支持向量机在桩基低应变检测中的应用

The Application of Support Vector Machine to Low Strain Integrity Testing of Foundation Piles

桩基础属于隐蔽工程,其完整性检测的优劣受人为的影响因素较多。由于支持向量机的分类算法在模式识别方面的优越性,其学习性能出色,推广能力强。为了克服桩基础检测质量受人为因素影响较多之缺点,通过小波分析提取影响桩身质量的因素,利用支持向量机的分类算法判断桩基完整性。训练时以影响桩身质量的因素作为输入参数,训练和测试结果表明该方法能快速有效的对桩基质量进行模式识别。

Foundation piles are covert engineering, whether the testing result is good rest with the testing man. The support vector machine has advantages in pattern identified and it is also good in learning and application prospects. In order to overcome the disadvantage that the results rest with the man, the wavelet analysis method is used to get the affected factors of piles, then using the classification algorithm of support vector machine to estimate the integrity of foundation piles. The affected factors of piles are input parameters during learning. The results of training and examination show that this method is speediness and convenience on the pattern identification of pile integrity.