基于ELM的大坝变形分析与预报模型

传统的诸如BP神经网络等学习方法训练时需要设置大量的参数,并且容易产生局部最优解。极限学习机(Extreme Learning Machine,ELM)可以随机选择输入权重以及隐藏层偏差且不需要调节,最终只产生唯一最优解。将ELM引入大坝变形分析建模中,建立了基于ELM的变形预报模型。实例表明,相比传统的逐步回归模型与BP神经网络模型,基于ELM的大坝变形预报模型在效率和精度上都有提高。

Dam Deformation Analysis and Prediction Model Based on Extreme Learning Machine

Traditional learning methods such like Back Propagation (BP) neural network training need to set a num-ber of parameters,and prone to local optimal solution. Extreme Learning Machine (ELM) randomly chooses the input weighs and the hidden layer biases and does not necessarily tuned. Finally it generates a unique optimal solution. In this paper,ELM algorithm is introduced in dam deformation analysis modelling,establishing a dam deformation and prediction model. Experimental results show that compared with BP neural networks model,the dam deformation prediction model based on ELM have improved on efficiency and accuracy.