论文部分内容阅读
通过实时采集正常操作条件下及发生异常工况时的甘草提取液的动态近红外光谱数据,结合主成分分析法(PCA)、偏最小二乘回归法(PLSR)和平行因子-偏最小二乘回归联用法(PARAFAC-PLSR)建立3种甘草提取过程的实时监测模型,并分析各种模型的特点。结果表明,基于3种方法建立的模型均能在一定程度上预测异常加热工况的发生,但同时也存在一定误判。其中,PCA方法建立的模型出错率最高,在60 min之前就出现3次“故障”误判,不适用于该过程的分析应用。而PLSR和PARAFAC-PLSR模型基本效果相似,校正集相关系数分别高达0.934 2,0.928 1,验证集相关系数也分别达到了0.856 7,0.828 3;并且这2种方法建立的预测模型误判率较低,首次成功预测的故障均发生于75 min。此外,PLSR和PARAFAC-PLSR模型均能在一定程度上预测出系统状态的走势。说明基于动态近红外光谱动态数据建立的PLSR和PARAFAC-PLSR模型均具有良好的在线监测和预测功能,为中药提取过程动态监测方法的优化选择提供了参考依据。
The dynamic near-infrared spectroscopy data of licorice extract under normal operating conditions and abnormal working conditions were collected in real time. Combined with principal component analysis (PCA), partial least-squares regression (PLSR) and parallel factor-partial least squares Regression combined method (PARAFAC-PLSR) to establish three real-time monitoring of licorice extraction process model, and analyze the characteristics of various models. The results show that all the models based on the three methods can predict the abnormal heating conditions to a certain extent, but there are also some misjudgments. Among them, the error rate of the model established by the PCA method is the highest, and 3 times of “fault ” misjudgment occurred before 60 minutes, which is not suitable for the analysis and application of this process. The PLSR and PARAFAC-PLSR models have similar basic effects, the correlation coefficients of the calibration set are as high as 0.934 and 2.0.928 1, respectively, and the correlation coefficients of the validation set also reach 0.856 7 and 0.828 3, respectively. And the false positive rates of the two models Low, the first successful prediction of failure occurred in 75 min. In addition, both the PLSR and PARAFAC-PLSR models can predict the state of the system to a certain extent. The PLSR and PARAFAC-PLSR models established based on dynamic near-infrared spectroscopy data have good on-line monitoring and prediction functions, which provide a reference for the optimization of dynamic monitoring methods of traditional Chinese medicine extraction process.