提出一种7075铝合金非等温固溶-锻造一体化热成形工艺。将固溶后铝合金直接放入温模中进行锻造，然后淬火并进行人工时效处理，通过构建温度-时间-性能（temperature-time-property，TTP）曲线，研究本工艺下入水温度和时效参数对7075铝合金微观组织和性能的影响，并结合机器学习对关键工艺参数进行优化匹配。结果表明：TTP曲线鼻端温度为315 ℃，合金时效后力学性能随入水温度的升高而升高，非等温锻时效后会出现双峰现象。在入水温度为380 ℃时，最佳时效参数为115 ℃-26 h，峰值硬度为182.38HV。训练后BP神经网络预测准度为94.9977%，对模型预测的最优工艺参数进行实验验证表明，其预测相似度为96.9%。与传统锻造工艺相比，本工艺能够在减少工序、降低能耗的同时，获得比传统锻造T6态7075铝合金更高的力学性能。

This paper proposes a non-isothermal solid solution-forging integrated hot forming process for 7075 aluminum alloy. After solid solution treatment， the aluminum alloy is directly placed into the mold for forging， then quenched and subjected to artificial aging treatment. The influence of water entry temperature and aging parameters on the microstructure and properties of 7075 aluminum alloy is studied under this process， through the construction of a temperature-time-property（TTP） curve. Additionally， machine learning techniques are integrated to optimize and match the key process parameters. The results reveal that the nose temperature of the TTP curve is 315 ℃， and the mechanical properties of the alloy increase with the increase of water temperature after aging， a double-peak phenomenon after non-isothermal forging and aging is observed. When the inlet temperature is 380 ℃， the optimal aging parameters are 115 ℃-26 h and the peak hardness is 182HV. After training， the prediction accuracy of the BP neural network model is 94.9977%. Experimental verification of the optimal process parameters predicted by the model shows that its prediction similarity is 96.9%. Compared with traditional forging processes， this process can achieve high mechanical properties than traditional forged T6-state 7075 aluminum alloy while reducing procedural steps and energy consumption.