The increasing penetration of PV into the distribution grid leads to congestion, causing detrimental power quality issues. Moreover, the multiple small photovoltaic (PV) systems and battery energy storage systems (BESSs) result in increasing conversion losses. A low-voltage DC (LVDC) backbone to interconnect these assets would decrease the conversion losses and is a promising solution for a more optimal integration of PV systems. The multiple small PV systems can be replaced by shared assets with large common PV installations and a large BESS. Sharing renewable energy and aggregation are activities that are stimulated by the European Commission and lead to a substantial benefit in terms of self-consumption index (SCI) and self-sufficiency index (SSI). In this study, the benefit of an LVDC backbone is investigated compared to using a low-voltage AC (LVAC) system. It is found that the cable losses increase by 0.9 percent points and the conversion losses decrease by 12 percent points compared to the traditional low-voltage AC (LVAC) system. The SCI increases by 2 percent points and the SSI increases by 6 percent points compared to using an LVAC system with shared meter. It is shown that an LVDC backbone is only beneficial with a PV penetration level of 65% and that the BESS can be reduced by 22% for the same SSI.

This paper proposes a novel feature construction methodology aiming at both clustering yearly load profiles of low-voltage consumers, as well as investigating the stochastic nature of their peak demands. These load profiles describe the electricity consumption over a one-year period, allowing the study of seasonal dependence. The clustering of load curves has been extensively studied in literature, where clustering of daily or weekly load curves based on temporal features has received the most research attention. The proposed feature construction aims at generating a new set of variables that can be used in machine learning applications, stepping away from traditional, high dimensional, chronological feature sets. This paper presents a novel feature set based on two types of features: respectively the consumption time window on a daily and weekly basis, and the time of occurrence of peak demands. An analytic expression for the load duration curve is validated and leveraged in order to define the the region that has to be considered as peak demand region. The clustering results using the proposed set of features on a dataset of measured Flemish consumers at 15-min resolution are evaluated and interpreted, where special attention is given to the stochastic nature of the peak demands.

Residential electricity consumers at present pose a major challenge to simulate due to their inherent stochastic nature. However, as this sector represents approximately one fourth of the total electricity consumption across the European Union, it is necessary to gain more insight in their consumption, both on an individual as well as on an aggregated level. In this work, we propose the use of load duration curves (LDCs) to characterize residential consumers. We construct LDCs for individual as well as artificially aggregated groups of residential consumers, and present how these LDCs can lead to a better understanding of what influence the aggregation level has on key characteristics of the consumption profile, how aggregation affects the height and duration of the peak demand, and what inherent trade-offs are present at higher aggregation levels.