THE IMPACT OF DISTRIBUTED HYBRID PHOTOVOLTAIC BACKUP SYSTEMS ON SHARED RESIDENTIAL FEEDERS Courtney K. Rhoda 1 , Justice Chihota 2 and Bernard Bekker 3 1,2,3 Department of Electrical and Electronic Engineering, Stellenbosch University, South Africa E-Mail: 1 crhoda@sun.ac.za Abstract Battery energy storage systems will increasingly be connected to shared low voltage (LV) feeders, as the uptake of electric vehicles (EVs), hybrid photovoltaic (PV) backup systems (i.e. grid-interactive PV systems with self-consumption and uninterruptable power supply functionalities) and other behind-the-meter storage technologies rises. While there are many benefits from the increase in these technologies they may also pose several issues. This paper discusses the potential impacts of hybrid PV system installations on LV networks in various scenarios of net load capacity (the offset between generation and consumption), grid access regulations and the customer’s battery-use behaviour. Using one of the scenarios, the paper demonstrates the potential impacts of increased hybrid PV system penetration on voltage levels, phase unbalance and thermal loading of the feeder, referenced against the relevant quality of supply standards. A stochastic-probabilistic approach is used to conduct the simulations; the Monte Carlo Simulation method is used to simulate the stochastic nature of the unknown hybrid PV system placement while the extended Herman Beta transform accounts for the uncertainty and variability in both the PV generation and loads. The results show that hybrid PV systems can cause the violation of voltage unbalance limits even if injection into the grid is not allowed. Further simulations demonstrate that the distribution of customers along the feeder affects the extent of the unbalance and thus the permissible penetration. Keywords: hybrid PV systems; stochastic PV distribution, probabilistic load flow; LV network hosting capacity, PV grid impacts. 1. Introduction The placement, size and usage patterns of battery energy storage systems (BESS) on shared LV feeders are not centrally planned, but rather decided by the end customer, informed by technology pricing and electricity pricing signals, amongst other factors. This is similar to the roll- out of embedded generation (EG) on shared feeders. The technical impacts of the random and subsequently difficult-to-predict roll-out of BESSs will also be similar to that of EG: the introduction of current flows for which the LV feeder was not designed, impacting feeder voltage profiles, thermal loading and phase unbalance. Regulations like NRS097-2-3 [1] provide some guidelines on managing the impact of the roll-out of EGs but does not include BESS yet. Currently in South Africa, the BESS with the highest uptake is likely to be hybrid PV backup systems, primarily due to frequent load shedding, favourable return on investment of PV and self-consumption requirements. This paper focuses on the impacts of these hybrid systems on shared LV feeders, further limited to residential applications to allow for a sufficient depth of analysis. The primary objective of this paper is to gain a better understanding of the effects on voltage level, thermal loading and phase unbalance as the number of hybrid PV backup systems connected to a shared residential feeder increases. The research applies a stochastic-probabilistic methodology initially developed by Gaunt et al. [2] and recently extended for enhanced accuracy and further applications [3]. This methodology is explained and was used successfully in [4] for LV feeders with PV EGs without storage and with feedback limits of 50% of the customer’s rated circuit breaker. In this paper, it is used to map the impacts of hybrid PV systems at different penetration levels, for a large number of placement scenarios that are randomly generated. The difference with BESSs compared to EGs is however that many variations of charge-discharge schemes exist, defined by customer behaviour, pricing signals, regulations to name a few, compared to EGs based mainly on solar irradiation. The value of this work will be in understanding how these different charge-discharge schemes correlate to the technical impacts as a function of uptake. The next section discusses the potential technical impacts of BESSs on the LV grid. It also illustrates how various system configuration constraints affect these technical impacts. Section 3 explains various power generating energy system configurations and how they affect the