灰色系统理论与神经网络的色谱保留值预测模型研究

灰色建模允许样本点少,不要求样本有较好的分布规律,而且计算量少,操作简便。而BP网络在对样本进行学习时,会对输出误差进行反馈校正,具有并行计算、分布式信息存储、容错能力强、自适应学习功能等优点。本文将灰色预测建模和神经网络技术融合起来建立灰色神经网络组合模型,应用于色谱保留值的预测。实证结果表明:该组合模型精度方面比常规灰色模型要好;组合的算法概念明确,计算简便,有较高的拟合和预测精度。组合模型的提出拓宽了灰色模型的应用范围。

Predictive model of chromatographic retention of bases on a combination of grey system theory and artificial neural networks

The basic grey model GM(1,1) does not require a large number of samples or a high degree of sample distribution,does not require a large amount of computation,and is easy to use. A back propagation (BP) network uses back propagation with learning samples,and has the advantages of being able to carry out parallel calculations,a distributed information memory and error tolerance. In this paper,we describe the combination of the grey prediction method with the neural networks method to establish a combined gre...