Search

Icing disasters on power grid transmission lines can easily lead to major accidents, such as wire breakage and tower overturning, that endanger the safe operation of the power grid. Short-term prediction of transmission line icing relies to a large extent on accurate prediction of daily minimum temperature. This study therefore proposes a LightGBM low-temperature prediction model based on LassoCV feature selection. A data set comprising four meteorological variables was established, and time series autocorrelation coefficients were first used to determine the hysteresis characteristics in relation to the daily minimum temperature. Subsequently, the LassoCV feature selection method was used to select the meteorological elements that are highly related to minimum temperature, with their lag characteristics, as input variables, to eliminate noise in the original meteorological data set and reduce the complexity of the model. On this basis, the LightGBM low-temperature prediction model is established. The model was optimized through grid search and crossvalidation and validated using daily minimum surface temperature data from Yongshan County (station number 56489), Zhaotong City, Yunnan Province. The root mean square error, MAE, and MAPE of the model minimum temperature prediction after feature selection are shown to be 1.305, 0.999, and 0.112, respectively. These results indicate that the LightGBM prediction model is effective at predicting low temperatures and can be used to support short-term icing prediction.

GF-3 satellite is a remote sensing satellite, and it can provide data support for multiple fields, including transmission line and tower location surveillance. However, the existence of system errors reduces the accuracy of location, which restricts the generalization of GF-3 to those applications. Geometric rectification technique is focused on the elimination of system errors and geometric distortion in SAR images. In this paper, focused on the accurate transmission line and tower location, the DEM-based system errors calibration method is proposed, which is used for electrical power system. The experimental results show that the location accuracy of our method reaches the level of no more than 20 meters compared with the geolocation accuracy of 100 meters with real time ephemeris.

The unique geographical features and changeable climatic of Yunnan Province make the forest fire prone and frequent. Aiming at the problem of forest fire that may be suffered by overhead transmission lines and power grid equipment in Yunnan Province, the research on forest fire spread was carried out, and the method of combining Rothermel model and Huygens principle was mainly used to simulate forest fire spread. On this basis, the powerful graphic processing capabilities and spatial analysis capabilities of ArcGIS are used to graphically display the spread process to improve usability and visibility.

How to detect forest fire on time is an important social problem. Remote sensing detection with large detection areas and high time resolution has become a popular method in detecting forest fire. Therefore, we propose a new method of detecting forest fire based on Himawari-8 geostationary satellite data. The detection algorithm used is the GBDT(Gradient Boost Decision Tree) machine learning combined with multi-temporal information referred to as MT-GBDT algorithm. Considering the characteristics of Yunnan Province, the algorithm of this paper uses contextual algorithms to filter potential fire points. Then, combined with multi-temporal information, we processed the potential fire point by using the diurnal temperature cycle (DTC) model. Finally, we used GBDT machine learning model to process the information obtained in the above steps. Result shows that the fire point information with high accuracy is obtained. The probability of false detection is 18%, and the probability of detection is 86%. Compared with the existing three methods, MT-GBDT algorithm successfully detects most forest fires, while other algorithms omit many forest fires, and the probability of detection is 40% to 50%.

The Yunnan power grid has the characteristics of wide distribution and large geometric size, which makes the power grid vulnerable to natural disasters such as wildfires. The unique topography and climatic conditions of transmission lines crossing mountains, forests and other zones can easily lead to the outbreak of mountain fires, or even the tripping of transmission lines. In serious cases, transmission towers may be destroyed, thus causing major power accidents. The occurrence of mountain fire is closely related to topographic factors, and the change of each topographic factor (elevation, slope, slope direction, etc.) has a certain influence on fire. This paper mainly studies the relationship between three terrain factors and fire. Based on the statistical analysis of remote sensing data from 2005 to 2020, the terrain features and their influence of fire spots in Yunnan province were analyzed. The results show that most of the fires occur within the range of 1000-2000m above sea level, and the fires occur frequently. The frequency of fire occurring below 1000m and above 2000m is low. The occurrence frequency of fire is higher in the gentle slope range, among which, the number of fire point samples is the largest within the range of 5°-15°, followed by 0°-5°and less than 15°.In this paper, eight slopes are divided to make statistics of fire points, and the number of fires in different slopes also fluctuates, among which the south slope has the highest fire risk, accounting for 16.0%.The West Slope accounted for only 10.0%, with the lowest fire risk. Terrain factor is a fixed variable that has a fixed influence on the regional fire risk. Therefore, terrain analysis of fire is of great significance for the prevention of mountain fires in power grid.