Abstract:Wire icing is a disastrous phenomenon that can harm the operation of the power system,and has been identified a meteorological disaster focused on by the government,electric power,and meteorological departments.Accurately predicting conductor icing is crucial for ensuring the safe operation of transmission lines and reducing disaster losses caused by wire icing.In mountainous environments,the prediction accuracy of power line ice accumulation depends on the downscaling capability of numerical models for meteorological fields,but the research on coupled dynamical downscaling methods for icing prediction is insufficient.In this study,by using the CALMET downscaling model coupled with the WRF model,the meteorological field of the icing event in the Zhongtiao Mountain of southern Shanxi Province during the period of 5—10 January,2022 was simulated and evaluated.On this basis,these meteorological fields were used to drive the Makkonen model to simulate the conductor icing process under different topographies.The simulation results of the WRF and CALMET models were examined separately,and the conclusions are as follows:1) Compared with the WRF model,the meteorological field downscaled by the CALMET model can more appropriately represent the distribution of the near-surface temperature and wind fields under the complex topography.The low-level wind field simulated by the CALMET model during the icing period is more consistent with the distribution pattern of slope surface flow and terrain bypass flow in actual topography.Meanwhile,the simulated cold region (air temperature <0 ℃) has a larger extent than that of the WRF simulation,which is more consistent with the observation.2) The air temperature and wind field simulated by the CALMET model show high agreement with the observation.The root mean square error (RMSE) of air temperature in CALMET is reduced by 0.5—1 ℃ and the correlation coefficient is improved from 0.5—0.8 to 0.6—0.85 compared with the WRF model.In addition,the RMSE of wind speed in CALMET is reduced by 1 m/s and the correlation coefficient is improved by 0.2 compared with the WRF model.This also indicates that the meteorological field simulated by the WRF model combined with CALMET is closer to the icing environment.3) The Makkonen method coupled with the CALMET model is able to reasonably reproduce the spatial and temporal distribution of icing process under different topographies.The error of the simulated ice thickness at each tower is significantly reduced by 2 mm compared with that of the WRF model,and the lag time of the simulated ice initiation is significantly reduced.In addition,the spatial distribustion of several high icing thickness areas simulated by the CALMET model fits better with the distribution of terrain elevation,particularly at the higher elevation of the south slope of Zhongtiao Mountain,where the simulated ice thickness ranges from 3.2 to 6.4 mm,which is consistent with the observation.In summary,the results of this study will help to optimize the simulation effects of meteorological factors and wire icing events in mountainous areas,which in turn will provide further assistance to exploring the deeper influence mechanisms of different microtopographies on conductor icing.
