SmartNet interviews: a talk with the project leader, G. Migliavacca

  • SmartNet was born from the need to propose practical and realistic solutions to the integration of Renewables in the electricity transmission network. What were the main difficulties that you encountered – or are still encountering – during your endeavour?

The SmartNet project is built around the idea to compare five different coordination schemes, representing different ways to conceive the interaction between TSO and DSO for the activation of flexible resources connected to distribution grids in an ancillary services market context. The basic assumption is that the steadily increasing penetration of renewable sources of generation increases the need for system reserve to be used for system balancing, congestion management and voltage control. Such reserve was traditionally provided by big power plants connected to transmission networks. For this reason, nowadays the ancillary services markets are usually managed by the national TSO. However, in these years we are assisting on one side to a progressive dismissal of big power plants, which, based on coal and gas technologies, result more and more out of merit order and on the other to a significant increase of distributed generation and flexible loads. So, the challenge of SmartNet is to test whether local flexible resources may be involved in the provision of ancillary services to the system and, consequently, different schemes could prove more efficient, in which the role of the DSO increases and even surpasses the one of sheer local congestion management manager (as foreseen by the Winter Package). This will be investigated by developing an ad hoc simulation platform able to model both physical network and real-time markets while considering the aggregator as the key figure staying between them, and ICT as the enabling technology (as well as a source of costs for the implementation of each coordination scheme).

Clearly, the simulation environment must be large enough to allow to simulate transmission network, distribution networks, real-time markets, aggregation and disaggregation processes, as well as the bidding processes of the ultimate flexibility providers, hence an important scale problem. This constitutes an important hurdle from the numeric point of view, but we are close to solve it with state-of-the-art software and modelling. Moreover, we encountered enormous problems in collecting the huge amount of data which allow to characterize correctly the simulated scenarios (RES generation and imbalanced profiles, flexibility characterization, distribution and transmission network data, etc). We are about to solve also these problems and promise to present the first results of the simulations along with the associated cost-benefit analysis assessment by end of January 2018, when we are going to hold our next Advisory Board meeting, thus informing the main European stakeholders and asking for their validation of our results. In any case, we are posting all significant documents and on-going consultations on our official website ( This should foster enlarging the debate to a wider public, allowing them to provide their opinions on our activities.

  • SmartNet is also aiming at facilitating the coordination between the grid operators at a national and local level. Given the differences in infrastructure between not only the various member countries of the EU, but also within the various municipalities of the countries themselves, how optimistic are you that this will prevail? And what kind of method (solutions) do you adopt?

The present regulatory situation is very complicated, especially on distribution side. There are huge distribution companies, able to afford direct R&D participation and small municipalities which have even not a technical structure which could allow them to take part into a future management of ancillary services. Even more, there are (at least in Italy) different kinds of distribution companies. On a given territory, typically one “level 1” distributor is directly connected to the transmission network. However, this doesn’t prevent to have smaller distributors that are only connected to the one at level 1, and even others at more underlying levels (up to three, in my knowledge). The smallest distributors (mainly still existing for historical reasons) could just count a dozen of nodes or so. Taking stock of this manifold reality, I think it couldn’t be realistic to assign every DSO critical operational tasks, but, nonetheless, it could still be thinkable that all the subjects in a given region could ally and appoint a common satellite organization able to manage on their behalf a local congestion management market or to take other roles in the ancillary services provision. Clearly, the regulatory aspect is still completely to define, both at national and at EU-level. Within the SmartNet project, we are going to dedicate an important slice of activities to regulatory considerations during the last phases of the project (i.e. during 2018), when the results of the simulations and of the cost-benefit analysis will be known and consolidated. In fact, it is clear to everybody that, yet important, a sheer cost-benefit perspective is still too reductive. There are for sure regulatory constraints that derive from a lot of reasons and which have to be taken into account in order to decide which TSO-DSO coordination scheme could be more appealing and realistic by 2030, which is the time horizon of the SmartNet project.

  • How do you envision the distribution networks of the future?

Depending on their changing role and the functions they will be dedicated to, distribution networks could change considerably in the future. A well-known problem is constituted by the remounting power coming from an increasing number of distributed generation located in distribution, which conflicts with the fact distribution networks are usually planned for a mono-dimensional flow coming to the connection point to transmission to the loads. This problem is very known, but it is not the only one.

Congestion has been so far a rare problem in distribution (apart from some specific areas), but the increase of distributed energy sources connected to distribution could make it possible that congestion will be an important operative issue in the future. In this case, the present fit-and-forget strategy of the distributors could result no longer acceptable and their active participation in the electricity markets could become unavoidable (e.g. providing real-time measurements reconstructing network status, thus making a more realistic market clearing including congestion management possible). Hence, the need for more advanced TSO-DSO coordination schemes, as studied by SmartNet. Consequently, the rigidly radial structure of nowadays’ distribution grids could become a limitation, pushing planners to replace static reconfiguration options with a meshed network structure resembling the one of transmission grids, more robust against real-time congestion. Finally, the present environmental constraints and the NIMBY syndrome negatively affecting any transmission planning project and preventing to actuate a technically optimal expansion policy should affect also the distribution segment in the future. So, an optimized use of local flexibility as well as of a support of network devices could become even more interesting and economically appealing.

This interview was conducted at the European Utility Week and originally published here.