以Al-Cu合金为研究对象,研究不同含量的稀土元素Ce对Al-Cu合金的热疲劳性能的影响。通过热疲劳和力学拉伸实验,研究了Al-Cu合金的疲劳裂纹萌生及扩展规律。结果表明,在Ce含量为0.3%时,Al-Cu合金有最优的抗热疲劳性能。采用BP神经网络的方法预测Al-Cu合金热疲劳性能,预测结果表明,实验值与预测值间的相对误差均在5%以内,网络的验证误差在2%以内,说明使用BP神经网络的方法预测合金热疲劳性能是可行的。 Taking Al-Cu alloy as the research object, the effects of Ce with different contents on the thermal fatigue properties of Al-Cu alloy were studied. The fatigue crack initiation and propagation of Al-Cu alloy were studied by thermal fatigue and mechanical tensile tests. The results show that the Ce-content of 0.3%, Al-Cu alloy has the best thermal fatigue resistance. The prediction of the thermal fatigue properties of Al-Cu alloys by BP neural network shows that the relative errors between the experimental and predicted values ​​are within 5% and the verification errors of the network are within 2%, indicating that the method using BP neural network It is feasible to predict the thermal fatigue properties of the alloy.