Evaluating flexibility and adequacy in future EU power systems: model coupling and long-term forecasting

There is an increasing need for improving the representation of the power system in energy system models. Recent research has shown that the economic value of each MWh of Variable Renewable Energy (VRE) declines with growing deployment. There are two main reasons for this: (1) even with limited shares of VRE, there is a reduced utilisation of dispatchable power plants and (2) with high shares of VRE, there can be
possible excess electricity that might have to be curtailed in case of insufficient storage capacity. These findings tend to be completely ignored by models with low temporal resolution and high shares of VRE. In this work, we aim at endogenizing the flexibility constraints into long-term planning models (such as the JRC-EU-TIMES model) by taking profit of Dispa-SET, a unit commitment and optimal dispatch model. The main idea is to derive generic functions from Dispa-SET that can be used by JRC-EU-TIMES and all other energy system models. In particular, we evaluate the impact of increased variable renewable power and different configurations of power systems (production, storage, EVs, demand and flexible demand) (1) on the operations of power plants and the different storage options, determining their economics such as operating hours, number of cycles and (2) on the amount of possible excess power (power that has to be curtailed, stored, or transformed. Simulations are run in Dispa-SET for a wide set of exploratory configurations of the power system in Dispa-SET. Then, a statistical analysis of the results is performed to derive generic flexibility functions. An example of the use of such flexiblity function is finally proposed for a long-term simulation in the JRC-EU-TIMES model at the European level, showing a non-negligible impact on the value of VRE, on the utilization of dispatchable generation, and on the required storage capacity.