(中南大學(xué) 材料科學(xué)與工程學(xué)院， 長(zhǎng)沙 410083)
摘 要: 利用人工神經(jīng)網(wǎng)絡(luò)對(duì)7055鋁合金二次時(shí)效熱處理工藝參數(shù)與時(shí)效性能樣本集進(jìn)行訓(xùn)練和學(xué)習(xí)， 采用改進(jìn)的BP網(wǎng)絡(luò)算法Levenberg-Marquardt算法， 建立7055鋁合金二次時(shí)效熱處理工藝BP神經(jīng)網(wǎng)絡(luò)模型。 針對(duì)二次時(shí)效工藝特點(diǎn)， 研究的工藝參數(shù)包括: 預(yù)時(shí)效溫度、 預(yù)時(shí)效時(shí)間、 二次時(shí)效溫度和二次時(shí)效時(shí)間。 結(jié)果表明: 神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)值與實(shí)驗(yàn)值吻合較好， 說(shuō)明神經(jīng)網(wǎng)絡(luò)模型具有較高的精度及良好的泛化能力, 可有效地用于預(yù)測(cè)和分析二次時(shí)效工藝參數(shù)對(duì)7055鋁合金時(shí)效性能的影響。
關(guān)鍵字: 7055鋁合金； 二次時(shí)效； 人工神經(jīng)網(wǎng)絡(luò)； Levenberg-Marquardt算法
（School of Materials Science and Engineering, Central South University,
Changsha 410083, China）
Abstract: A model was developed for modeling the correlation between process parameters of second aging treatment and properties of 7055 Al alloy by applying the artificial neural networks (ANN). According to the feature of second aging, the process parameters were preliminary aging temperature, preliminary aging time, second aging temperature and second aging time. The model was based on error back-propagation (BP) algorithm and trained by Levenberg-Marquardt training algorithm. After the ANN model was trained successfully, the model achieved a very good performance. The results show that the model has high precision and good generalization performance, and can be successfully used to predict and analyze the influence of secondary aging treatment on the mechanical properties of 7055 Al alloy.
Key words: 7055 Al alloy; secondary aging; neural networks; Levenberg-Marquardt algorithm
