An Application of Grey-General Regression Neural Network for Predicting Landslide Deformation of Dahu Mine in China

The prediction of landslide sliding trend is complicated and dynamic system problem. Firstly, there are many uncertain factors in engineering geology, hydro-geological conditions and surrounding environment. Secondly, the engineering geology and surrounding environment are changing continually in the mining process. To solve above two problems, we proposed a model coupled Gray method and General Regression Neural Network (GM-GRNN). In this model, a time series of landslide displacement is expressed as trend item and its random item. Trend item of deformation, which is related by time, is predicted using Gray theory. Random item of trend item, which is a complex non-linear sequence, is calculated by General Regression Neural Network. The model uses the advantages of “accumulative generation” of a Gray prediction method, which weakens the original sequence of random disturbance factors, and increases the regularity of data. It also makes full advantage of the GRNN approximation performance, which has a fast solving speed, describes the non-linear relationship easily, and avoids the defects of Gray theory. The model is applied to the prediction of sliding deformation of Dahu landslide from Henan province in China. It shows that the approximation error of proposed model is about 0.1 m, and forecast error is about 0.2 m to meet project needs, and it is reasonable and reliable. The coupled model can forecast changes in the course of follow-up operation and development of landslides for mining.