EFFICIENT RE-FORMULATION OF LINEAR COGENERATION PLANNING MODELS

摘要

Recent development in the energy sector has made it necessary to create more realistic production planning and trade optimization models and to solve them more rapidly. Different kinds of cogeneration planning models can be formulated as linear programming (LP) or mixed integer programming (MIP) models, for which general solution techniques exist. We present a technique for reformulating long-term cogeneration planning models so that they can be solved much more efficiently using specialized algorithms. The technique is based on decomposing the long-term model into hourly models for individual production plants, identifying automatically the extreme points of the plant models, and re-formulating the model in terms of the extreme points as an LP/MIP model with a specialized structure that can be exploited for solving the problem much more rapidly.