Study of the growth rate of wire icing under the conditions of coexisting freezing rain and supercold fog Weather

Using two case studies of icing growth observed in Lu Shan during the winter of 2016 under freezing rain and supercooled fog coexistence weather conditions, and combining meteorological elements with supercooled fog microphysical parameters, the correlation between icing growth rate and these elements was investigated. In addition, four commonly used machine learning models: Random Forest (RF), Support Vector Machine (SVM), Convolutional Neural Network (CNN), and Extreme Learning Machine (ELM) were used to simulate the icing growth process. The results show that in both cases, rain rate (R), temperature (T), wind speed (WS ≥ 1 m/s), and wind direction (WS) were significantly positively correlated with the icing growth rate. In the case with stronger WS (Case 1), the icing growth rate was negatively correlated with the number concentration (N), liquid water content (LWC), mean diameter (MD) and mean volumetric diameter (MVD), with WS being the most important factor in the RF model. The growth of icing was mainly driven by the enhancement of R, which caused an increase in LWC. In the case with weaker WS (Case 2), the icing growth rate was positively correlated with N, LWC and MD, with the MVD of supercooled fog droplets being the most important factor in the RF model. The growth of icing was primarily due to an increase in the number of water-phase particles and an overall increase in particle size, leading to an increase in LWC. All four machine learning models were able to successfully simulate the icing growth process, with results superior to traditional empirical formulas and numerical simulations. Among them, the RF, SVM and CNN models performed particularly well.