基于马尔科夫链的云模型在某碾压混凝土重力坝渗透压力预测中的应用

Application of Cloud Model Based Markov Chain in Dam Seepage Forecasting

为了同时反映大坝渗流序列的整体规律和多种影响因素造成的不确定性,提出了基于马尔科夫链的云模型的预测方法,即先建立基于云理论的预测模型,再通过马尔科夫链对其误差进行修正,并在建立云预测模型的过程中,提出设立过渡时期的改进方法以提高部分时期预测精度。通过对某重力坝渗透压力的预测分析,并与ARIMA、ANN模型结果进行对比,发现该方法预测值精度较高,具有较好的实用性。

To reveal the overall rule and randomness of dam seepage series caused by influencing factors, a cloud pre- diction model based Markov Chain is proposed. Firstly, a prediction method is established based on the cloud model. And then the Markov chain is used to correct the predictive errors of the model. Meanwhile, an improved method of setting a ＂transition period＂ is proposed to increase the prediction accuracy of some periods. By analyzing seepage pressure predic- tion of a certain gravity dam and comparing with the ARIMA and ANN models, it shows that the prediction precision of the proposed method is higher and it has good practicability.