Research On Spatiotmeporal Dynamic Deformation Analysis And Prediction Methods Based On Grey Model

Deformation process of the deformable body is a complicated nonlinear dynamics system affected by many factors of man himself can’t fully clear comprehensive effect, the great uncertainty are embodied in the deformation system and its potential emergent behavior. The existence of system inside and outside disturbance, as well as the limitations of human cognitive ability, cause the deformation of the inner operation mechanism, the relationship between structure, are shown as grey, therefore, take the deformation system as a grey system to study will be perfect. This article revolves the topic about "dynamic deformation analysis and prediction", to further research the several problems of modeling analysis of dynamic deformation based on the deformation of grey system point of view, the main research contents and results are as follows:1、Refined the gray of dynamic deformation of engineering deformation based on the grey theory basic principle and characteristics, and on this basis, studied the main content architecture of the three key technologies (includes "gray modeling preprocessing", "the gray model", " the error identification and control of grey model") about grey deformation analysis and prediction theory (that deformation analysis and forecast based on grey System theory).2、Resolved the modeling mechanism and the whole process of the classic GM(1,1) prediction model which take the discrete displacement time series of spatial single point as the modeling object; Further constructed a spatial single-point multidimensional grey forecasting model based on GM(1,h), example analysis shows that multidimensional grey model is better than unidimensional grey model.3、Analyzed the limitations of single point prediction model, constructed a spatial multipoint static grey forecasting model based on space sequence which has an important significance for fitting and forecasting subsidence trend in a certain space direction; Further constructed a spatial multipoint grey forecasting model which take into account the spatio-temporal correlation between the deformation monitoring points, example analysis shows that the model prediction effect is superior to single-point grey model.4、Analyzed a series of error problem about modeling mechanism of classical model of GM(1,1). Build a set of optimization mechanism about grey forecasting model. And then, Explores the selection optimization problem of albino background parameter by using the genetic algorithm (GA), example analysis shows that GA can effectively improve the precision of gray model, and IRCGA-GM is better than SBCGA-GM, SRCGA-GM and GM in the aspect of modeling prediction effect.