Research On The Energy Systems Optimization Models Under Uncertainty

Energy is of importance to thedevelopment of human society; is the strategic material related to national livelihood of people, the national economic lifeline and national security; and provides a steady source for the progress of human society and is a direct impact on the country's modernization. However, in recent years, with the rapid social and economic development, energy consumption has been rising increasingly, leading to a serious contradiction between energy supply and energy demand. In addition, the energy structure based on coal, oil and natural gas has led to deterioration of the ecological environment. Therefor, integrated management for energy system has become an urgent issue for building a resources-saving society with low energy consumption, little environmental pollution, and high economic benifit. To begin, this research investigates the international issues regarding energy supplies and consumptions, as well as their the relationships with the environment; then narrates review of energy model and energy optimal techniques. Subsequently, an interval chance-constrained semi-infinite programming (ICCSILP) model is developed to support integrated enengy environmental systems planning under parameter uncertainty. Two scenarios are used to compare the optimized eletricity supply strategeis:one scenario assumes the environmental constriants can be neglected, and the other one stressed the necessity of the environmental constraints. Results obtained through solving the model under the two scenarios provide decision support for energy allocation and environmental control. Furthermore, an interval fuzzy chance-constrained semi-infinite programming (IFCCSILP) model is developed to simultaneously deal with fuzzy and interval parameters in an integrated enengy and environmental system. By solving the model, not only decisionmaking strategis can be obtained, but the satisfactory level of the each strategy can be known. Thus, IFCCSILP provides a more flexible decision-making tool compared to than ICCSILP. Finally, the limitations of the two models and future studies to overcome such limitations are dicussed.