西风扰动是一种西风带风暴，更好的提取西风扰动的运动轨迹进而确定其活动范围对其经过的国家具有重要的价值。西风扰动轨迹提取中的关键步骤是最近邻的提取，目前较为常用的最近邻提取算法是k-dimension-tree（kd-tree）算法，但是由于地理数据的特性以及算法自身存在的缺点，在算法进行搜索时，存在许多搜索无关问题，降低了算法的时间效率。于是本文使用Best Bin First（BBF）算法对kd-tree算法进行优化，主要方式是对kd-tree算法中的回溯路径进行优先级的排序以减少搜索时间。并利用一年的西风扰动的质心的数据进行实验验证分析，结果表明BBF优化后的kd-tree算法，在运行时间上，减少了24%，并且可以保证结果的正确率。BBF算法的一个特征是可以对算法进行运行时长限制，但是当对BBF算法设定超时即停止时，随着运行时长的降低，结果的正确率也随之降低，这是BBF算法的一个缺陷。

Western disturbance is a kind of storm in the westerly zone. It is of great value for the countries it passes through to better extract the track of westerly disturbance and determine its range of activities. The key step of western disturbance trajectory extraction is the extraction of nearest neighbors. The k-dimension-tree (kd-tree) algorithm is the most commonly used nearest neighbor extraction algorithm at present. However, due to the characteristics of geographical data and the shortcomings of the algorithm itself, there are many search-independent problems in the search of the algorithm, which reduces the time efficiency of the algorithm. So this paper uses Best Bin First (BBF) algorithm to optimize the kd-tree algorithm. The main way is to rank the backtracking paths in the kd-tree algorithm in order to reduce the search time. The results show that the BBF algorithm can replace the kd-tree algorithm, reduce the running time by 24%, and ensure the accuracy of the results. One of the characteristics of BBF algorithm is that it can limit the running time of BBF algorithm, but when the BBF algorithm is timed out and stopped, with the decrease of running time, the correct rate of results will also decrease, which is a defect of BBF algorithm.