关键词: 微地形, 等级划分, 风速反演, 计算流体力学, 多元回归分析, 输变电系统

Abstract:

Micro-topography significantly influences the distribution of wind speed， temperature， and other factors. Different micro-topographic grades exhibit differences in meteorological amplification or shielding effects， resulting in significant differentiation of disaster risks such as icing， wind deflection， and tower collapse of transmission lines. Traditional qualitative micro-topographic classification relying only on topographic morphology cannot meet the requirements of refined disaster-prevention design for power grids. To achieve precise disaster prevention and control and ensure the safety of power transmission and transformation lines， a quantitative classification method for micro-topography linked to meteorological responses was established. Taking the Liupan Mountain area in Ningxia as an example， a quantitative framework for inverting micro-topographic grades based on wind speed variability characteristics was proposed. 61 852 sets of topographic data were extracted and inputted into a digital elevation model. Principal component analysis （PCA） and K-means clustering were combined to classify the regional micro-topography into three types： saddle， topographic uplift， and high-mountain watershed. Based on typical micro-topographic classification results， a coupled model of the relationships among slope， elevation， and local wind speed was established using multiple regression analysis （R²=0.987， significance level p<0.01）， statistically verifying the controlling effect of micro-topography on wind speed. Multicollinearity was eliminated through variance inflation factor and residual normality diagnostics， and model reliability was verified. On this basis， high-resolution wind field numerical simulations using computational fluid dynamics （CFD） were conducted to extract the spatial variability characteristics of wind speed under complex topographic conditions. In accordance with standards and regulations， a four-level wind speed classification standard was established and mapped onto micro-topographic units， enabling preliminary classification of topographic grades based on wind speed response. The CFD simulations clearly revealed that the constriction effect at saddles was the core cause of extreme wind and icing hazards on mountainous transmission lines， while turbulence intensity could directly reflect the degree of airflow disturbance caused by micro-topography. By integrating wind speed classification results with the quantitative relationship between topographic parameters， a classification framework for micro-topographic grades with explicit parameter ranges was established. The grade classification framework provides a quantitative basis for icing zone division， route optimization， and engineering wind load assessment of transmission lines in mountainous areas， and offers a methodological reference for differentiated prevention and control of micro-topographic disaster risks in similar areas. However， regional limitations still exist. Cross-regional applications require recalibration of the topography-wind speed regression relationship and adjustment of CFD boundary conditions and classification thresholds. In the future， field observations across multiple mountainous areas can be conducted to validate the generalization ability of the model.

Key words: micro-topography, grade classification, wind speed inversion, computational fluid dynamics, multiple regression analysis, power transmission and transformation system