A generic framework for the valorisation of high-dimensional virtual power plants integrating multiple flexibility technologies
In response to the requirement of a sustainable energy system, diversified energy resources, enhanced energy efficiency and liberalization of electricity markets, the energy sector is experiencing worldwide a huge penetration of Distributed Energy Resources (DERs). This entails significant challenges mostly related to the intermittent nature of Renewable Energy Sources (RESs) and the small size of DERs, which prevents their individual participation in the current electricity market and their active participation in the grid management. On the other hand, the central role that DERs will play in the future of the energy sector calls for immediate solutions to support their integration.
The full valorisation of DERs can be achieved through their virtual aggregation in a large portfolio, which is then managed as a single system to participate in multiple electricity markets and provide ancillary services to the grid. This is commonly known as the concept of Virtual Power Plant (VPP) and represents an appealing solution to overcome the difficulties related to the individual participation of DERs, as well as promote the diffusion of RESs.
The efficient exploitation of such a large-scale portfolio of DERs is far from being an easy task. A complete decision-making framework for the energy management of such a high-dimensional system should include a chain of interconnected models involving forecasting of system’s uncertainties, advanced optimization and real-time control. In addition to this, the decision framework should be flexible enough to accommodate the evolution in time of the portfolio. Moreover, a new trend is arising in the business model of VPPs, providing for a diversification in the aggregator’s assets through the integration of power-to-X (P2X) reconversion technologies (e.g. power-to-hydrogen conversion) and other sources of flexibility such as electrical vehicles and demand response programs.
Luca’s study aims to answer at least part of these challenges. First in term of modelling, providing a flexible formulation which is able to describe the dynamic nature of the VPP portfolio (e.g. integration of new assets, unavailability of resources due to maintenance or faults). Then in the decision-making framework, providing a forecasting model that ensures spatial-temporal coherence between the individual forecasting of each unit, as well as optimization and real-time control strategies that can meet the industrial needs of scalability, privacy and resilience. To this aim, this project involves the real-world data, the complex models and the experience of the French aggregator Compagnie Nationale du Rhône (CNR).
Figure 1: A Virtual Power Plant integrating several types of DERs to participate in multiple electricity markets and provide ancillary services to the grid.
More about Luca Santosuosso
On his background
“I started my undergraduate at Roma Tre University in Rome (Italy) with a bachelor in computer engineering, then moved to Sapienza University of Rome in Rome (Italy) for my master in control engineering.”
On his research interests
Sequential decision-making, power system optimization, control and automation.
On choosing Smart4RES and Mines Paris for his research
“I was looking for a stimulating project for my master thesis when I first received the proposal to join the Smart4RES team for an internship. Since the first meeting I had with Simon Camal and Georges Kariniotakis to discuss about my contribution to the project, I knew it was going to be an incredible opportunity to grow both professionally and personally in a stimulating environment. The decision to join the project was straightforward.”
Where he sees himself in 5 years
“Either in the academia or in the industry, but always contributing as much as I can to face the research challenges of the energy sector.”