In order to study the predictive ability of BP neural network on compressive strength of CFRP-confined concrete and the output performance of neural network model，this paper establishes a BP neural network prediction model for the compressive strength of CFRP-confined concrete based on a large number of experimental data. The influence of data combinations on the prediction accuracy of the neural network model is investigated. Based on the theory of neural network，the high-precision BP network model is generated into general formulas and simplified formulas for application convenience，and the prediction results of the neural network models and the empirical formulas are compared and discussed. The analysis results show that: BP neural network can well mine the data information of the input and output parameters and obtain a prediction model with high accuracy; compared with the traditional regression models，the simple linear equation derived by using purelin instead of sigmoid as the transfer function only adds a constant term，and the average value of prediction result/test result is 1.011，and the coefficient of variation is 0.112，showing a higher prediction accuracy and stability.