论文部分内容阅读
针对悬臂施工的连续刚构桥施工过程中标高误差产生的因素多,各因素与误差值之间存在高度非线性关系并属于小样本事件,所以准确的标高误差预测一直成为一项难题。该文利用最小二乘支持向量机,借助Matlab中的工具箱,建立预测模型,对一座大跨连续刚构桥施工中节段标高误差进行预测,并与BP神经网络预测模型进行对比,可知LS-SVM预测结果准确、稳定性优良。
There are many factors that affect the elevation error of continuous rigid frame bridge during cantilever construction. There is a highly nonlinear relationship between each factor and the error value and it is a small sample event. Therefore, accurate elevation error prediction has been a difficult problem. In this paper, the Least Squares Support Vector Machines (LS-SVM) are used to predict the height error of a large span continuous rigid frame bridge by using the toolbox in Matlab. The prediction model is compared with the BP neural network prediction model. -SVM accurate prediction results, excellent stability.