针对频率规划局部搜索算法中存在的算法性能度量、性能比较及局部极小等问题 ,提出对多种结构实例模型和实例规模 ,以给定质量限下首次到达时间作为算法的性能度量 ,以搜索过程到达的多个质量限观察和比较算法之间的性能差距及其变化趋势 ,以全面评价算法的性能 ,并据此提出局部搜索算法设计的改进设想和方法 Aiming at the problems such as the performance metric, the performance comparison and the local minima existing in the algorithm of local search for frequency planning, this paper proposes a performance metric for a variety of structural example models and instance sizes with the first arrival time of a given quality limit as an algorithm, The process of reaching multiple quality limits observation and comparison of the performance gap between the algorithm and its changing trend in order to fully evaluate the performance of the algorithm and proposed local search algorithm design ideas and methods to improve