Bayesian Estimates of Transmission Line Outage Rates That Consider Line Dependencies

Abstract

Transmission line outage rates are fundamental to power system reliability analysis. Line outages are infrequent, occurring only about once a year, so outage data are limited. We propose a Bayesian hierarchical model that leverages line dependencies to better estimate outage rates of individual transmission lines from limited outage data. The Bayesian estimates have a lower standard deviation than estimating the outage rates simply by dividing the number of outages by the number of years of data, especially when the number of outages is small. The Bayesian model produces more accurate individual line outage rates, as well as estimates of the uncertainty of these rates. Better estimates of line outage rates can improve system risk assessment, outage prediction, and maintenance scheduling.

Applying Bayesian estimates of individual transmission line outage rates

“…This section summarizes from [1] the inputs and outputs of the Bayesian hierarchical method of estimating line outage rates and some typical results. For reasons of space and complexity, all explanations and details of the Bayesian method itself are referred to [1].…”

“…Pooling does not work when there are multiple partial similarities. However, recent work [1] shows that the Bayesian hierarchical model can leverage multiple partial similarities to get better estimates of individual line outage rates from utility data. Typical results are that, for the lines with less frequent outages and using one year of data, the annual outage rate estimated with Bayesian methods has less than half the variance of the straightforward estimate.…”

“…If, more typically, the line has infrequent outages and the data is sparse, Bayesian methods can calculate a more accurate outage rate than non-Bayesian methods because any available prior information can be taken into account. The Bayesian method in [1] is a hierarchical model -data is shared across lines that have some partial similarity, such as similar location, proximity, rating, or length. As shown in [1], sharing data in this way can improve estimates of line outage rates.…”

“…The Bayesian method in [1] is a hierarchical model -data is shared across lines that have some partial similarity, such as similar location, proximity, rating, or length. As shown in [1], sharing data in this way can improve estimates of line outage rates. This is because the hierarchical model can learn from similar lines when estimating the outage rate for a line with little associated data.…”

“…This is because the hierarchical model can learn from similar lines when estimating the outage rate for a line with little associated data. There is much more explanation of the Bayesian hierarchical model and the detailed results of applying it to estimate line outage rates in [1].…”