摘要

Ground movement data of a monitoring point from 2001 to 2010 in a mining area were analyzed by statistical methods. The results show that the autocorrelation AC) and partial autocorrelation PAC) of the sample data are clearly tail-dragged, which meet with ARMA(p, q) model. On this basis, parameters estimation and checking of model was done by the software EViews, ARMA prediction model of ground surface movement was established. Through comparing the forecasting results from the model with the monitoring data, it shows that the prediction model is a useful method with good precision, especially under short time step prediction.

全文