Robust model for weather-related contingency probability estimation used for risk based security assessment

Abstract

Weather is one of major factors that profoundly has an impact to the performance of a transmission line. Advanced knowledge and accurate estimation of the failure rate related to weather provides better overview on the probability of transmission line outage. Precise probability estimation is vital in risk based security assessment so as to reflect more accurate risk index value that enable correct judgments on power system security level. This paper introduced two approaches in the construction of an improved model to estimate the failure rate of a transmission line considering lightning and rain weather. The first approach is based on the identification of the best linear model that has the least error and significant correlation between the independent and dependent variables. The second focused on the robust estimator that has resistance to the presence of outliers in solving the multiple linear regression. In realizing the effectiveness of the proposed method, the results are compared with the traditional Ordinary Least Square (OLS) method. The results revealed that the proposed method using MM-estimator shows a significant improvement over OLS performance. © 2014 Praise Worthy Prize S.r.l. - All rights reserved.