In May the workshop paper Physics-Informed recurrent neural networks for the identification of a generic energy buffer system (view pdf) from Manu Lahariya (imec) was published.

This paper proposes 2 physics-informed neural networks-based grey-box modeling methods for the identification of energy flexibility. The same physics based approach is used within BIGG to characterize the flexibility of residential gas boilers and offer flexibility services to gas suppliers, while respecting the comfort limits of the users.