基于三层BP网络的分馏产品质量综合评价

提出了一种基于人工神经网络的分馏产品质量综合评价方法.首先介绍了一种归一化效用函数,把不同类型、不同量纲的原始评估值转换到[-1,1]区间,该效用函数较好的体现了"奖优罚劣"原则,同时对于神经网络又更容易学习和训练;其次,详细介绍了基于人工神经网络的分馏产品质量综合评价原理,并将之运用于实际,取得了较满意的效果;最后,简述了基于神经网络的分馏产品质量综合评价的优点.

Synthetic evaluation to the quality of fractionation production based on neural networks

In this paper,a method of synthetic evaluation to the quality of fractionation productions based on artificial neural networks is presented.First,a kind of utility function,which changes evaluation values with different dimensions and different kinds of indexes into values in the range of-1,1]with no dimension,not only emphasizes on the principle of"rewarding good and penalizeing bad"but also makes it easy for us to train neural networks to be used in this paper.Second,the principle of synthetic evaluation to the quality of fractionation production based on artificial neural networks is described in detail and applied to practice.The results are satisfactory.Last,the characteristics of the method proposed at this paper are mentioned in the end of this paper.