A Holistic Approach to Dependability Modeling and Analysis of Smart Distribution Grids

Abstract

Lately, the distribution grid has been under a significant transformation, with a pervasive integration of Information and Communication Technology (ICT) for an enhanced operation and management of the grid. Some of the driving forces behind these changes are the spread of new and cheap technologies for generation and storage of electrical energy, along with the policies to reduce greenhouse emission, and the incentives for an increasing contribution of renewable energy into the global energy industry. These massive changes, especially the large scale use of distributed energy resources, have not only brought benefits, but also some new challenges. In order to deal with the new challenges, an extensive use of ICT has been introduced. In fact, the smart grid has now become a complex interdependent system of system in which the dependability of ICT has a significant impact on the overall dependability. Considering the grid’s role in the society, it is very important to carefully investigate the interdependency between the power system and the newly introduced ICT support systems. The thesis aims to develop a modeling framework that can be used to study and analyse the dependability of future distribution grid where ICT plays a major role. In doing this, a literature survey is first conducted to identify the open challenges in modelling the future distribution grid, as well as the new behaviours that may arise due to the extensive use of ICT in the future grid. A modelling framework is then developed and the impact of the identified new behaviours and challenges on the dependability of the distribution grid is assessed using the framework. The framework is based on Stochastic Activity Networks (SAN) using the Mobius tool. The core part of the modelling framework mainly uses the SAN based simulation to model the most relevant properties, often described as discrete activities, such as failure processes, repair processes as well as the interactions and dependencies between the ICT support system and the underlying power system. The method employed for the SAN modelling is a bottom up approach. First, individual models are created for all the components. Then, these models are combined to represent a subsystem or the whole system in a second tier model where interaction and dependencies between the component models are defined. The framework is also extended with an integration of external C++ based simulator that are used to model some activities that can only be expressed in the continuous time domain. In the process of the research work, the framework has been customised and extended for different use cases which are defined to study the impact of the identified future challenges on the dependability of the grid. The investigations on these challenges show that the growing dependency on ICT has a significant influence on the dependability of the grid. The analysis shows that investigation of the ICT’s impact should not be limited to the usual conventional (omission) failure mode assumptions. Investigation on the effect of assuming different failure modes of ICT showed how a small variations on these assumptions have a significant influence on the dependability of the grid. Meanwhile, studies conducted on the effect of introducing new ICT technologies and architectures revealed that the IEC 61850 for substation automation and the use of 5G for monitoring and protection systems could improve the dependability measures significantly.