A New Method of Background Value Optimization in Grey Prediction Model

In order to improve the prediction accuracy of grey prediction model, the paper uses the compound idea of Simpson formula, which combines with dynamic model of non-homogeneous exponential function. It reconstructs background value in GM (1,1) model, to change the white differential equation of Verhulst model, a new Verhulst model is created and presented in the paper. Also three examples are given to verify the advantage of new Verhulst model, it is not only enriches researches of GM (1, 1) model, but also have high reference value for further study in Verhulst model.