基于支持向量机的双桥静力触探地层自动划分方法

Automatic Method of Soil Layers Division Using Double Bridge Static Cone Penetration Test Data based on Support Vector Machine

目前,根据双桥静力触探曲线进行地层划分多采用目测经验法和分类图法,目测经验法人为主观性强,并且耗费人力和时间,而分类图法在某些地区适用性不高。为解决此问题,研究了一种基于支持向量机的双桥静力触探地层自动划分方法。以锥尖阻力和侧摩阻力作为输入属性,土类名称作为数据标签,对模型进行训练,将训练好的模型用于土类识别和地层自动划分。应用此方法对天津某地铁工程的双桥静力触探孔进行了地层自动划分,准确率达100%。研究结果表明,基于支持向量机的双桥静力触探地层自动划分方法可以较好进行土类识别和地层划分。

At present,the experience method and classification charts are mostly used to divide the soil layers according to the curve of double bridges static cone penetration test.The experience method is highly subjective and needs excessive manpower and time.The applicability of classification charts is different in different regions.In order to solve this problem,the automatic method of soil layers division was studied using double bridge static cone penetration test data based on support vector machine,which took advantage that support vector machine could solve the classification problem.The model was trained with cone tip resistance and side friction resistance as input attributes and soil type name as data label.The trained model was used for soil type identification and automatic soil layers division.This method was applied to the automatic soil layers division using double bridges static cone penetration test data in a project in Tianjin,and accuracy reached 100%.The method carry out soil classification and divide soil layers effectively.