多元回归法在复杂地形风资源微尺度模拟中的应用

随着风电事业的发展,复杂地形下的风资源模拟成为一项重要的课题。利用广东沿海某海岛南部9个测风塔共22个测风高度的完整一年的测风资料及该海岛高分辨率的数字地形资料,采用地理空间因子的多元回归法进行风资源的微尺度模拟并与WASP软件的计算结果进行了比较。结果表明：（1）年平均风速与地理空间因子存在显著的相关关系;（2）模拟的年平均风速平均相对误差为3.2%,年平均风功率密度的平均相对误差为3.5%;（3）经假设检验,对无测站的地点的年平均风速模拟效果良好;（4）该方法和WASP软件计算的风资源大值区分布基本一致。

Application of Multiple Regression to Wind Resource Micro-scale Simulation over Complex Terrain

With the development of wind electricity, wind resource simulation over complex terrain is becoming an important question for discussion. On the basis of wind observational data of one year of 9 observation towers at 22 levels in the southern zone of a certain coastal island of Guangdong Province, and the digital terrain data with high resolution of the island ,wind resource micro-scale simulation by multiple regression was done. In addition, the simulated result was compared with that obtained by WASP program. It was found that:(1) There is a significant correlation between annual mean wind speed and geographical spatial factors;(2) The mean relative errors of simulated annual mean wind speed and wind power density are 3.2% and 3.5%, respectively; (3) The assumption test shows that there is a good effect in annual mean wind speed simulation at those sites without observational data;(4) The distribution of the wind resource is basically consistent with that obtained by WASP program.