A probabilistic approach to power system security assessment under uncertainty

The deepening penetration of renewable resources, such as wind and photovoltaic solar, has introduced additional uncertainty into power system operation and control. This added uncertainty, together with the conventional sources of uncertainty, the loads and the availability of resources and transmission assets, makes clear the limitations of the conventional deterministic power flow in power system analysis and security assessment applications. Therefore, the explicit consideration of uncertainty requires the deployment of probabilistic approaches so as to provide the ability to manage the wide spectrum of all possible values of the input and state variables. In this paper, we make use of cumulant-based probabilistic power flow methodology to account for correlations among the input random variables. Extensive testing indicates good performance of probabilistic power flow. We illustrate application of the probabilistic power flow on the 14-bus IEEE test system and present a comparison with the result obtained by the computationally more demanding Monte Carlo approach. The probabilistic power flow results provide valuable information for power system analysis and security assessment and, in particular, provide insights into issues associated with line overloading, over-/under-voltage, and the critical ramping requirements from conventional generators in system with deep penetration of highly variable resources, such as wind farms.