摘要
本文提出了用改进的 BP网络对嵌岩桩承载力进行预测的新方法 ,并编写了相应的训练程序。实例计算表明 ,改进的 BP网络算法比一般的 BP模型结果误差小、收敛快 。
A new method of using improved BP network is presented to make a training program. The calculations show that the improved BP network is precisier and has a quick effect compared with the common BP model. And the method of artificial neural network is effective in forecasting the bearing capacity of the rock-socketed piles.
出处
《广西工学院学报》
CAS
2002年第3期9-12,共4页
Journal of Guangxi University of Technology
关键词
BP网络
嵌岩桩
承载力
人工神经网络
artificial neural network
rock socketed pile
bearing capacity