Abstract: There are many methods for analyzing the data of low strain integrity testing on foundation piles, but there are a lot of the artificial interferences in the data processing. Based on the powerful nonlinear reflection and training function of artificial neural networks,the model of BP neural network for foundation piles integrity testing is put forward. According to in-situ measurements, every interference in the data processing can be avoided in this model. At last, the model is applied to the analyzing a case history. The results of training and examination show that this method is speediness and convenience on the pattern identification of pile integrity.