BP人工神经网络在光谱定量预测中的应用

人工神经网络是模仿大脑神经元网络结构和功能而建立的一种信息处理系统,广泛的应用于各种波谱数据处理。误差反向传播多层前馈式网络(back-propagation network,简称BP网络)应用最广,发展最为迅速。将BP神经网络用于紫外-可见吸收光谱和拉曼光谱数据的定量分析和预测,与原文的一元线性回归模型数据处理方法相比,获得了比较满意的预测结果,预测精度有显著提高。这为相关的光谱分析和数据处理提供了一种更有效、更精确的方法。

The Application of BP Artificial Neural Network in Quantitative Prediction of Optical Spectroscopy

The artificial neural network is an information processing system established by imitating the structure and the function of cerebral neuron networks.It is widely used in data processing and analysis of various spectroscopies.Back-Propagation neural network is one of the most useful and perspective networks,therefore it is used in quantitative prediction of UV-Vis absorption spectroscopy and Raman spectroscopy in this paper.The results showed that the prediction accuracy has been remarkably improved compared with linear regression,and the satisfied prediction results were obtained.This provides a method more effective and more accurate for spectroscopic analysis and processing of spectral data.