The impact of introducing distributed energyresources at the low voltage side of the distribution grid iscurrently raising new challenges for utilities. In particular, thehigh penetration of photovoltaic panels (PVs) in radial grids isincreasing the active power losses in the branches and thevoltage level at some of the nodes. Principally nodes next to PVarray installations. This paper presents a methodology based ondesign of experiments (DOE) to detect such exposed nodes andbranches, together with the identification of the main scenariosthat cause such problems, characterized by: season, type of day,solar radiation and outdoor temperature levels. Themethodology is simulated on a LV network based on the Cigrebenchmark Grid with real utility data. The exposed nodes areclassified for each feeder from most to least problematic andshowed (as expected) sensitivity to seasonality (summertime),characterized by high solar radiation and outdoor temperatures.