Canal
Statistical Analysis and Modeling in Evaluation of Effectiveness of the Reliability Standards for the North American Bulk Power System (303709)
*Svetlana Ekisheva, North American Electric Reliability Corporation
Keywords: Generalized Linear model, Explanatory and Predictive models, Reliability metrics, Reliability standards, Risk reduction evaluation
The North American Electric Reliability Corporation is an international regulatory authority whose mission is to assure the effective and efficient reduction of risks to the reliability and security of the North American Bulk Power System (BPS) that serves 334 mln. people. As part of its mission, NERC develops Reliability Standards for the BPS. Periodic reviews of a standard can include the analysis of the reliability data that evaluates the standard effectiveness in reduction of risk to the grid. Depending on a standard, reduction of risk could mean a decrease in number of occurrences of different types of disturbance events on the grid and their severity as well as a better BPS response to these events. Some conclusions about performance are results of routine statistical inference about significant changes in relevant reliability metrics (e.g. frequency of transmission outages), another require advanced statistical modeling. In this talk, we present case study analyses of effectiveness of several reliability standards. The discussed statistical approaches are easily generalized to different industries and fields where practitioners of statistics track changes in complex data.