This paper presents two different approaches to model expansion planning in electricity markets under imperfect competitive conditions. Both approaches ocnsider a market in which firms compete in quantity as in the Nash-Cournot game. However, the two models differ from each other in how earch firm makes its expansion-planning decisions. In the first model, the key assumption that permits its formulation as a Mixed Linead Complementarity Problem (LCP) is that firms not only decide their output in a Cournot manner but also their new generating capacity. In contrast, in the second model, there is a leader firm that anticipates the reaction of the follower firms as in the Stackelberg game. The latter model is formulated as a Mathematical Program with Equilibriun Constraints (MPE), wherein the leader firm decides its optimal new capacity subject to a set of market equilibrium constraints. Both models have been implemented in GAMS. A simple example is also presented to illustrate their comparison and application.
Keywords: Complementarity problem, nash-cournot, mathematical programming with equilibrium constraints, stackelberg
14th Power systems computation conference -PSCC, Session 43-4, pp. 1-8, Seville, 24-28 June 2002
Publicado: junio 2002.