ESRI Research Seminar:"Quantifying uncertainty in energy system models"

Venue : ESRI, Whitaker Square, Sir John Rogerson's Quay, Dublin 2

Speaker : Dr Amy Wilson

Abstract:

Computer simulators of energy systems are widely used to help make evidence-based policy decisions in both industry and government. In order to make good decisions, any uncertainties in these computer simulators should be quantified and accounted for in the decision-making process. Such uncertainties may arise because: i) no computer simulator can perfectly replicate the complex real-world interactions which make up an energy system; ii) outputs from computer simulators are often heavily dependent on uncertain inputs such as future electricity demand; and iii) the computational demands of running a large energy system simulator may mean that the simulator has only been evaluated at a limited number of inputs.

Emulators, or statistical approximations of simulators, have been widely used in areas such as climate science for the quantification of uncertainty, but there have been few examples of this methodology in the field of energy systems. This talk will present an introduction to emulation for energy system models. The analysis of a specific exemplar, the Dynamic Dispatch Model (DDM), will be discussed. The DDM is a generation investment model, used by the UK Department for Energy and Climate Change and National Grid to study UK energy policy. An emulator for the DDM was constructed based on a limited number of publicly available model evaluations. The results of this analysis, and the impact that a thorough uncertainty analysis could have on UK energy policy will be presented.

Speaker Bio:

Amy Wilson is a post-doctoral research associate in the department of mathematical sciences at Durham University and a DEI fellow. She is a researcher on an EPSRC project which aims to improve methods for quantifying uncertainty in energy system models. Prior to joining Durham University, Amy completed a PhD in forensic statistics at the University of Edinburgh in collaboration with Mass Spec Analytical Ltd., a private forensic science provider.

Amy's research is in the area of statistical modelling for energy systems, with a focus on the modelling of uncertainty in large computer models. Particular applications she has worked on include the development of methods for capacity adequacy assessment in Great Britain and the study of the effect of uncertainties on the outputs of a generation investment computer model used by the Department for Energy and Climate Change.