Massive data analysis to assess PV/ESS integration in residential unbalanced LV networks to support voltage profiles

Significance Statement

Photovoltaic systems are the most preferable form of distributed generation in low voltage distribution networks. The photovoltaic units can significantly enhance the performance of the system. When loads are supplied locally, developing new photovoltaic-based ancillary services presents a superior opportunity for the distribution system operator to enhance power quality, minimize losses, and defer investment.

However, increasing the number of photovoltaic installations, the stochastic and intermittent production poses technical and economic challenges for the distribution system operator. Precisely, in case the photovoltaic surpasses the local demand, the excess power may lead to reverse power flows in the feeder, and sometimes may lead to voltage rise. In addition, in case the power demand is too high as compared to the photovoltaic generation, there may occur consistent voltage drops. In the two cases, voltage violations appear in the network.

Distributed generation units can support the grid in a bid to fix these challenges providing an ancillary service. Italian researchers, Francesco Lamberti, Vito Calderaro, Vincenzo Galdi at University of Salerno, in collaboration with Giorgio Graditi at Research Center of Portici proposed a comprehensive analysis based on the Monte Carlo analysis which was able to evaluate the effects that a local integration of photovoltaic units with energy storage systems have in offering voltage support. They were motivated by the fact that it is important to develop an analysis based on various periods of the year that permit a complete view of the combined photovoltaic energy storage systems potential support the network. Their research work is now published in Electric Power Systems Research.

Both stochastic and solar production and demand behaviors under varying photovoltaic energy storage system penetration levels must be taken into consideration in a bid to offer voltage support via the energy storage system in low voltage networks. Therefore, the authors needed a precise model for the entire system.

The researchers adopted a system that was made of feeders, one transformer, loads, photovoltaic units, meters, and energy storage systems. They used a shunt reactance and a series impedance for every feeder section. In addition, they modelled the photovoltaic units and the loads as P-Q buses. Generator as well as load profiles were modelled by the means of time-series profiles.

The researchers then conducted massive data analyses in a bid to exhibit the advantages that the integration of the energy storage systems with photovoltaic units possesses in a low voltage network in terms of voltage quality enhancement. For this reason, the energy storage system allowed for promoting self-consumption and reduced the risk of photovoltaic disconnection owing to voltage infringements.

The outcomes indicated that an indirect voltage support could be reached by energy storage installations to support distribution system operator as well as local consumption. The co-located systems were monitored my customers without extra meters on the grid. The potential to offer ancillary services by residential energy storage systems can therefore help the deployment of battery systems in a bid to become economically attractive.


Francesco Lamberti, Vito Calderaro, Vincenzo Galdi, Giorgio Graditi. Massive data analysis to assess PV/ESS integration in residential unbalanced LV networks to support voltage profiles. Electric Power Systems Research 143 (2017) 206–214

Go To Electric Power Systems Research