Abstract: Wavelet neural network is a kind of neural networks, which can closely combine wavelet theory with neural network theory, and avoid the blindness of framework designs for BP neural networks and the problem of nonlinear optimizations, such as local optimization.So it can greatly simplify the training of neural networks.It has good abilities of function learning and dissemination with a vast range of prospects for application.In this paper, the wavelet neural network is applied to predict the strength of concrete.The results show that the convergence speed of wavelet neural network is faster and its predicting result is more accurate than that of BP network