The aircraft engine is a crucial component of aircraft. The failure of aircraft engines usually leads to serious accidents and significant economic losses. Accurate remaining useful life (RUL) prediction is crucial for enhancing the reliability of aircraft engines and flight safety. This paper proposes a method for RUL prediction based on performance trajectory similarity and transfer fusion. Firstly, time-domain moving for historical performance data is used to select similar degradation trajectory. Degradation trajectory data of similar engines is selected for model training. Subsequently, a RUL prediction model based on long short-term memory (LSTM) networks is constructed to explore the relationship between engine monitoring data and RUL. Multiple prediction models based on similar engines are used for RUL prediction of the test engine. Finally, the different prediction results are fused according to the weights of similar engines, thereby obtaining the final RUL prediction result. Besides, the feasibility and effectiveness of the proposed method are validated based on the C-MAPSS simulation dataset.