摘要
首次提出神经网络与马尔可夫链相结合的数学模型,用于随机波动数据序列变形预测,克服随机观测数据固有的随机和波动特征,较好地实现了工程中大量随机波动性数据无法采用灰色GM(1,1)模型精确预测问题。采用神经网络完成观测数据变化动态基准模型的计算,在此基础上应用马尔可夫链确定系统状态转移概率矩阵,通过系统状态的划分、样本值与模型拟合值之间的残差及其中误差等指标的分析,完成观测数据变化值的准确计算。该模型被应用于隧道围岩工程实例,计算证明取得了较好的效果。
For the first time,Markov chain model based on artificial neural network are integrated to forecast for random series of measurement data which are great randomness and fluctuation.As usual it used GM （1,1）model.A neural network is used to get the dynamic baseline for measurement data.Then Markov chain is applied to achieve state transition probability matrix.The data interval is forecasted and analyzed in the form of probability by the system state classification,the calculation of the residue between true value and model fitting value.This model is applied to analyze the stability of surrounding rock in tunnel.The result shows that it is practicable to predict.
出处
《计算机工程与应用》
CSCD
北大核心
2006年第6期225-226,232,共3页
Computer Engineering and Applications
关键词
BP-马尔可夫模型
围岩
沉降变形
预测
BP-Markov chain,surrounding rock,settlement deformation,forecast
