On energy modelling for a range of spatial, temporal and technological scales

Abstract

In this dissertation a number of modeling studies, ranging from process specific mathematical formulations to global long term climate consequences of energy and emission scenarios, are presented and the commonalities, and differences, across the existing energy modeling methodologies and applications are reviewed. For the regional modeling work mathematical representations are developed for the operation of small biofuel fired CHP plants. These mathematical descriptions are then used in a non-linear optimization model, in order to evaluate the economic feasibility of connecting such a CHP plant to a local district heating grid. The results indicate that under the assumed policy and economic conditions, such a CHP plant would not be able to compete with a biofuel fired heat-only boiler. The global, long term part of this work describes the modification and application of a global energy system model for a range of climate related issues. An endogenous description of "learning by doing" is implemented in the model and the subsequent model results show that although technological progress alone is unlikely to lead to climate stabilization, and therefore specific policies aimed at emission mitigation are a necessity, the lowered costs resulting from the "learning by doing" effect can reduce the mitigation costs considerably. The further studies establish that if the climate policy regime is incomplete, in the sense that some regions join it considerably later than others, more stringent targets might be difficult to reach and they will certainly be more expensive. Finally, a probabilistic study shows that reaching ambitious temperature targets with a high likelihood might not only require a wide portfolio of mitigation options and relatively early action, but it may also be that scenario specific indicators, such as demographic, economic and technological developments would need to progress in favorable manner. In the summary section the field of energy modeling is reviewed and presented in terms of methodologies and applications on the one hand and in terms of the system borders of the models on the other. The studies presented in this thesis and in the literature are placed within this framework and the commonalities of models appearing very different on the first look are pointed out.