A modelling framework for characterising the impacts of uncertainty on energy systems

A modelling framework was developed for characterising the impacts of variability and uncertainty of renewable generation and load on a multi-vector and multi-scale energy system. A time series synthesis algorithm was proposed to produce a large number of daily profiles for wind, PV and load, representing possible variation features. Based on the synthetic time series, four models with different uncertainty characteristics were built and applied to generate forecast scenarios for wind and PV. Using the generated scenarios, the operation of a combined gas and electricity system was formulated as a two-stage stochastic mixed-integer linear optimisation problem. A simplified Great Britain's energy system was investigated under different flexibility options and uncertainty characterisations. Results indicate that electricity storage was the most effective measure to reduce operation cost. The impacts of uncertainty characterisation methods were significant only when system flexibility was insufficient