It is of great significance to accurately predict and evaluate the development and management of polymer flooding. Reservoir micro pore structure parameters are important factors affecting the development effect of polymer flooding, and there is a nonlinear and uncertain complex relationship between them. The parameters which have great influence on the EOR of polymer flooding are selected by correlation analysis. The nonlinear and Uncertain Multivariable System is predicted by polynomial regression analysis and BP neural network. The results show that the artificial neural network method has better adaptability, It can better reflect the internal relationship between various micro parameters affecting polymer flooding effect and EOR value, and the prediction accuracy is high. Therefore, it is considered that the application of BP neural network to predict polymer flooding effect is feasible and effective.