偏最小二乘(Partial least square,PLS)是一种基于数据驱动可以处理多个因变量对多个自变量的回归建模方法,因其具有提取质量相关信息的特性,在质量相关复杂工业过程监控中得到广泛的应用,成为近几十年复杂工业过程故障检测和诊断领域...偏最小二乘(Partial least square,PLS)是一种基于数据驱动可以处理多个因变量对多个自变量的回归建模方法,因其具有提取质量相关信息的特性,在质量相关复杂工业过程监控中得到广泛的应用,成为近几十年复杂工业过程故障检测和诊断领域的研究热点.对此,介绍线性、非线性、动态PLS模型及其故障检测技术.首先,介绍标准PLS模型,在此基础上对传统PLS模型进行细化分并指出其优缺点,针对标准PLS存在的两个问题以及工业过程数据的两种极端情况,从数据预处理类、多空间类和分块类三方面梳理线性PLS模型的发展和改进历程;其次,将非线性PLS模型扩展方法分为两类,重点介绍核函数非线性PLS模型的研究现状;再次,指出动态扩展方法的两种基本思路,对PLS动态模型进行分类,阐明动态特性的成因,从本质上揭示两种动态扩展方法的原理,按照分类综述动态PLS模型的发展现状;最后,指出该领域亟需解决的问题和未来研究方向.展开更多
针对具有动态特性的质量相关的故障检测问题,提出了一种基于自回归移动平均模型(auto-regressive moving average exogenous,ARMAX)的动态全主成分回归(dynamic total principal component regression,DT-PCR)方法.该方法基于输入的时滞...针对具有动态特性的质量相关的故障检测问题,提出了一种基于自回归移动平均模型(auto-regressive moving average exogenous,ARMAX)的动态全主成分回归(dynamic total principal component regression,DT-PCR)方法.该方法基于输入的时滞值,形成增广输入矩阵;然后将形成的增广矩阵分解成质量无关和质量相关两个正交部分,并根据这两个部分对应子空间的统计量设计出一个更简单的故障诊断策略.该方法对于输出的预测精度也优于以往的方法.最后,通过一个数值例子和田纳西—伊斯曼过程将DT-PCR与全潜结构投影模型(total partial least squares,TPLS)进行对比,验证了DT-PCR的输出预测性能及与质量相关的故障检测性能.展开更多
As a dynamic projection to latent structures(PLS)method with a good output prediction ability,dynamic inner PLS(DiPLS)is widely used in the prediction of key performance indi-cators.However,due to the oblique decompos...As a dynamic projection to latent structures(PLS)method with a good output prediction ability,dynamic inner PLS(DiPLS)is widely used in the prediction of key performance indi-cators.However,due to the oblique decomposition of the input space by DiPLS,there are false alarms in the actual industrial process during fault detection.To address the above problems,a dynamic modeling method based on autoregressive-dynamic inner total PLS(AR-DiTPLS)is proposed.The method first uses the regression relation matrix to decompose the input space orthogonally,which reduces useless information for the predic-tion output in the quality-related dynamic subspace.Then,a vector autoregressive model(VAR)is constructed for the predic-tion score to separate dynamic information and static informa-tion.Based on the VAR model,appropriate statistical indicators are further constructed for online monitoring,which reduces the occurrence of false alarms.The effectiveness of the method is verified by a Tennessee-Eastman industrial simulation process and a three-phase flow system.展开更多
Tomato fruits (Solanum lycopersicum, L.) were exposed to origanum oil, vinegar, or chlorine at 12℃ and 95% RH. Quality-related attributes were examined during or following vapour treatment. Fruit pre-treated with nat...Tomato fruits (Solanum lycopersicum, L.) were exposed to origanum oil, vinegar, or chlorine at 12℃ and 95% RH. Quality-related attributes were examined during or following vapour treatment. Fruit pre-treated with natural volatiles and transferred to ambient air had less decay compared to control fruits. Vinegar-enrichment maintained fruit firmness and acidity during exposure, though the effect was sustained only for acidity following storage in ambient air. Glucose and fructose levels were higher in origanum oil-treated fruit, while ascorbic acid content increased in volatile-treated fruits following transfer/storage in ambient air. Total phenolics declined during vinegar and chlorine exposure but tended to increase following transfer to ambient air. Vinegar and chlorine increased lycopene concentration during exposure to vapours, and this effect persisted only for vinegar following exposure and transfer in ambient air. Origanum oil-treated fruit increased lycopene concentration following exposure and transfer in ambient air. Vapour- treatment did not affect fruit weight and ethylene production. Natural volatiles may maintain fruit quality in addition to the well documented antimicrobial protection during fresh produce storage and transit.展开更多
