Distribution networks have been traditionally conceived for
transporting electricity downstream into low voltage demand
nodes. However, the connection of significant amounts of
distributed generation may reverse this condition, resulting in
distribution nodes exporting power to other parts of the
network. The current planning standard of the UK distribution
networks (Engineering recommendation P2/6) requires making
available sufficient capacity and redundancy for downstream
flows under peak demand levels. However, it does not
explicitly consider the implications of DG-mediated flow
reversals that may cause flow constraints under circuit outage
conditions. Relying on a Monte Carlo approach to sample wind
and demand with adjustable correlations, this paper provides
insights into the risks associated with an increase of variable
distributed generation to the point where reverse flows may
exceed the connection capacity under circuit outage conditions.
Remote tripping schemes that disconnect distributed
generators upon occurrence of a fault are explored to mitigate
outage related costs. The latter strategy carries benefits but also
novel risks in the form of a reliance on real-time
communication and control, which may malfunction. It is
shown that even unreliable corrective actions convey
significant benefits to system reliability.