(武漢理工大學(xué) 機(jī)電工程學(xué)院，武漢 430074)
摘 要: 提出一種最小二乘支持向量機(jī)的Cu-0.75Cr銅合金反擠壓力預(yù)測(cè)新模型。以斷面縮減率、凸模錐角和擠壓溫度這3個(gè)主要工藝參數(shù)作為影響因素，以反擠壓過程的擠壓力為影響對(duì)象，通過最小二乘支持向量機(jī)模型建立影響因素和影響對(duì)象之間的復(fù)雜非線性關(guān)系。以正交實(shí)驗(yàn)數(shù)據(jù)為樣本對(duì)模型進(jìn)行訓(xùn)練，用訓(xùn)練好的模型預(yù)測(cè)在一定反擠壓條件下Cu-0.75Cr銅合金的擠壓力。結(jié)果表明：該模型不僅預(yù)測(cè)精度和處理速度大大高于人工神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)模型，而且建模速度也比標(biāo)準(zhǔn)支持向量機(jī)快，實(shí)際預(yù)測(cè)誤差小于3%。
關(guān)鍵字: Cu-0.75Cr銅合金；反擠壓；擠壓力；預(yù)測(cè)；最小二乘支持向量機(jī)
（School of Mechanical and Electronic Engineering, Wuhan University of Technology, Wuhan 430074, China）
Abstract:A novel prediction model for extrusion force of Cu-0.75Cr alloy reverse extrusion process based on least square support vector machine(LS-SVM) was proposed. With fault plane contraction rate, convex model awl angle and extrusion temperature as influence factors, and with extrusion force as influence object, the complex nonliner relations among the influence factors and influence object were fitted by LS-SVM model. Orthogonal experiment was taken to obtain data samples, and LS-SVM model was established through the data samples, so that the extrusion forces of Cu-0.75Cr alloy under different backward extrusion process conditions can be predicted by this model. The results show that not only the prediction accuracy and treatment speed by this model are much higher than those of artificial neural networks(ANN), but also the construction speed is higher than that of standard SVM, and the practical prediction errors are less than 3.0%.
Key words: Cu-0.75Cr alloy; backward extrusion process; extrusion force; prediction; least square support vector machine
