In this milestone project we will verify software that uses machine learning for real-time production planning and energy market bidding for renewable energy producers. To achieve the global environmental targets necessary for a sustainable future, today’s energy systems are transitioning rapidly. This require less fossil fuel based energy production and more renewable production, like solar and wind. Unfortunately, these sources lack the flexibility of fossil fuels, meaning that producers do not control when they produce. This causes a mismatch between supply and demand, which raises two fundamental questions. First, how do we deal with the surplus of renewable energy at times when there is too much supply? Second, how do we supply demand at times when there is too little production? Actors that can either store energy (e.g. in the form of batteries or pumped-hydro storage plants) or produce flexibly (i.e. hydropower producers), therefore have an increasingly important role in balancing the grid on short notice. Market mechanisms incentivise this behaviour, meaning that power producers can increase their profit margins while simultaneously benefiting the overall system by planning their production and bids smarter and more responsively. In the milestone project, we will verify software for real-time production planning and bidding for renewable energy producers, to enable them to achieve this goal.
Project leader: Kristin Jørstad
Institution: NTNU TECHNOLOGY TRANSFER AS