在工业生产中,对系统进行故障检测具有十分重要的作用.改进的偏最小二乘(modified partial least squares,MPLS)是在PLS基础上提出的一种扩展算法,在质量相关故障检测中具有良好的检测效果,但当测试数据中含有质量无关故障时,MPLS算法...在工业生产中,对系统进行故障检测具有十分重要的作用.改进的偏最小二乘(modified partial least squares,MPLS)是在PLS基础上提出的一种扩展算法,在质量相关故障检测中具有良好的检测效果,但当测试数据中含有质量无关故障时,MPLS算法漏报率较高.另外,MPLS算法的阈值为固定值会导致其误报率增加,这些问题会对工业过程监控产生较大影响.鉴于此,提出一种基于局部信息增量与MPLS的质量相关故障检测方法(local information increment-MPLS,LII-MPLS).在MPLS基础上,通过使用局部信息增量技术对测试数据进行实时更新检测后,质量相关故障的漏报率明显降低.同时,过程复杂化导致静态控制限不能满足故障检测的需求,现存的动态控制限适用范围具有一定局限性,因此改进静态控制限将其推广为局部动态阈值.最后,通过田纳西伊士曼过程(Tennessee Eastman process,TEP)仿真实验验证了所提出算法的有效性.展开更多
文摘偏最小二乘(Partial least square,PLS)是一种基于数据驱动可以处理多个因变量对多个自变量的回归建模方法,因其具有提取质量相关信息的特性,在质量相关复杂工业过程监控中得到广泛的应用,成为近几十年复杂工业过程故障检测和诊断领域的研究热点.对此,介绍线性、非线性、动态PLS模型及其故障检测技术.首先,介绍标准PLS模型,在此基础上对传统PLS模型进行细化分并指出其优缺点,针对标准PLS存在的两个问题以及工业过程数据的两种极端情况,从数据预处理类、多空间类和分块类三方面梳理线性PLS模型的发展和改进历程;其次,将非线性PLS模型扩展方法分为两类,重点介绍核函数非线性PLS模型的研究现状;再次,指出动态扩展方法的两种基本思路,对PLS动态模型进行分类,阐明动态特性的成因,从本质上揭示两种动态扩展方法的原理,按照分类综述动态PLS模型的发展现状;最后,指出该领域亟需解决的问题和未来研究方向.
文摘针对具有动态特性的质量相关的故障检测问题,提出了一种基于自回归移动平均模型(auto-regressive moving average exogenous,ARMAX)的动态全主成分回归(dynamic total principal component regression,DT-PCR)方法.该方法基于输入的时滞值,形成增广输入矩阵;然后将形成的增广矩阵分解成质量无关和质量相关两个正交部分,并根据这两个部分对应子空间的统计量设计出一个更简单的故障诊断策略.该方法对于输出的预测精度也优于以往的方法.最后,通过一个数值例子和田纳西—伊斯曼过程将DT-PCR与全潜结构投影模型(total partial least squares,TPLS)进行对比,验证了DT-PCR的输出预测性能及与质量相关的故障检测性能.
基金supported by the National Natural Science Foundation of China(62273354,61673387,61833016).
文摘As a dynamic projection to latent structures(PLS)method with a good output prediction ability,dynamic inner PLS(DiPLS)is widely used in the prediction of key performance indi-cators.However,due to the oblique decomposition of the input space by DiPLS,there are false alarms in the actual industrial process during fault detection.To address the above problems,a dynamic modeling method based on autoregressive-dynamic inner total PLS(AR-DiTPLS)is proposed.The method first uses the regression relation matrix to decompose the input space orthogonally,which reduces useless information for the predic-tion output in the quality-related dynamic subspace.Then,a vector autoregressive model(VAR)is constructed for the predic-tion score to separate dynamic information and static informa-tion.Based on the VAR model,appropriate statistical indicators are further constructed for online monitoring,which reduces the occurrence of false alarms.The effectiveness of the method is verified by a Tennessee-Eastman industrial simulation process and a three-phase flow system.
文摘Tomato fruits (Solanum lycopersicum, L.) were exposed to origanum oil, vinegar, or chlorine at 12℃ and 95% RH. Quality-related attributes were examined during or following vapour treatment. Fruit pre-treated with natural volatiles and transferred to ambient air had less decay compared to control fruits. Vinegar-enrichment maintained fruit firmness and acidity during exposure, though the effect was sustained only for acidity following storage in ambient air. Glucose and fructose levels were higher in origanum oil-treated fruit, while ascorbic acid content increased in volatile-treated fruits following transfer/storage in ambient air. Total phenolics declined during vinegar and chlorine exposure but tended to increase following transfer to ambient air. Vinegar and chlorine increased lycopene concentration during exposure to vapours, and this effect persisted only for vinegar following exposure and transfer in ambient air. Origanum oil-treated fruit increased lycopene concentration following exposure and transfer in ambient air. Vapour- treatment did not affect fruit weight and ethylene production. Natural volatiles may maintain fruit quality in addition to the well documented antimicrobial protection during fresh produce storage and transit.
文摘在工业生产中,对系统进行故障检测具有十分重要的作用.改进的偏最小二乘(modified partial least squares,MPLS)是在PLS基础上提出的一种扩展算法,在质量相关故障检测中具有良好的检测效果,但当测试数据中含有质量无关故障时,MPLS算法漏报率较高.另外,MPLS算法的阈值为固定值会导致其误报率增加,这些问题会对工业过程监控产生较大影响.鉴于此,提出一种基于局部信息增量与MPLS的质量相关故障检测方法(local information increment-MPLS,LII-MPLS).在MPLS基础上,通过使用局部信息增量技术对测试数据进行实时更新检测后,质量相关故障的漏报率明显降低.同时,过程复杂化导致静态控制限不能满足故障检测的需求,现存的动态控制限适用范围具有一定局限性,因此改进静态控制限将其推广为局部动态阈值.最后,通过田纳西伊士曼过程(Tennessee Eastman process,TEP)仿真实验验证了所提出算法的有效性.
