基于人工神经网络的6063铝合金时效工艺的研究

Study on Ageing Regime for 6063 Aluminium Alloys Based on Artificial Neural Network

采用人工神经元网络对时效硬度进行预测,并对两阶段时效工艺参数对时效硬度的影响规律进行研究.研究结果表明:在本研究的温度时间范围内,第2阶段时效温度对时效硬度的影响非常显著,而第1阶段时效温度和时间及第2阶段时效时间对时效硬度的影响不太明显,这与正交试验结果相同.人工神经网络与正交试验分析方法相结合,确定的最优时效工艺参数为170℃×70 min+210℃×50 min.

Ageing hardness for 6063 Aluminium alloys has been predicted, and the effects of ageing process parameters on ageing hardness have been studied by means of artificial neural network. The results show： in the range of temperature and time of this study, the temperature and time of the firststaged ageing process and the time of the second-staged ageing process have less influence on ageing hardness, but the temperature of the second-staged ageing process has obvious influence on ageing hardness. This result accords with that of orthogonal experiment. Through combining artificial neural network and orthogonal experiment, the optimum two-staged ageing process parameters are 170 ℃ for 70 mins and 210 ℃ for 50 mins.