Modern distribution grid utilities are steadily adapting to the concepts of Smart Grids by augmenting distribution grids with Distribution Automation (DA) to enhance visibility and control for the purpose of enhanced system availability. Existing methods to place and evaluate DA overlook the important enhancement it provides to the resilience of a system. Resilience, a much-discussed but poorly defined measure for power systems, represents a system’s ability to withstand and recover from High-Impact Low-Probability (HILP) events such as storms and earthquakes. This thesis argues that exisiting resilience quantification methods do not capture the direct contribution which DA can make to enhance system resilience. It develops a novel model and methodology to analyse distribution grid resilience using the formalisms of Reliability Graphs (RGs), and Stochastic Reward Nets (SRNs). These two models capture the different parts of the complex recovery process which distribution grids perform to recover from faults using DA.

There are three novel contributions in this thesis. Firstly, a three-tier hierarchical model which contains an RG is developed to assess the enhancement which DA equipment provides to load point and feeder availability and resilience. Next, and SRN is used to develop a load point (LP) model which incorporates the dependence of feeder assets during the fault isolation phase of the recovery process. Finally, the SRN model is augmented with a phased recovery model to represent the complex recovery process for distribution grids. Utilising these models, the placement of switch automation and fault indicators is evaluated, and the contribution they make to resilience demonstrated. Collectively, these models give a novel means of assessing the the availability, sensitivity and resilience of distribution grids which utilise DA.