一种基于 BP 神经网络的尾矿坝沉降预报方法

基于BP神经网络建立尾矿坝沉降预报模型,重点对BP神经网络的拓扑结构和学习算法进行研究。并以某尾矿库初期坝的沉降监测数据为例,对模型的拟合、预测精度进行验证。实例表明,BP神经网络自学习、自组织能力强,具有极强的线性逼真能力,能够准确地反映输入、输出变量之间的非线性关系,有效地表征尾矿坝的沉降变形规律,对即将发生的变形情况做出科学、合理的预报。

A way to predict the settlement of tailings dam based on BP neural network

The tailings dam sedimentation forecast model is established based on BP neural network ,with emphasis on the topological structure of BP neural network and the learning algorithm .And the fromer subsidence monitoring data of tailings dam has verified the model fitting and prediction accuracy .T he practice shows that ,BP neural network can do well in self‐learning and self‐organizing ,and have a strong linear realistic ability to accurately reflect the input and output of a nonlinear relationship between variables in order to present the settlement law and to make the scientific and rational prediction against the upcoming deformation .