Imprecise probabilistic framework for power grids risk assessment and sensitivity analysis

Reliability of electric power supply is major concerns for the future interconnected power grid. Probabilistic frameworks has been used to assess the power supply reliability and network risks considering both consequences and likelihood of face unexpected events. In general, those models relies on a variety of assumptions and on the availability of good quality data. Nevertheless, many are the cases affected by partially corrupted information, lack of knowledge and poor data quality. In the past just few works considered problems of imprecision when computing reliability index for power networks. The authors believe that further efforts should be devoted to the improvement of probabilistic frameworks and power grid risk probabilistic indices by including rational treatment of imprecision and without introducing unwarranted assumptions. In this study, a power grids risk assessment framework based on imprecise probabilistic approach is introduced. Adopting an imprecise probabilistic approach, relevant sources of uncertainty have been propagated through numerical model of the power gird and their effects quantified in the risk metrics. In particular, the effect of imprecise stochastic model for load demand and lack of knowledge on the system components (e.g. line failure rate) has been investigated. Computational time is generally a strong burden for imprecise probabilistic frameworks. Hence, linear approximation of sensitivity measure, i.e. line outage distribution factors, and parallelization strategy have been used as valuable tool to reduce the overall wall-clock time for the analysis. The framework has been tested on a realistic IEEE test power network, two different propagation strategies adopted and sensitivity analysis for imprecise input performed. The results are compared and discussed, both aleatory and epistemic uncertainties propagated in the risk index and the most relevant sources of epistemic uncertainty identified using sensitivity
analysis