The electrical power grid is a backbone of the US infrastructure: Every year, about $400 billion of electricity is being transmitted across the country. The operation underlying the electrical infrastructure has been, and remains, a success story for optimization technology which solves unit commitment problems daily and computes power flows every 5 minutes or so. The grid would not operate without decades of progress in linear programming, mixed integer programming, and nonlinear optimization.
The electrical grid however will face tremendous challenges in the next decades, as it transitions from fossil fuels to renewable energy, power electronics, storage technology, and demand response. This paradigm shift requires a reconsideration of the fundamental assumptions underlying the grid design. The future grid will need to cope with the inherent unpredictability introduced by solar and wind energy “in front and behind the meter”, the loss of inertia induced by power electronics, the electrification of transportation systems, and the increasing frequency of extreme weather events that will necessitate new approaches to risk and resilience. It is largely recognized at this point that “the expected growth in system complexity will require the development of substantially improved software optimization and control tools to assist grid operators, and deliver the societal benefits of improved grid performance”.
This proposal is a catalyst to develop critical mass and expertise on campus in the area of computational energy and to foster collaborations with leading universities, national laboratories, and industry. It is rooted in the recognition that the future grid requires a deep integration of high-performance computing, forecasting methods, optimization technology, power engineering, and control algorithms.