Prediction Model of Runway Settlement at a Southwest Airport
The settlement of airport pavements seriously affects the safe operation of airports.Accurately predicting the settlement of runway construction is extremely important for airport construction and operation.Based on the observed data of runway settlement and deformation at a southwest airport,the hyperbolic model,logarithmic model,exponential model,and GM(1,1)model were used to predict and compare the runway settlement.The issue of low prediction accuracy of curve prediction under small sample size and the problem of poor accuracy of the GM(1,1)grey model in nonlinear prediction were addressed to improv the prediction accuracy.Furthermore,by using Back Propagation Neural Network to correct the residuals of the combined prediction model,the accuracy and effectiveness of the model prediction were maximized.Reference is provided for predicting ground settlement.