Correlating Grid-operators' Performance with Cascading Failures in Smart-Grids

In this paper, the role of human factors associated with the grid operators, e.g., human-error probability (HEP) are modeled as a function of the grid conditions as well as operators' training and experience levels. Moreover, the HEP is embedded in a previously reported Markov-chain model that generates the probability distribution of blackout as a function of time following a trigger. Specifically, through the HEP, the Markov-chains transition matrix includes the dynamics of detailed smart-grid operator attributes. To derive the grid-state dependent HEP, three real-valued performance shaping factors (PSFs), representing key human attributes of the operators, are mapped to the grid-state variables, thereby capturing the correlation between the evolution of the PSF levels and the propagation of transmission-line failures. This mapping is established based on a histogram-equalization principle, which utilizes the experimentally-estimated probability distribution of the PSF levels while assuming a monotone relationship between the HEP values and number of line failures. Further, the distribution of the PSF levels was used to identify the critical combinations of PSF levels that corresponds to an event with high joint probability as well as a high HEP.