Modeling future cyber-physical energy systems

Abstract

In this paper a model of a future combined cyber-physical energy system is introduced. We view such systems as the intertwined physical-cyber network interconnections of many non-uniform components, such as diverse energy sources and different classes of energy users, equipped with their own local cyber. This modeling approach is qualitatively different from the currently used models that do not explicitly account for the effects of sensing and communications. The proposed approach is based, instead, on representing all physical components as modules interconnected by means of an electric network. However, not all physical components can be modeled from first principles because of the extreme non-uniformity and the complexity of various classes of components. Instead, many components and/or groups of components have to be monitored and their models have to be identified using extensive signal processing, sensing, and model identification. We illustrate such combined cyber-physical models of key components, and use these to introduce a structure preserving model of a cyber-physical infrastructure of the interconnected system. Such a model becomes a basis for deciding what to sense and at which rate, what level of data mining is needed for which (groups of) physical modules to achieve predictable performance for cyber-physical future energy systems. This model rests on the premise that the performance of future energy systems can be shaped in major ways by means of broadly available cyber technologies. In order to make the most out of the available cyber technologies, the first step is to establish models which capture these interdependencies. This paper is a step in such direction.