As a high precision mechanical component, aero-engine has important influence on aircraft performance and reliability. Accurate prediction of remaining useful life can reduce maintenance costs and reduce the occurrence of safety accidents. The existing prediction methods only focus on the temporal relationship between sensor data, ignoring the spatial relationship between sensors. This paper proposes a network model that integrates spatial-temporal features, and uses graph convolutional networks and long short-term memory to extract spatial and temporal features, respectively. The parallel structure is used to integrate the temporal and spatial features.The RMSE and Score of subdataset FD001 are 12.81 and 252.04 respectively.The experimental results show that the proposed method has higher prediction accuracy than other prediction methods.