机构地区: 中国地质大学
出 处: 《莱阳农学院学报》 2002年第4期307-309,316,共4页
摘 要: BP神经网络具有很强的非线性映射和自适应学习功能,可用于模式识别和预测评估等领域。本文应用BP神经网络的理论和方法,并使用MATLAB语言编制了相应的程序和用户界面,选取孔压、锥尖阻力、孔压参数作为输入参数,进行土层划分。应用结果表明,将BP神经网络用于土层划分是可行的。 The BP neural network has ability of nonlinear-mapping and self-accommodating.It can be used to recognize mode, forecast and so on. In this article the theory and method of BP neural network are used, corresponding program and UI are written by the MATLAB language. The pore water pressure, cone resistance and the ratio of side friction to pore water pressure are selected as input-parameter to compartmentalize the soil. The results indicate that it is feasible to compartmentalize soil by BP neural network.