Foundation settlement forecasting using the new BP-Gompertz model

摘要

In view of each advantages of Gompertz and BP neural network, the Gompertz growth curve model was combined with BP neural network. The BP-Gompertz foundation forecasting model was proposed through using the capability of approximating the true value of BP neural network to optimize the curve fitting capability of Gompertz. An example shows that compared with the traditional Gompertz model, the prediction accuracy of the new BP-Gompertz model is significantly improved. The model provides a new method for the foundation settlement prediction.