膨胀土等级判别的遗传支持向量机多分类方法.pdfVIP

学兔兔 第31卷第4期 土 木 建 筑 与环 境 工程 Journal of Civil，ArchitecturalEnvironmental Engineering Vo1．31 No．4 2009年8月 Aug．2009 膨胀土等级判别的遗传支持向量机多分类方法 师旭超，郭志涛 (河南工业大学土木建筑学院，郑州450052) 摘 要：针对支持向量机模型中的参数难以确定的状况，提出了遗传支持向量机方法，即利用遗传 算法来搜索支持向量机与核函数的参数，避免了人为选择参数的盲目性，同时提高了支持向量机的 推广预测能力，并将该方法应用于膨胀土胀缩等级的判别分类问题。考虑影响膨胀土判别的重要 因素，选用液限、胀缩总率、塑性指数、天然含水量和自由膨胀率5个指标作为模型的判别因子，以4 类膨胀土胀缩等级作为相应的输出，以膨胀土实测数据作为学习样本进行训练，建立相应分类函数 对待判样本进行分类。研究结果表明：遗传支持向量机模型分类性能良好，预测精度高，是膨胀土 胀缩等级分类的一种有效方法，可以在实际工程中进行推广。 关键词：支持向量机；遗传算法；膨胀土；等级；判别 中图分类号：TU443 文献标志码：A 文章编号：1674—4764(2009)04—0044—05 Multi-·classification Method of GA—SVM on Identifying Grade of Expansive Soils sHt Xu-chao，GUO zhi．tao (Department of civil engineering，Henan University of technology，Zhengzhou 450052，P．R．China) Abstract：Aiming at the fact that parameters in support vector machine(SVM)model were difficult to be identified，a genetic algorithm SVM(GA-SVM)was proposed to avoid the blindness in parameter choosing and ireprove the estimation ability of SVM 。in which the parameters in SVM and kernel function were searched by genetic algorithm．And it was then applied to the classification for the swell and shrink grade of expansive soils． Five indexes including liquid limit，total swell—shrink ratio，plasticity index，water contents and free expansive ratio were adopted as discriminated factors． And the four grades of the expansive soils were the outputs correspondingly． Classification function was obtained through training a 1arge set of expansive samples．And it was shown that the classification