Storage in WITCH completed
In the past weeks my research efforts were mostly dedicated to the improvement of WITCH, focusing in particular on storage. I remind that for this activity I can count on the collaboration of Marco and Simone, the two master students from the Polytechnic of Milan whom I have been mentoring for their dissertation.
As I anticipated in the relevant post in July, the new storage modeling substantially had two objectives.
The first objective was to fully integrate storage in the electricity system. This means describing the charge/discharge cycles, which can be fed either by the energy overproduced by Variable Renewable Energies (VRE) and that would be otherwise curtailed or by the energy generated by the other technologies and supplied to the grid. The description of curtailment of the energy overproduced by VREs (that is, in periods when supply is higher than demand) was introduced in the new system integration modeling implemented in the first part of the year, and it is sorted between short-term and seasonal storage.
This distinction is the introduction to the second objective, which was technology differentiation. Until the beginning of this work, in fact, storage was an undefined technology only contributing to the flexibility and capacity constraints. Concerning short-term storage, the model now considers Pumped Hydroelectric Storage (PHS), Compressed Air Energy Storage (CAES), and batteries. Concerning seasonal storage, we now consider that the curtailed electricity is firstly converted into hydrogen via alkaline electrolyzers, and then converted again in electricity via fuel cells (with the advantage that now it actively contributes to demand, obviously).
The next step regards the development of a new modeling scheme for the electricity grid. Grid is already modeled in WITCH, but in a simplified way. Plus, it is not linked to the rest of the system integration modeling. Thus the objective is both to add detail to the grid description and build an integrated modeling comprising system integration, storage, and grid.
This activity should not take up too much time, so in the very next weeks I will likely be able to write a more detailed post on the topic.