Abstract :
This paper presents the results of an analysis of outage data recorded over a 12-year period on 112 345-kV transmission lines of the Commonwealth Edison Company in Illinois; the lengths range from 3.5 to 188 miles. A linear relationship is developed between outage rate and line length; it does not pass through the origin but above it, indicating a residual outage rate corresponding to zero line length. This residual outage rate is attributed to terminal conditions and equipment, and is viewed quantitatively by equating it to a fixed number of miles of line length. Because of the wide scatter of data points, the data were grouped by line-length in 10-mile increments. The outage and years in service for each group were combined, and then analyzed by groups. The Gaussian method of least squares is used to fit a line to these points. The important value, is the outages/year due to the terminal conditions and equipment (ie, where the length of the transmission line equals zero and outages are still indicated). For accurate representation of this relation, only the values representing the largest number of data points, are included in the calculations. These show an annual failure rate of 0.7 which compares most favorably with the 0.6 found with the Markov Model (ie, by developing a failure and repair rate matrix due to the dependent outage combined with its probability matrix. A graphic method for determining the effects of terminal conditions and equipment has also been introduced.
