S. Kolluri (University of Washington), and V. R. Subramanian (University of Washington, Seattle)
Batteries are complex electrochemical devices whose short-term and long-term operational lives are highly dependent on their construction, materials, and use cases. Models have been developed and used to improve the performance of batteries by providing different material design, model based control and model based battery management systems1,2.

In recent literature, data science techniques have been used to accomplish a variety of tasks related to representing simulated data for control or estimation purposes2,3,4. Data science techniques have also been used to approximate optimal solutions to computationally difficult problems, such as the traveling salesman problem5. Rather than create a surrogate model to use with a typical optimization scheme, we will examine framing the problem as a parameter ‘correction’ endeavor. By doing so, it may be possible to retain all of the information present in a time series of error and leverage this information to very quickly go from a poor initial guess to a viable solution, after which time iterative optimization schemes can take over. The effectiveness of this approach (both in terms of efficiency and robustness) will be compared with the standard approaches. This method is inspired by DeepChess, a comparative deep neural network structure which plays successfully against world-class chess algorithms6.

Acknowledgments

The authors thank the United States Department of Energy (DOE) for the financial support for this work though the Advanced Research Projects Agency – Energy (ARPA-E) award #DEAR0000275, as well as the Clean Energy Institute (CEI) through the DIRECT program at the University of Washington. Part of the research has been supported by the Assistant Secretary for Energy Efficiency and Renewable Energy, Office of Vehicle Technologies of the U. S. Department of Energy through the Advanced Battery Materials Research (BMR) Program (Battery500 Consortium).

References

  1. Y. Dai, and V. Srinivasan,. J. Electrochem. Soc., 163, A406 (2016).
  2. N. Dawson-Elli, S. B. Lee, M. Pathak, K. Mitra, and V. R. Subramanian, “Data Science Approaches for Electrochemical Engineers: An Introduction through Surrogate Model Development for Lithium-Ion Batteries,” J. Electrochem. Soc., vol. 165, no. 2, pp. A1–A15, Jan. 2018.
  3. D. A. C. Beck, J. M. Carothers, V. R. Subramanian, and J. Pfaendtner, “Data science: Accelerating innovation and discovery in chemical engineering,” AIChE J., vol. 62, no. 5, pp. 1402–1416, May 2016.
  4. Blake R. Hough, author. Jim Pfaendtner, degree supervisor. 2016 Thesis (Ph. D.)--University of Washington,2016.
  5. B. La Maire and V. Mladenov, “Comparison of Neural Networks for Solving the Travelling Salesman Problem,” in 11th Symposium on Neural Network Applications in Electrical Engineering,NEUREL 2012 - Proceedings, 2012.
  6. E. David, N. S. Netanyahu, and L. Wolf, “DeepChess: End-to-End Deep Neural Network for Automatic Learning in Chess,” arXiv:1711.09667 [cs, stat], vol. 9887, pp. 88–96, 2016.