基于MATLAB的ANN方法在桩基岩石强度预测中的应用  

Application of ANN Based on MATLAB to Forecasting the Strength of Pile Foundation Rock

作　　者：张向东[1] 于崇[1] 刘功勋[1] 

机构地区：[1]辽宁工程技术大学土木建筑工程学院,辽宁阜新123000

出　　处：《地下空间与工程学报》2005年第6期1143-1146,共4页Chinese Journal of Underground Space and Engineering

基　　金：辽宁省教育厅高等学校科学研究资助项目（2004D120）.

摘　　要：人工挖孔灌注桩广泛应用于工业与民用建筑、桥梁、港口、码头等工程建设领域，对于高层建筑，此种方法应用的更为普遍。桩基岩石强度直接决定工程质量和安全。利用MATLAB软件提供的改进人工神经网络算法。调用PRESTD、POSTSD、POSTREG函数进行样本处理和网络结果分析，建立数学分析模型。并且结合工程实例，对桩基岩石抗压强度进行预测。将预测结果与室内实验得到的抗压强度值及回归分析的预测值比较，得到良好的预测精度，有一定的实际应用价值。Manual drilling cast piles are used widely in the fields of architecture and civil-industrial buildings,bridges, seaports,docks and so on. This method is used more widely to construct high- rise buildings. The strength of pile foundation rock decides directly engineering quality and security. In order to present mathematical analysis model, the MATLAB software, which provides improved artificial neural network algorithm, is used and the functions of RRESYD, POSIS3D, POSTREG which treat the samples and analyze the net result are called. Moreover, combined with real engineering case, this method can be used to forecast the compressive strength of pile foundation rock. By comparing the predicted results with the compressive strength is gained by indoor testing and the regression analysis prediction value, the high forecasting precision was gained. This method has some value in practice.

关 键 词：桩基岩石强度 MATLAB编程 BP人工神经网络 回归分析 

分 类 号：TU473[建筑科学—结构工程]