// Research needs in statistical modelling for energy system planning Chris Dent Amy Wilson / Meng Xu / Antony Lawson / Edward Williams Stan Zachary / Matthias Troffaes / Michael Goldstein ATI Energy Workshop 29 January 2016

Mathematics and energy systems  Variability and uncertainty (renewable generation)  Complexity (smartgrids, demand side participation)  Doing the same things cheaper and more robustly (asset management)

What I do  Maths/OR background, worked in Engineering since 2007  Mainly work in planning  All papers since 2010 in applied probability and statistics  Why planning – skills and opportunities

Risk of absolute supply shortages

∂ The GB Capacity Assessment Study ‘Will the lights stay on’, ‘Will the wind be there when we really need it’ ‒ Or ‘What is the risk of generating capacity shortfalls in a future season’ Originally report by Ofgem, technical modelling designed by NG ‒ Project team included me and Stan Zachary What did mathematical sciences bring ‒ Model specification, clarification of assumptions ‒ Probability theory work to understand drivers of model outputs ‒ Applied statistical work, particularly wrt sparse data

Current work: ‘traditional’ statistics  ‘Will the wind be blowing when we really need it?’ (or what’s prob dist)  Wind-demand link: temperature, time of day/week/year  Conditional independence  Report for Grid on non-sequential model (Wilson, Zachary)

Current work: UQ  Don’t always have traditional data  Expert judgment  Computer models  Embedded ops models  Limited # runs  Capital planning  Economic projection  EMR  Wilson/Lawson/Xu

∂ Sources of uncertainty (MG) Parametric uncertainty (each model requires a, typically high dimensional, parametric specification) Condition uncertainty (uncertainty as to boundary conditions, initial conditions, and forcing functions) Functional uncertainty (model evaluations take a long time, so the function is unknown almost everywhere) Stochastic uncertainty (either the model is stochastic, or it should be) Solution uncertainty (as the system equations can only be solved to some necessary level of approximation) Structural uncertainty (the model only approximates the physical system) Measurement uncertainty (as the model is calibrated against system data all of which is measured with error) Multi-model uncertainty (usually we have not one but many models related to the physical system) Decision uncertainty (to use the model to influence real world outcomes, we need to relate things in the world that we can influence to inputs to the simulator and through outputs to actual impacts. These links are uncertain.)

∂ Managing Uncertainty in Complex Models (

Need for statistics / data science  Lots of statistical problems  Planning, operation, futurology, policy  Until recently little involvement of statisticians  Or math sci in general  Except in short term forecasting  Need to integrate methodology and application communities  Research  Funding (RCUK and industry)