Energy and data science
The following sections illustrate some aspects of the interplay of energy and data science (see also here).the
Forecasting is important in many different contexts (for an introduction see here .) Price or revenue forecasts are important across all industries. For energy, fossil fuel prices are particularly relevant, since the currently form the majority of the consumption. For renewables both supply and demand forecasts are very important. Particularly, supply shows considerable variability, e.g., in the case of wind. Hence forecasts are very important and data science can be instrumental to achieve that.
The following two links give some assessment of the live UK energy generation and consumption gridwatch, ukenergywatch. A nice overview for the German energy generation is compiled by the Fraunhofer institute and can be found here. With more volatile sources connected to the grid supply-demand matching will become a more difficult problem. Some companies specialise in helping to regulate supply-demand matching by adjusting capacities (enernoc, redtree). PV supply is estimated here.
The UK plans to roll out electricity smart meter to 26 million households by 2020 ( see here). Even though this is unlikely to happen by this data smart meters will eventually become an integral part of every household in the future. We expect that data from real time electricity meters and smart home applications which determine the time electricity and other energy consumption takes place in the household will become more important. In this context data science will be crucial to provide relevant insights as well as directives in order to optimally coordinate consumption.
Some data for energy trading in the UK can be found here .