The OPENTUNITY project aims to empower end-users by leveraging their flexibility assets within the ecosystem, supporting local system operators and creating new revenue streams. Aggregators play a key role by gathering flexibility from various sources and offering it to energy markets, optimizing asset activation when bids are accepted. This is where the University of Ljubljana (UL) contributes significantly.
UL’s main involvement in OPENTUNITY is in Task 4.4, focusing on developing a method for optimal flexibility selection from units of various sizes and types. The outcome will be an algorithm for optimal selection based on market participation (DSO or TSO) and minimal cost criteria.

We’ve adopted a bottom-up approach, splitting the task into:

  1. Baseline and flexibility forecast models for EV fleet (AVANT) and HEMS fleet (AMIBIT).
  2. Optimal selection algorithm.

Currently, we’re focused on developing forecast models for EV and HEMS fleets using Python, Keras, and Tensorflow. The analysis of charging point (CP) data for EV fleets shows significant flexibility potential during night hours by postponing charging sessions.

The difference between charging session length can be seen in the figure below.

charging session length

This flexibility potential is further illustrated in the figure below.

flexibility potential

In the coming months, UL will continue to delve into baseline and flexibility forecasting for EV and HEMS fleets, marking the initial step towards developing the optimal selection innovation and empowering end-users within the flexibility ecosystem.