Icing load accretion prognosis for power transmission line with modified hidden semi-Markov model

Ice accretion on power transmission lines is one of the major causes for cable failure in Zhaotong area, Yunnan Province, South China. This study proposes a method to predict the remaining-dangerous time (RDT) of the icing load accretion on an interval of the power transmission lines with modified hidden semi-Markov model (HSMM). Based on the predicted RDT of the cables during ice accretion, the appropriate preventative measures can be scheduled in advance by electric power companies. The estimation model with the learning algorithm of support vector machine for icing load accretion is built through historical icing load accretion data and meteorological conditions first. Then, the estimated icing load accretion sequence can be obtained through the estimation model by forecasting the meteorological conditions. The modified HSMM method can eliminate the possible underflow issue during computation, and be used to build the RDT prognosis model. With the estimated icing load accretion sequence and RDT prognosis model, the authors can predict RDT of the icing load accretion on an interval of the power transmission lines. The developed prognosis algorithm is verified through collected meteorological conditions and icing load accretion data on the Dazheng 73# power transmission line in Zhaotong area, Yunnan Province, South China.