摘要
应用遗传算法解决边坡稳定性评价的聚类分析问题。针对问题特点,提出了基于自然数的编码方案并设计了相应的选择、交叉和变异算子。计算结果表明,该方法能以较大概率找到全局最优解并能有效克服常规聚类方法中对初始聚类中心敏感以及聚类结果与样本输入次序有关等缺点。
A dynamic clustering method for evaluation of slope stability is developed based on genetic algorithm. By incorporating features of the problems discussed, the corresponding genetic operators such as selection strategy, crossover operator, and mutation operator are designed to promote global search. Analyzing results show that the method is easy to find global optimum solution and can resolve the defects in the ordinary dynamic clustering methods such as sensitivity to the original cluster center and clustering results depending on the order of input patterns.
出处
《四川建筑科学研究》
北大核心
2008年第5期124-127,共4页
Sichuan Building Science
基金
国家自然科学基金(70371046)
河北省科技厅资助项目(06547004D)
关键词
遗传算法
边坡稳定性评价
聚类分析
genetic algorithm
evaluation of slope stability
clustering analysis
