Assess: anomaly sensitive state estimation with streaming systems

Abstract

Information and communication technology (ICT) is an increasing part of modern power systems, which are, therefore, recognised as cyber-physical energy system (CPESs). The increase of ICT affects the situational awareness in CPESs, which is traditionally solely based on information about the power system but not about the ICT system. However, CPESs are facing various challenges regarding the integrity, correctness, and availability of process data due to the interconnection with ICT. Examples are stealthy false data injection attack (FDIAs). This paper pursues a holistic approach to describe the quality of process data, which brings together aspects like integrity, correctness, and availability in multivariate trust values. The arising research question this paper deals with is, how multivariate trust in physical measurements in a CPES can be modelled, estimated, and integrated into situational awareness. A proposed framework implements a context-sensitive and multivariate trust model as well as a trust sensitive state estimation. While these two artefacts are already published, the focus of this paper is on the implementation of the framework and the fulfilment of the requirements for timeliness, interoperability, flexibility, and scalability. It is evaluated in three different scenarios with CIGRE and IEEE benchmark grids.