非晶态合金磁性能与结构参数的定量关系

QUANTITAIVE RELATIONSHIP BETWEEN MAGNETIC PROPERLY AND STRUCTURAL PARAMETER FOR AMORPHOUS ALLOYS

用结构参数一模式识别／人工神经网络（ＡＮＮ）方法对非晶态合金系的磁性（包括饱和磁致伸缩系数、饱和磁感应强度、矫顽力）与组成、结构之间的关系进行了定性分析和定量计算，采用的结构参数有平均价电子数、混合熵、原子半径比、电负性差、功函数差和电子密度差等结果表明，定性分析结果与实验结论一致，定量计算结果与实验测定值符合较好。

The magnetic properties of thirty amorphous alloys have been qualitatively analyzed by means of partial least squares (PLS) method with structural parameters as features, and quantitatively calculated using artificial neural network (ANN) with the same parameters as inputs. The structural parameters involve electron concentration, ratio of radii, difference of electron density, difference of work function and mixed entropy. The qualitatively or quantitatively calculated results are in agreement with experimental ones basically.