摘要
本文介绍了在一个土钉支护工程中土钉应力的监测结果 ,同时将神经网络理论引入土钉应力的预测中 ,建立了土钉应力预测的时间序列神经网络模型 ,并用遗传算法改善BP网络的局部极小值问题 .最后 ,将预测结果与实测数据进行了比较 ,结果表明本文方法是有效的 .
This paper presents the results of inspection and prediction of the stress of nails for soil nailing.The time series neural network based model is proposed for predicting the stress of the nails. Genetic algorithm is employed to solve the local minimum problem of back\|propagation learning algorithm.The calculation results are compared with the observation data and good agreement is attained.
出处
《水利学报》
EI
CSCD
北大核心
2000年第10期61-65,共5页
Journal of Hydraulic Engineering
关键词
深基坑
土钉支护
土钉应力
监测
预测
Deep excavation
soil nailing
stress
inspection
prediction
artificial neural network
genetic algorithm
