岩石本构关系的径向基函数神经网络快速逼近模型
A fast approaching model for rock constitutive equation by radial basis function neural network
-
摘要: 建立了径向基函数神经网络快速逼近模型 ,基于具体条件下岩石力学实验数据 ,对相应岩石的本构关系进行逼近 ,实例分析表明 ,该模型不仅对具体应力 -应变关系能够很好逼近和预测 ,而且逼近速度快、稳定性好 ,对岩石力学快速、高效数值方法的进一步发展具有重要参考价值。Abstract: A fast approaching model for rock constitutive equation using radial basis function neural network has been built. It is shown that the model can approach the stress-strain curve accurately, rapidly and stably by laboratory data. The model is valuable for developing the numerical method in rock mechanics.