The report covers energy management algorithms (EMS) for virtual aggregations consisting of distributed generation (mainly from renewable sources), storage and loads, possibly grouped locally in microgrids.

The share of generation from non-programmable renewable energy sources is constantly increasing, so there is a need to expand the energy resources for the various frequency and voltage regulation services, such as secondary and/or tertiary reserve. To this end, the Italian Regulatory Authority for Energy, Networks and the Environment (ARERA) has opened the Market for Dispatching Services (MSD) to new types of users, including on the distribution grid, through the figure of the aggregator. The management of aggregate units, called Virtually Aggregated Mixed Units (UVAMs), poses several challenges due to the number and heterogeneity of users, the overlap of system services with local services for the user, the greater uncertainty related to measurements and forecasts, and the complexity of the ICT architecture.

This system research activity will be developed over three years, and in the current first year, an optimization/control algorithm (MAGC) has been developed that, starting from the day-ahead production program of microgrids and other resources that are part of the aggregate, reconstitutes in real time the secondary and/or tertiary reserve sold to MSD in the face of imbalances due to generation and load forecast errors. The designed control approaches, tested with simulations on real grid benchmarks, have demonstrated their effectiveness.

ARERA has also introduced the role of the Balance Service Provider (BSP), which has the task of aggregating the different resources and connecting them to the MSD as a single equivalent unit. This report describes a techno-economic analysis procedure to evaluate the flexibility limits on a case study of an aggregation of residential users with photovoltaic systems and storage systems used mainly for self-consumption. The activities carried out during 2019 in order to launch two pilot projects with two aggregates of this type are also reported.

Finally, an activity has been launched to develop an EMS based on a stochastic optimization algorithm: in fact, a relevant factor for the connection of new RES plants is their intrinsic intermittence, which could lead to a deterioration in the quality of the electricity service or an excessive increase in costs for the user. A stochastic approach to the definition of the production program of the microgrid or the aggregation of grid resources can minimize the cost impact of the inevitable forecasting errors of RES generation and load. The report starts with an examination of possible models for stochastic optimization.