提出一种以相对误差平方和(Sum squared relative error,SSRE)作为误差性能函数的反向传播(Back propagation,BP)神经网络算法(SSRE-BP),针对3种不同镁合金AZ31B、ZK60和AZ61A在单轴拉压、纯扭、45°比例和90°圆形非比例等4种不同加载路径下的疲劳寿命进行预测。并与以均方误差(Mean squared error,MSE)作为误差性能函数的传统BP神经网络(MSE-BP)以及基于临界平面法的SWT疲劳损伤模型预测的结果进行比较。结果表明,在3种镁合金材料总共138组疲劳数据中,神经网络只有一组预测值在3倍偏差界限外,而用SWT预测结果分别有16组、13组、10组数据在3倍偏差界限外。两种BP神经网络能够较好地预测镁合金不同加载路径下的疲劳寿命,相比于SWT疲劳模型预测的寿命在精度上有较大幅度的提升。其中,SSRE-BP算法的精度略高于传统的MSE-BP算法。
An improved BP network which use sum of squared relative error (SSRE) as the performance function is applied to predict the fatigue life of three kinds of magnesium alloy under different loading paths. Stain-controlled fatigue experiments are conducted on AZ31B and ZK60 magnesium alloy under four loading paths, which including fully reversed tension-compression, cyclic torsion, 45° in-phase axial-torsion and 90° out-of-phase axial-torsion. In addition, the fatigue data of AZ61A magnesium alloy from literature are also adopted. Two fatigue life prediction methods, namely, a standard BP network which use mean squared error(MSE) as the performance function, and Smith-Watson-Topper(SWT) critical plane fatigue models, are evaluated based on the experimentally obtained fatigue results. Result shows that all of the predicted results except one date by both BP network are within factor-of-three boundaries, there are 16 date, 13 date and 10 date predicted by SWT model outside factor-of-three boundaries even factor-of-five boundaries, respectively. Both BP network are found to be able to correlate the fatigue experiments reasonably well in comparison with SWT model, and SSRE-BP network is better than MSE-BP network.
[1] ALBINMOUSA J,JAHED H,LAMBERT S. Cyclic behaviour of wrought magnesium alloy under multiaxial load[J]. International Journal of Fatigue,2011,33：1127-1139.
[2] ALBINMOUSA J,JAHED H. Multiaxial effects on LCF behaviour and fatigue failure of AZ31B magnesium extrusion[J]. International Journal of Fatigue,2014,67：103-116.
[3] ALBINMOUSA J,JAHED H. Multiaxial behaviour of wrought magnesium alloys：A review and suitability of energy-based fatigue life model[J]. Theoretical and Applied Fracture Mechanics,2014,73：97-108.
[4] EXEL N,SONSION C M. Multiaxial fatigue evaluation of laserbeam-welded magnesium joints according to IIW-fatigue design recommendations[J]. Weld World,2014,58：539-545.
[5] XIONG Y,YU Q,JIANG Y Y. Multiaxial fatigue of extruded AZ31B magnesium alloy[J]. Materials Science and Engineering A,2012,546：119-128.
[6] 熊缨,程丽霞. 挤压AZ31B镁合金多轴疲劳寿命预测[J]. 金属学报,2012,12(48)：1446-1452.
XIONG Ying,CHENG Lixia. Multiaxial fatigue life prediction for extruded AZ31B magnesium alloy[J]. Acta Metallurgica Sinica,2012,12(48)：1446-1452.
[7] YU Q,ZHANG J X,JIANG Y Y. Multiaxial fatigue of extruded AZ61A magnesium alloy[J]. International Journal of Fatigue,2011,33：437-447.
[8] REIS L,ANES V,FREITAS M D. AZ31 Magnesium alloy multiaxial LCF behavior：theory,simulation and experiments[J]. Advanced Materials Research,2014,891-892：1366-1371.
[9] YU Q,ZHANG J X,JIANG Y Y,et al. An experimental study of cyclic deformation of extruded AZ61A magnesium alloy[J]. International Journal of Plasticit,2011,27(5)：768-787.
[10] FATEMI A,SOCIE D F. A Critical plane approach to multiaxial fatigue damage including our-of-plane loading[J]. Fatigue & Fracture of Engineering Materials & Structures,1988,11(3)：149-166.
[11] SMITH R N,WATSON P,TOPPER T H. A stress-strain parameter for the fatigue of metals[J]. Journal of Materials,1970,5(4)：767-778.
[12] JAHED H,VARVANI-FARAHANI A. Upper and lower fatigue life limits model using energy-based fatigue properties[J]. International Journal of Fatigue,2006,28：467-473.
[13] JIANG Y,SEHITOGLU H. Fatigue and stress analysis of rolling contact[R]. Urbana-Champaign：University of Illinois,1992.
[14] JIANG Y. A fatigue criterion for general multiaxial loading [J]. Fatigue & Fracture of Engineering Materials & Structures,2000,23(1)：19-32.
[15] VENKATESH V,RACK H J. A neural network approach to elevated temperature creep-fatigue life prediction[J]. International Journal of Fatigue,1999,21：225-234.
[16] MATHEW M D,KIM D W,RYU W S. A neural network model to predict low cycle fatigue life of nitrogen-alloyed 316L stainless steel[J]. Materials Science and Engineering A,2008,474：247-253.
[17] RODRÍGUEZ J A,HAMZAOUI Y EI,HERNÁNDEZ J A,et al. The use of artificial neural network (ANN) for modeling the useful life of the failure assessment in blades of steam turbines[J]. Engineering Failure Analysis,2013,35：562-575.
[18] LOTFI B,BEISS P. Application of neural networking for fatigue limit prediction of powder metallurgy steel parts[J]. Materials and Design,2013,50：440-445.
[19] LIN J W,ZHANG J H,ZHANG G C,et al. Aero-engine blade fatigue analysis based on nonlinear continuum damage model using neural networks[J]. Chinese Journal of Mechanical Engineering,2012,25(2)：338-345.
[20] GENEL K. Application of artificial neural network for predicting strain-life fatigue properties of steels on the basis of tensile tests[J]. International Journal of Fatigue,2004,26：1027-1035.
[21] XIANG K L,XIANG P Y,WU Y P. Prediction of the fatigue life of natural rubber composites by artificial neural network approaches[J]. Materials and Design,2014,57：180-185.
[22] ABDALLA J A,HAWILEH R. Modeling and simulation of low-cycle fatigue life of steel reinforcing bars using artificial neural network[J]. Journal of the Franklin Institute,2011,348：1393-1403.
[23] ARTYMIAK P,BUKOWSKI L,FELIKS J,et al. Determination of S-N curves with the application of artificial neural networks[J]. Fatigue & Fracture of Engineering Materials & Structures,1999,22：723-728.
[24] AL-ASSADI M,KADI H E,DEIAB I M. Predicting the Fatigue life of different composite materials using artificial neural networks[J]. Appl. Compos. Mater.,2010,17：1-14.
[25] AL-ASSAF Y,KADI H E. Fatigue life prediction of composite materials using polynomial classifiers and recurrent neural networks[J]. Composite Structures,2007,77：561-569.
