Abstract
This chapter discusses strategies to coordinate charging of autonomous plug-in electric vehicles (PEVs). The chapter briefly reviews the state of the art with respect to grid level analyses of PEV charging, and frames PEV coordination in terms of whether they are centralized or decentralized and whether they are optimal or near-optimal in some sense. The bulk of the chapter is devoted to presenting centralized and decentralized cost-optimizing frameworks for identifying and coordinating PEV charging. We use a centralized framework to show that "valley filling" charge patterns are globally optimal. Decentralized electricity cost minimizing frameworks for PEV charging can be framed in the context of noncooperative dynamic game theory and are related to recent work on mean field and potential games. Interestingly, in this context it can be difficult to achieve a Nash equilibrium (NE) if electricity price is the sole objective. The decentralized algorithm discussed in this chapter introduces a very small penalty term that damps unwanted negotiating dynamics. With this term, the decentralized algorithm takes on the form of a contraction mapping and, in the infinite system limit, the NE is unique and the algorithm will converge to it under relatively loose assumptions.
Original language | English |
---|---|
Title of host publication | Control and Optimization Methods for Electric Smart Grids |
Publisher | Springer New York |
Pages | 259-273 |
Number of pages | 15 |
ISBN (Electronic) | 9781461416050 |
ISBN (Print) | 9781461416043 |
DOIs | |
Publication status | Published - 1 Jan 2012 |