The report presents an advanced model of overhead line vulnerability to the indirect effects of high wind and wet snow, which includes Lidar information on potential interference with the line by vegetation from outside the shear zone. Tests performed show that the advanced model avoids the failure probability overestimation errors, committed with the “base” model developed in the previous year, by providing failure return times consistent with the evidence from the exercise.

A descriptive statistical analysis is then presented of the hydrogeological events that have affected the national transmission power grid, highlighting two primary causes, floods and landslides. The analysis shows that the components most frequently impacted by floods are station equipment, while those most impacted by landslides are line supports. The non-negligible role of erosive phenomena on pylon legs is also noted.

With reference to these phenomena, we present the vulnerability model of pylon legs for the scouring phenomenon in which the flow of water (e.g., due to the overflow of a river) results in the removal of soil from the leg, causing potential mechanical instability. The model takes into account both the geometric characteristics of the pylon and the geotechnical characteristics of the soil on which it sits. The case study confirms the dependence of the conditional probability of failure (output of the vulnerability model) on both soil class and foundation design characteristics (e.g., foundation depth).

The report then proposes a definition of system vulnerability and a mathematical framework for formulating probabilistic indicators of system vulnerability, which can be used in control applications to minimize system degradation in the face of a specific threat.
Finally, with reference to uncertainties in the Terna-RSE methodology for network resilience assessment, the types of uncertainties affecting the methodology are identified and an overview of possible methods for quantifying them is presented.

A probabilistic method is then proposed for evaluating the effect of climate model uncertainty on the calculation of component failure return time: application on examples of climate maps and vulnerability curves shows the ways in which uncertainty on return times also varies spatially in different areas of the grid.