基于近红外光谱技术的石油组分定量分析新方法

A Novel Fuzzy Neural Network Method for Diesel Quantitative Analysis with Near Infrared Spectroscopy

针对石油化工产品生产控制和质量检查的需要,为提高测定产品组成的效率,将近红外光谱法作为基础测定方法,以直馏柴油、加氢精制柴油和催化裂化柴油为校正模型的训练样本,测定其中饱和烃、胶质、单环芳烃、双环芳烃、三环芳烃和环烷烃的组成,论述了采用模糊神经网络建立校正模型测定石油化工产品组成的可行性。基于dSPACE硬件平台,用验证样本对模糊神经网络校正模型进行了检验,实验结果表明,该方法响应快、误差小、鲁棒性强,在近红外长波区内,校正样品和验证样品的均方误差小于10-6。该方法可用于石油化工产品的生产工艺研究中。

To satisfy the requirement of the production control and quality inspection of petrochemical products,a novel fuzzy neural network control method is proposed for determining product composition by near infrared spectroscopy.For the data analysis three different diesel products were selected as samples and six analytical models,such as saturated hydrocarbons,polar compounds,monoaromatics,dicyclic aromatics,tricyclic aromatics and naphthenes,were developed with the proposed fuzzy neural network method.Based on dSPACE,the near infrared spectroscopy system real-time experimental platform has been established to testify and analyze different diesel samples.The experimental results show that the improved performance is superior because of its advantage of quick response and good robustness.The mean squared error（MSE） of calibration and prediction samples is of the order of 10^-6 in the spectral range of 800-2 300 nm.The developed method can be used in the research on petrochemical products processing.