摘要
提出了深基坑变形预测的进化支持向量机方法。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。利用优化后的模型对基坑实例进行了变形预测,并将预测结果与监测结果进行了对比。研究结果表明,该模型与神经网络模型相比,具有预测精度高、泛化能力强等优点,对基坑安全监控具有实用价值。
It proposes a new method to predict deformations in deep excavation based on genetic arithmetic and support vector machine.The parameters of the SVM model optimized by genetic arithmetic and best parameters are obtained.The model is verified with the experiment datum,result of prediction by the optimized SVM model is compared with the test datum.The application shows that this model is better than the models based on BP neural networks;It possesses the advantage of high accuracy of forecasting and high ability of generalization.The calculation result shows that the deep excavation deformation predicted by this model is in good agreement with the observation data.
出处
《武汉理工大学学报》
CAS
CSCD
北大核心
2010年第1期158-161,共4页
Journal of Wuhan University of Technology
基金
河南省高校青年骨干教师资助项目(2004099)
关键词
深基坑
变形
遗传算法
支持向量机
deep excavation
deformation
genetic arithmetic
support vector machine
