Research on Global Uncertainty and Sensitivity Analysis(SA) Algorithms Based on the Development of Energy Internet

Abstract

Conversion and interconnection of multiple energies have become the main form of energy development in China. This paper analyzes the evolution of energy internet and proposes its basic framework based on summarized typical features. As we all know, the mathematical model in energy internet plays increasingly key role in arranging and optimizing resources as well as predicting work. In real applications, we find many complicated problems of electrical model, such as high-dimensional, multiple peaks, non-linear, inconsistent, and nonconvex. In order to solve them, this paper combines model simulation with Global Uncertainty and Sensitivity Analysis to analyse and predict the scientific problem from a probabilistic viewpoint. Global Sensitivity Analysis aims to find out how changes of multiple parameters affecting working result of electrical model, and to analyze how interactions among parameters influencing model results. While modeling each type of electrical model, Global Sensitivity Analysis are used less, this paper also lists various sensitivity analysis algorithm, and describes kinds of methods of both local and global sensitivity analysis. The research will provide a complete method data base of sensitivity analysis for each model development in electrical system. Furthermore, the combination of uncertainty and sensitivity analysis at home and abroad is beneficial to figure out the difficulties and focuses of connections among researches on sensitivity analysis and its common features and the spatially explicit landscape modeling.