A probabilistic risk-based security assessment tool allowing contingency forecasting

Abstract

Extreme weather events are getting more and more frequent due to climate changes and often determine large load disruptions on power systems; this calls for the analysis of the vulnerabilities to natural threats which may cause multiple, dependent contingencies. In this context, exploiting the data coming from forecasting systems in a risk-based security assessment environment can help anticipate the most risky contingencies provoked by the weather event itself. The paper proposes an in-depth risk-based security assessment methodology, based on an extended definition of risk and aimed to predict the most risky contingencies which will affect the power system (contingency forecasting) on the basis of the k-hour ahead forecasts of the weather events. Big data analytics can be useful to get an accurate model for weather-related threats. The relevant software platform integrates the security assessment methodology with weather prediction systems. The application to a realistic wet snow threat scenario in the Italian transmission grid shows the added value of the proposed approach with respect to conventional security analyses, by defining a set of single and multiple contingencies evolving with the weather disturbance, thus complementing the conventional N-l security criterion.