Develop new system-theoretic data analytics (AI & ML) methods to enhance the coordination of local energy systems (LESs) and green buildings
In order to meet the 2030 emissions reduction targets of the Climate Agreement, the Dutch and Chinese government requires the active participation of all energy systems users. Only by properly coordinating energy generation and consumption among all users will a safe energy provision with minimum cost be obtained. In the DATALESS project, different data sources combined with advanced AI and engineering methods will be employed to achieve this required level of coordination. New understanding methods will also be developed to increase residential users' acceptance of renewable technology.
The main objective of the DATALESs project is to develop new system-theoretic data analytics (AI & ML) methods to enhance the coordination of local energy systems (LESs) and green buildings. The research outcomes will boost the Netherlands and China's globally leading position on the digitalization of the energy system and the realization of the Climate Agreement GHG emissions reduction targets of 2030.
From a technical perspective, we will accomplish this by developing new models that LESs operators can directly use to enhance operation by properly coordinating all the distributed energy resources (DERs) available to reduce renewable energy curtailment, reduce operational costs, etc. These results will strengthen aggregators' and energy traders' business cases by providing them with state-of-the-art analytical tools to estimate energy flexibility availability. We will also develop a smart control framework to enhance LESs reliability based on edge computing and 5G communication. These technologies can provide a guarantee for efficient data processing and transmission. The project's success will also result in an information model and architecture being accepted and adopted by society and stakeholders. In practice, the digital twin obtains the energy performance of various buildings (interventions) and detects relevant patterns through sensor networks and machine learning. Enhanced actors' integration through the exchange of required information will lead to more comprehensive and effective implementation of green building and positive energy districts (PEDs).
From a societal perspective, understanding the interaction between traders (peer-to-peer) in the market can help determine how new network users can be enticed. Still, it could also help to understand the impact of the individual decision-making process. The latter is especially important since decisions that can cause energy wastages are Pareto efficient, meaning that actions from users to improve comfort usually lead to larger wastages.
The DATALESS consortium involves four academic institutions (TU Delft and The Hague University of Applied Sciences (HHS) in the Netherlands, Tsinghua University and Zhejiang University in China) and three industry partners (Alliander and Geodan in the Netherlands, PEPITe in Belgium, Jibei Power Exchange Center in China).