Optimization for Power Systems and the Smart Grid

Practitioners in the field of power systems operations are keen users of optimization techniques. Several fundamental problems in the area are solved every day using optimization algorithms as part of the real-time operation of the power grid. One such fundamental problem is the unit commitment problem that is concerned with scheduling power generation so as to meet demand at minimum cost. Realistic instances of unit commitment are typically large-scale, and because the time available in the real-time context is limited, practitioners sometimes have to settle for solutions that are not globally optimal. Beyond the well-known fundamental problems, the advent of the smart grid introduces new challenges for power system researchers and for optimizers. The smart grid is the combination of a traditional electrical power distribution system with two-way communication between suppliers and consumers. This combination is expected to deliver energy savings, cost reductions, and increased reliability and security. However it also raises new difficulties for managing of the resulting system. These include integrating renewable energy sources such as wind and solar power generation, managing bidirectional flows of power and of information, and incorporating demand-response. This chapter begins with an overview of the area of smart grid and some of the challenges relevant to optimization researchers. We then summarize two recent examples of optimization research in power systems, the first consisting of an application to demand-response for the smart grid, and the second of a new technique to solve certain types of unit commitment more efficiently.