DEFORMATION FORECASTING WITH A NOVEL HIGH PRECISION GREY FORECASTING MODEL BASED ON GENETIC ALGORITHM

Ning Gao, Cai-Yun Gao

School of Geomatics and City Spatial Information, Henan University of Urban Construction, Pingdingshan City, Henan Province, China, 467036

The precision of prediction of grey forecasting model depends on the conformation of background vale and the selection of the initial condition. Existent literatures optimized grey forecasting model just from one side, respectively. Therefore, a novel model named BIGGM (1,1) is proposed in this paper by integrated optimizing background value and initial condition. In addition, genetic algorithm has also been integrated into the new model to solve the optimal parameter estimation problem. An illustrative example of deformation of Lianzi cliff dangerous rock along the Yangtze River in china is adopted for demonstration. Results show that the BIGGM (1,1) model can increase the prediction accuracy, and it is suitable for use in modelling and forecasting of deformation.