Application of improved grey system in tunnel monitoring

Abstract

The prediction of deformation law of tunnel surrounding rock is an important measure to ensure construction safety. Affected by engineering geology and construction factors, tunnel deformation prediction has volatility and randomness. Based on the grey theory, the background value of GM(1,1) prediction model was optimized by using the cotes integral formula, and the grey model (ncbc-gm (1,1)) was established according to the minimum rule of the sum of squares of relative error. The ncbc-gm (1,1) grey prediction model is used to predict the deformation trend of tunnel surrounding rock, and the prediction results are compared with the traditional GM(1,1) prediction model and the GM(1,1) prediction model based on background value construction optimization. The results show that the root mean square error of the ncbc-gm (1,1) grey model is 2.66mm and 0.10mm lower than that of the traditional GM(1,1) prediction model and the optimized GM(1,1) prediction model based on the construction of background value, respectively, and the average relative error is 6.76% and 0.23% lower, respectively. The prediction accuracy is obviously improved, which provides an important basis for tunnel deformation prediction.