Global average surface temperature is predicted to rise by 3.7-4.8° above pre-industrial levels by 2100 if we fail to mitigate our GHG (Green House Gas) emissions. However, even under best case scenarios, disruptions to our global climate are resulting in an increasing frequency of high-impact low-probability (HILP) events which can cause disruptions to our power systems. As a result, traditional reliability metrics alone are no longer sufficient to ensure secure operation of our networks, and resilience against these events is an important consideration. The increasingly distributed generation landscape facilitates local networks to operate as microgrids — where supply of the most essential loads can be sustained following contingencies.

In this thesis, a model for day-ahead-operation of a multi-energy microgrid, using rolling time horizons, is developed for the purpose of assessing resilience during disruptions. Due to its linear nature, the model is suitable for analysing both small and large-scale microgrids and non-exhaustively includes demand side response, renewable energy sources, network constraints — including a novel linear power loss implementation, and robustly considers the effects of uncertainty.

Microgrid resources such as vehicle-to-grid services (V2G), routed EVs, and thermal energy storage using inherent building mass and pipe heat networks, can help sustain the essential loads of an energy system during islanding, thereby increasing resilience. In this context, high-resolution, high-detail models are developed to assess how optimal operation of these resources can enhance the resilience in multi-energy microgrids (MEMG).

The results show the capacity of these resources to substantially enhance resilience during contingencies. However, the results also show that these resources can also diminish system resilience when operation or resources are non-optimal. Moreover, the smart operational strategies employed in this thesis can be used to maximise the resilience benefits of multi-energy microgrids, without considerable additional expenditure.