基于工艺参数的7005 铝合金力学性能的支持向量 - 中国有色金属学报.PDFVIP

20 2 2010 2 Vol.20 No.2 The Chinese Journal of Nonferrous Metals Feb. 2010 1004-0609(2010)02-0323-06 7005 ( 400044) 7005 ( )(σ b σ0.2 HB) (PSO)(SVR) (LOOCV)7005 (PLS) (BPNN)PLS-BPNN SVR-LOOCV 3 (σ σ HB)(RMSE)4.531 9 MPa14.550 8 MPa HB1.414 2 b 0.2 (MRE)0.72%2.61%0.66%PLSBPNN PLS-BPNN RMSE MRE 7005 TP18 TG146.2+1 A Quantitative prediction of mechanical properties of 7005 Al alloys from processing parameters via support vector regression CAI Cong-zhong, WEN Yu-feng, ZHU Xing-jian, PEI Jun-fang, WANG Gui-lian, XIAO Ting-ting (Department of Applied Physics, Chongqing University, Chongqing 400044, China) Abstract: The support vector regression (SVR) approach based on the particle swarm optimization (PSO) for its parameter optimization, combined with leave-one-out cross validation (LOOCV), was proposed to predict the mechanical properties (tensile strength σ , yield strength σ and hardness HB) of 7005 Al alloys under different processing b 0.2 parameters including extrusion temperature, extrusion velocity, quenching type and aging time. The results strongly support that the prediction precision of SVR-LOOCV method is superior to those of partial least squares (PLS), back-propagation neural networks (BPNN) and their combination PLS-BPNN model by applying the identical dataset. The root mean square errors (RMSE) for σ , σ and HB achieved by SVR-LOOCV are 4.531 9 MPa, 14.550 8 MPa and b 0.2 HB 1.