Distribution networks are faced with the growing challenge of hosting consumer-owned grid-integrated rooftop solar systems (RTS) or residential Solar PV on their networks. Exactly how much Solar PV capacity a LV feeder can comfortably host has been anybody’s guess. Estimates started at ~40% of feeder/transformer rating, when regulations allowing grid-integrated RTS were introduced some 20 years back. Due to growing pressure from regulators, consumer groups and policy makers, some distribution networks now consider up to 80% to be acceptable unless localised supply quality issues are experienced.
We believe that proper modelling using detailed data can provide evidence based reliable estimates. We therefore developed a model to simulate the operation of a typical LV feeder, under different scenarios to establish its maximum RTS hosting capacity. For simplicity, we considered the acceptable RTS hosting capacity to be one that does not result in reverse flow of energy into the grid.
We used this model to simulate the following ½ hourly profiles for one complete year (17,568 values per profile);
· combined generation of total installed RTS capacity – considering average yield
· total consumer demand without the effect of RTS, scaled to a maximum demand of 5MW
· net/balance supply from the grid
For consumer demand we first used a demand profile of a predominantly residential customer mix, with maximum demand appearing in the late afternoon. The simulation shows that peak generation output from a RTS capacity of 5.5MW would be just below the demand at that point (Figure 1), on a typical day. RTS capacities past this value will result in excess energies to produce reverse energy flows, varying throughout the year (Figures 2 & 3).
Fig 1
Fig 2
Fig 3
However, when the demand profile was changed to a day peaking commercial customer dominated mix, the same feeder could comfortably host 8MW of RTS (Figure 4).
Fig 4
Conclusion: The hosting capacity is not a hard and fast rule e.g. function of rating of feeder/ transformer, but rather the coincidence between RTS generation and demand profile.
Appropriate modelling using fine-interval data can provide investigative analytics; events of excess, and duration, magnitudes of excess power and excess energy during such events. Such investigative analytics can help DNOs, regulators and system operators to identify technically and economically most feasible options to increase the hosting capacity, e.g. network augmentation, network BESS, demand response and demand rationalization.
The fear is that market participant are using a current estimation approach that could be resulting in gross over and under estimation of actual hosting capacities of networks. Once a proper model is developed it can be deployed throughout the network for automated reporting of hosting capacities and follow up performance evaluations.
