Electrical and Computer Engineering Department, University of Wisconsin–Madison, Madison, WI 53706 USA.
Testing Branching Process Estimators of Cascading Failure with Data from a Simulation of Transmission Line Outages
Article first published online: 25 FEB 2010
© 2010 Society for Risk Analysis
Volume 30, Issue 4, pages 650–662, April 2010
How to Cite
Dobson, I., Kim, J. and Wierzbicki, K. R. (2010), Testing Branching Process Estimators of Cascading Failure with Data from a Simulation of Transmission Line Outages. Risk Analysis, 30: 650–662. doi: 10.1111/j.1539-6924.2010.01369.x
- Issue published online: 12 APR 2010
- Article first published online: 25 FEB 2010
- Blackout risk;
- branching process;
- cascading failure;
- electric power transmission system;
We suggest a statistical estimator to quantify the propagation of cascading transmission line failures in large blackouts of electric power systems. We use a Galton-Watson branching process model of cascading failure and the standard Harris estimator of the mean propagation modified to work when the process saturates at a maximum number of components. If the mean number of initial failures and the mean propagation are estimated, then the branching process model predicts the distribution of the total number of failures. We initially test this prediction on failure data generated by a simulation of cascading transmission line outages on two standard test systems. We discuss the effectiveness of the estimator in terms of how many cascades need to be simulated to predict the distribution of the total number of line outages accurately.