Development of GHG-Mitigation Oriented Models for the Planning of Integrated Energy-Environment Systems (IEES) Under Uncertainties

Abstract

Energy-related activities are a major contributor of anthropogenic greenhouse gas (GHG) emissions. The mitigation of GHG emissions should thus be incorporated within the framework of integrated energy-environment systems (IEES) planning. However, various uncertainties and dynamics of the systems pose difficulties for IEES planning. As well, variations associated with random weather/climatic conditions aggravate planning challenges. This dissertation aims to address such complexities through developing a series of inexact optimization models. The associated models include: (a) an interval multi-stage stochastic programming model (IMSP-IEES), (b) an interval fuzzy multi-stage stochastic programming model (IFMP-IEES), (c) a dual-interval mixed-integer linear programming model (DMLP-IEES), (d) a dual-interval multi-stage stochastic programming model (DMSP-IEES), (e) a fuzzy dual-interval multi-stage stochastic programming model (FDMSP-IEES), (f) a joint-probabilistic left-hand-side chance-constrained programming model (ILJCP-IEES), (g) an interval joint-probabilistic two-side chance-constrained programming model (IJTCP-IEES), and (h) an interval fuzzy two-side chance-constrained programming model (IFTCP-IEES) for energy systems planning and GHG-emission mitigation in Alberta. The main contribution of this research is the development of a series of innovative approaches for supporting robust planning of regional energy systems and scientific reduction of GHG emissions under uncertainties. The IMSP-IEES and IFMP-IEES improved upon the previous inexact optimization methods by allowing the stochastic uncertainties and dynamics within a multi-layer scenario tree to be incorporated within the planning systems. The integration of fuzzy programming further enhanced the robustness of IEES optimization through addressing the ambiguous system information. The DMLP-IEES, DMSP-IEES and FDMSP-IEES introduced the concept of dual interval into IEES planning to address dual uncertainties without distribution information but rough estimation of lower and upper bounds. The proposed methods have advantages in integrating inherent system uncertainties expressed not only as discrete intervals and dual intervals but also as possibility and probability distributions into solution procedures. The ILJCP-IEES, IJTCP-IEES and IFTCP-IEES enhanced the capabilities of previous inexact stochastic optimization approaches in dealing with left/two-side multi-randomness issues, especially when a joint-probabilistic requirement is imposed on multiple constraints. Through integrating interval and fuzzy programming, the three universal types of inexact information could be addressed and thus further enhance the robustness of the IFTCP-IEES. The IFTCP-IEES was then applied to GHG-emission reduction and energy system planning in Alberta. Results indicated that interactions among various system components would be sufficiently reflected in a mixed multi-uncertain environment.