Incorporating endogenous demand dynamics into long-term capacity expansion power system models for Developing countries

摘要

This research develops a novel approach to long-term power system capacity expansion planning for developing countries by incorporating endogenous demand dynamics resulting from social processes of technology adoption. Conventional capacity expansion models assume exogenous demand growth; however, literature suggests that this assumption is not appropriate for developing countries. The planning approach presented in this research explicitly represents the links between the social and technical components of the power system. As potential customers without electricity select between various supply options to meet their power needs and as existing customers alter their consumption in reaction to the price of electricity and the perceived performance of the grid, the demand for grid power is directly impacted. This thesis demonstrates that neglecting these feedbacks and resorting to simplified assumptions can result in suboptimal investment strategies. By comparing the investment strategies identified using this novel approach to that of more conventional approaches, this research highlights cases in which the incorporation of endogenous demand impacts capacity expansion planning. More specifically, this work proves that incorporating endogenous electricity demand is important when there is a large fraction of the population without access to power or when the improvement in reliability afforded by capacity expansion is large. Employing traditional capacity expansion methods in such cases may lead to the selection of inferior expansion strategies. This research has both academic and applied contributions. Methodologically, this research extends state-of-the-art power system models by combining two generally separate modeling approaches, system dynamics and optimization. These methods are integrated to capture both the technical details of power grid operation and endogenous electricity demand dynamics in order to simulate the performance and evolution of the electric power grid. This research also demonstrates a holistic approach to centralized power planning that enables a more realistic representation of grid demand in developing countries and the identification of strategies that, in some cases, perform better than the strategies identified using traditional approaches. Finally, while this research was inspired by the case of Tanzania, the approach was developed with the flexibility to be applied to other countries with similar power system structure and contextual features.