Statistical analysis of field data for precipitation icing accretion on overhead power lines

This paper addresses the analysis of field data gathered at the Mont Belair icing test site in Quebec. Load cell evaluations of actual icing loads on an existing 315-kV line are correlated to hourly measurements of ambient temperature, wind speed, precipitation rate, and number of signals of the Ice Rate Meter (IRM), in order to establish a numerical model for precipitation icing accretion on overhead line conductors. The correlation analysis is limited to precipitation ice events, or those mixed with relatively short periods of in-cloud icing. Emission of IRM signals is used as a criterion to distinguish the accumulation phase of an ice event, from persistence and shedding, characterized by no emission of IRM signals. The results from the analysis show that the icing rate corresponding to wet growth is much larger than that in dry conditions. What is more surprising, it was also found that the icing rate during periods when winds blow parallel to the line axis is significantly greater than that with perpendicular winds. The linear fit to the set of multivariate data is usually applicable. However, in some relatively rare cases of wet growth in heavy precipitations without IRM signals, the linear model may be inadequate and quadratic polynomials must be used. The results from the application of the numerical model are in excellent agreement with the field observations. This empirical model can be very useful for evaluation of icing loads on energized transmission lines, when there are not available measurements by load cells or other direct methods.