Forecasting and Probabilistic Methods for Power Systems

Probabilistic methods have attracted renewed attention in power systems engineering in recent years. Most prominently, they provide the natural framework in which to analyse the variable output from renewable sources and their integration into the system. Also, because probabilistic methods can often reflect underlying uncertainties better than deterministic approaches which have been used in the past, they can also support more efficient planning and operation of systems.

This paper reviews UK research capabilities in the area of probabilistic modelling applied to power systems, based on a broad survey of recent research outputs. The field is divided into four broad categories of planning timescale generation adequacy assessment, network planning, forecasting and short term operation, and network reliability analysis.

The survey is based on the author’s own knowledge and experience, combined with advice from a number of relevant researchers as to which of their own and others’ research outputs best represent current capabilities. While there is discussion of the context of current industry practices, the emphasis throughout is on original research originating in the UK.