Research on complex mountain wind power spectrum modeling based on frequency coupling mechanism

Traditional wind power spectrum models are often inadequate for complex mountainous terrain, where wind spectra deviate markedly from classical forms because of terrain heterogeneity and non-stationary wind conditions. To address this issue, this study proposes a wind power spectrum modeling method based on a frequency-coupling mechanism. By introducing low-frequency and high-frequency corrections, the proposed method simultaneously characterizes the large-scale energy distribution associated with terrain heterogeneity and wind-speed non-stationarity, as well as the small-scale energy distribution induced by turbulent intermittency and local shear, thereby improving the description of wind conditions in complex mountainous areas. Comparisons with field measurements from Yak Mountain in Sichuan, Yushe in Shanxi, and Mile in Yunnan show that, relative to classical models such as the Kaimal spectrum, the proposed method achieves better representations of both low-frequency energy and high-frequency fluctuations. The error between the proposed spectrum model and the measured data is generally controlled within 10%, with the average error reduced by approximately 20% and the maximum error within a local time interval reduced by more than 25%. In addition, time-series wind-speed reconstruction based on multi-period field data yields average errors of 8.17%, 9.42%, and 8.18% for the three sites, respectively, indicating better agreement with measurements than the classical spectral model. These results support the effectiveness of the proposed method for the tested complex mountainous sites and suggest its potential value for wind resource assessment, structural load simulation, and wind farm design.