动态预测软土路基沉降的神经网络模型研究  

Study on the neural networks model used for dynamically predicting the settlements of soft ground

作　　者：胡伍生[1] 方磊[1] 

机构地区：[1]东南大学交通学院,南京210096

出　　处：《测绘科学》2008年第6期110-112,共3页Science of Surveying and Mapping

基　　金：国家973项目资助(2006CB705501);江苏省测绘科研基金项目资助(JSCHKY200608)

摘　　要：人工神经网络具有较强的非线性映射能力。本文介绍了神经网络BP算法的一些改进措施。这些措施可以提高BP算法的学习收敛速度,同时也可以提高BP网络性能的稳定性。为避免软土路基沉降传统计算方法中各种人为因素的干扰,本方法利用实测资料直接建模。基于改进的BP神经网络模型,建立了可依据现场量测信息对软基路堤沉降量随时间而发展的过程进行动态预报的分析方法。本文所建立的BP算法模型比较独特,利用该模型预测软土路基沉降精度高,预测结果的稳定性好。Some improved steps for the BP neural networks are introduced in this paper. The improved steps can increase the convergence speed of BP neural networks, and can improve the performance of BP neural networks. In this paper, on the basis of the improvement of BP neural network and field measuring information, the dynamic analysis method for predicting the highway settlement which varying with time is established. The BP model used for predicting the settlements of ground is particular. Since the model of this method is directly based on real samples, it can avoid the mistakes due to factitiousness in traditional methods. It is proved that the prediction model is accurate and the performance of the BP model in this paper is stable.

关 键 词：改进的BP神经网络 软土路基 沉降量 

分 类 号：P25[天文地球—测绘科学与技术]