Impacts of distribution-connected variable generation on short-term load forecasting and operational planning

Summary form only given. A growing amount of variable generation is being connected to the distribution network, where it is not directly monitored or controlled by the system operator. This introduces a greater degree of uncertainty into bulk system operations. For short-term operational planning tasks, such as the prediction of network congestions, it is becoming increasingly necessary to forecast demand/generation profiles accurately on a locational basis, for example, at the bulk supply point level. This work examines the state-of-theart in methods for short-term forecasting of disaggregated, bulk supply point demands, including approaches for modelling variable generation embedded in the distribution networks. It looks at how wind and solar forecasts with various spatial resolutions can be integrated with demand profile forecasts to give the best results. Additionally, it discusses what ranges of accuracy can realistically be achieved for forecasting bulk supply point demand/generation profiles, and how this uncertainty can impact bulk system short-term operational planning and congestion management.