Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Urban Environment & Urban Ecology》 2008-01
Add to Favorite Get Latest Update

Integrated Application of Time Series Model and Markov Model in Land Subsidence Prediction

GUO Jia-wei1,SHAO Chuan-qing2,WANG Jie2,YI Li-xin1(1.Tianjin Key Laboratory of Environmental Remediation and Pollution Control,Tianjin 300071,China;2.College of Environment Science and Engineering,Nankai University,Tianjin 300071,China)  
The interannual variation process of land subsidence is non-stationary and random.Dynamic prediction based on observation data is a kind of important method in forecasting the land subsidence.When using time series model to predict,a difference approach should be carried out to make the observation data stable and there should be abundant in raw data for obtaining high-precision predicting results.Since the predicted value fluctuate flatly,this model is appropriate for short-term forecast.By using state transition probability matrix,Markov processes can accomplish the modeling with less data and predict with stronger data fluctuation.In this paper,we combine time series model with Markov model,and put forward the integrated application of the two models.This method selects time series model as a prediction tool,and revises the result by using state transition probability matrix in Markov model.The practical prediction showed that the integrated model was more suitable for the prediction when the observation data were limited and fluctuated greatly,and the results of the integrated madel had higher accuracy than the one predicted by using only time series model.
【CateGory Index】: F224
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved