|
" Addressing Uncertainty in Energy System Optimization Models Over a Long Planning Horizon "
Patankar, Neha Satish
de Queiroz, Anderson
| Document Type
|
:
|
Latin Dissertation
|
| Language of Document
|
:
|
English
|
| Record Number
|
:
|
1052363
|
| Doc. No
|
:
|
TL51480
|
| Main Entry
|
:
|
Patankar, Neha Satish
|
| Title & Author
|
:
|
Addressing Uncertainty in Energy System Optimization Models Over a Long Planning Horizon\ Patankar, Neha Satishde Queiroz, Anderson
|
| College
|
:
|
North Carolina State University
|
| Date
|
:
|
2019
|
| Degree
|
:
|
Ph.D.
|
| student score
|
:
|
2019
|
| Note
|
:
|
213 p.
|
| Abstract
|
:
|
Rapid technological changes and anthropogenic climate change are motivating a fundamental transformation of the global energy system. That transformation comes with significant uncertainties. Dealing with the effects of future uncertainties in energy systems is a major challenge, particularly over long time horizons. This thesis utilizes Tools for Energy Model Optimization and Analysis (Temoa), an open source energy system optimization model to conduct the analysis. Temoa is used to examine a wide range of future uncertainties across three different applications of the model. The first application uses stochastic programming to address how conflict uncertainty in South Sudan can affect the country’s electricity planning. The second application utilizes robust optimization to address parametric uncertainty and explore deep decarbonization pathways in the United States. The third application focuses on modeling consumer behavior regarding the choice between electricity consumption and the adoption of energy efficiency measures. Chapter 2 describes the ongoing armed conflict uncertainty in South Sudan, the newest country in the world. Assuming that armed conflict might lead to damage to power system infrastructure with a finite probability, we evaluate the importance of uncertainty analysis for power system capacity expansion planning. The model solution obtained with stochastic programming is evaluated by estimating the expected value of perfect information (EVPI), the value of the stochastic solution (VSS), and the expected cost of ignoring conflict (ECIC). The study finds that utilizing distributed solar generation hedges against the risk of conflict. The investment in centralized hydro generation, on the other hand, will lead to higher financial risks despite the lower cost of electricity generation when conflict uncertainty is ignored. Chapter 3 sheds light on the limitations of using stochastic programming for energy system capacity expansion problems, which leads to the formulation of a robust optimization problem. The objective is to better characterize robust pathways for US deep decarbonization while considering future uncertainty in future fuel prices and technology capital costs. The variations in fuel price projections affect the penetration of renewables and the deployment of more efficient technologies. Moreover, the fuel prices and investment cost of the technologies are autocorrelated. The correlated robust optimization formulation is proposed based on the nature of data uncertainty. Another key challenge in energy system models is modeling the response of consumers to prices as well as the option to invest in energy efficiency in order to reduce electricity costs. Chapter 4 revises the Temoa formulation to describe the substitution effect between electricity and energy efficiency. The revised model formulated in chapter 4, referred to as the “energy efficiency model,” uses the production function defined in microeconomics theory to consider the substitution effect between electricity supply and the adoption of energy efficiency measures. The model is tested on a hypothetical test case, and sensitivity analysis is performed to quantify the effects of uncertain parameters on total (consumer plus producer) welfare. The sensitivity analysis suggests that lower energy efficiency costs lead to higher welfare recovered for a given efficiency subsidy. Similarly, lower own price elasticity of end-use energy service demand increases the welfare recovered at a given efficiency level. Lastly, the energy efficiency subsidy needs to increase as the carbon tax increases in order to recover the maximum amount of welfare
|
| Descriptor
|
:
|
Energy
|
|
|
:
|
Operations research
|
|
|
:
|
Systems science
|
| Added Entry
|
:
|
de Queiroz, Anderson
|
| Added Entry
|
:
|
North Carolina State University
|
| |