Computational investigation of a model linking algorithm for the energy/economic interaction model

Abstract

Recently, there has been a great deal of energy planning models which combine the econometric submodels and the process analysis supply oriented submodels. These models were solved successfully by decomposition methods. Of these methods, the PIES (Project Independence Evaluation System) algorithm has received academic attention as well as real-world application. This along with increasing interests in modular modeling approach provided a motivation for this thesis. In this study, a model linking algorithm which is based on the concepts employed in the PIES algorithm is developed for a dynamic energy/economic interaction model. The convergence of the algorithm is tested empirically and the results show that the algorithm converges to the desired accuracy. But, more research efforts are needed on the validation of the convergence properties. The test model employed in this work can be applied for various sensitivity analyses of the energy policies such as the expansion of electric energy production capacity against various oil price scenarios, resource availabilities such as coal, gas, uranium, etc., Given the projection of GNP growth, the optimal energy production plan can be established, too. In addition, the algorithm in this study can be applied to generalized network problem provided that the relevant decomposition rules and proper visitation rules are developed.