基于近红外光谱分析的原油脱盐过程故障检测
|
摘要
为了解决原油脱盐过程的故障检测问题,提出一种利用近红外光谱技术从微观分子光谱数据角度进行故障检测的新思路。不同于传统基于宏观过程变量的故障检测手段,该方法从分子振动信号中获取过程信息从而完成对原油脱盐过程运行状态的监控。为了同时监控脱盐原油的质量,采用偏最小二乘(PLS)算法,选择Hotelling T2和平方预测误差(SPE)统计量作为判断指标,比较了基于宏观过程变量和微观分子光谱数据的故障检测方法的效果。结果表明,基于微观分子光谱数据的检测方法在时间上比基于传统宏观过程变量的方法快了约46分钟,并且对早期故障的感知更加灵敏,充分体现了近红外光谱技术在原油脱盐过程故障检测问题上的有效性和优势。
To solve the fault detection problem of crude oil desalting process, a method to detect the faults of crude oil desalting process from micro-molecular spectral data by using Near Infrared Spectroscopy(NIRS) was proposed. Different from the traditional fault detection methods based on macro-process variables, in this method, the process information was obtained from molecular vibration signals to monitor the crude oil desalting process. To monitor the quality of desalted oil at the same time, Partial Least Squares(PLS) algorithm was adopted and Hotelling T2 and Squared Prediction Error(SPE) were selected as indicators.. The results show that the fault detection based on micro-molecular spectral data is about 46 minutes faster than that based on the traditional macro-process variables, and is more sensitive to early faults, fully demonstrateing the effectiveness and superiority of NIRS for the fault detection of crude oil desalting process.
To solve the fault detection problem of crude oil desalting process, a method to detect the faults of crude oil desalting process from micro-molecular spectral data by using Near Infrared Spectroscopy(NIRS) was proposed. Different from the traditional fault detection methods based on macro-process variables, in this method, the process information was obtained from molecular vibration signals to monitor the crude oil desalting process. To monitor the quality of desalted oil at the same time, Partial Least Squares(PLS) algorithm was adopted and Hotelling T2 and Squared Prediction Error(SPE) were selected as indicators.. The results show that the fault detection based on micro-molecular spectral data is about 46 minutes faster than that based on the traditional macro-process variables, and is more sensitive to early faults, fully demonstrateing the effectiveness and superiority of NIRS for the fault detection of crude oil desalting process.
引文
[1] PARVASI P,HESAMEDINI A K,JAHANMIRI A,et al.A novel modeling and experimental study of crude oil desalting using microwave[J].Separation Science & Technology,2014,49(7):1029-1044.
[2] DADARI S,RAHIMI M,ZINADINI S.Crude oil desalter effluent treatment using high flux synthetic nanocomposite NF membrane-optimization by response surface methodology[J].Desalination,2016,377:34-46.
[3] AL-OTAIBI M,ELKAMEL A,AL-SAHHAF T,et al.Experimental investigation of crude oil desalting and dehydration[J].Chemical Engineering Communications,2003,190(1):65-82.
[4] ARYAFARD E,FARSI M,RAHIMPOUR M R.Modeling and simulation of crude oil desalting in an industrial plant considering mixing valve and electrostatic drum[J].Chemical Engineering & Processing Process Intensification,2015,95:383-389.
[5] YIN S,DING S X,XIE X,et al.A review on basic data-driven approaches for industrial process monitoring[J].IEEE Transactions on Industrial Electronics,2014,61(11):6418-6428.
[6] BOTRE C,MANSOURI M,NOUNOU M,et al.Kernel PLS-based GLRT method for fault detection of chemical processes[J].Journal of Loss Prevention in the Process Industries,2016,43:212-224.
[7] VILLEZ K,SRINIVASAN B,RENGASWAMY R,et al.Kalman-based strategies for Fault Detection and Identification (FDI):Extensions and critical evaluation for a buffer tank system[J].Computers & Chemical Engineering,2011,35(5):806-816.
[8] ALAM M A,SHI Z,DRENNEN J K,et al.In-line monitoring and optimization of powder flow in a simulated continuous process using transmission near infrared spectroscopy[J].International Journal of Pharmaceutics,2017,526(1/2):199-208.
[9] SANTOS J R,VIEGAS O,PáSCOA R N,et al.In-line monitoring of the coffee roasting process with near infrared spectroscopy:Measurement of sucrose and colour[J].Food Chemistry,2016,208:103-110.
[10] NAWAR S,BUDDENBAUM H,HILL J,et al.Estimating the soil clay content and organic matter by means of different calibration methods of vis-NIR diffuse reflectance spectroscopy[J].Soil & Tillage Research,2016,155:510-522.
[11] STANIMIROVA I.Practical approaches to principal component analysis for simultaneously dealing with missing and censored elements in chemical data[J].Analytica Chimica Acta,2013,796(17):27-37.
[12] FREITAS M R,BARIGYE S J,DARé J K,et al.Aug-MIA-SPR/PLS-DA classification of carbonyl herbicides according to levels of soil sorption[J].Geoderma,2016,268:1-6.
[13] ESQUERRE C,GOWEN A A,Burger J,et al.Suppressing sample morphology effects in near infrared spectral imaging using chemometric data pre-treatments[J].Chemometrics & Intelligent Laboratory Systems,2012,117(6):129-137.
[14] MOES J J,RUIJKEN M M,GOUT E,et al.Application of process analytical technology in tablet process development using NIR spectroscopy:blend uniformity,content uniformity and coating thickness measurements[J].International Journal of Pharmaceutics,2008,357(1/2):108-118.
[15] MERCIER S,MONDOR M,MARCOS B,et al.Estimation of missing values in a food property database by matrix completion using PCA-based approaches[J].Chemometrics & Intelligent Laboratory Systems,2017,166:37-48.
[2] DADARI S,RAHIMI M,ZINADINI S.Crude oil desalter effluent treatment using high flux synthetic nanocomposite NF membrane-optimization by response surface methodology[J].Desalination,2016,377:34-46.
[3] AL-OTAIBI M,ELKAMEL A,AL-SAHHAF T,et al.Experimental investigation of crude oil desalting and dehydration[J].Chemical Engineering Communications,2003,190(1):65-82.
[4] ARYAFARD E,FARSI M,RAHIMPOUR M R.Modeling and simulation of crude oil desalting in an industrial plant considering mixing valve and electrostatic drum[J].Chemical Engineering & Processing Process Intensification,2015,95:383-389.
[5] YIN S,DING S X,XIE X,et al.A review on basic data-driven approaches for industrial process monitoring[J].IEEE Transactions on Industrial Electronics,2014,61(11):6418-6428.
[6] BOTRE C,MANSOURI M,NOUNOU M,et al.Kernel PLS-based GLRT method for fault detection of chemical processes[J].Journal of Loss Prevention in the Process Industries,2016,43:212-224.
[7] VILLEZ K,SRINIVASAN B,RENGASWAMY R,et al.Kalman-based strategies for Fault Detection and Identification (FDI):Extensions and critical evaluation for a buffer tank system[J].Computers & Chemical Engineering,2011,35(5):806-816.
[8] ALAM M A,SHI Z,DRENNEN J K,et al.In-line monitoring and optimization of powder flow in a simulated continuous process using transmission near infrared spectroscopy[J].International Journal of Pharmaceutics,2017,526(1/2):199-208.
[9] SANTOS J R,VIEGAS O,PáSCOA R N,et al.In-line monitoring of the coffee roasting process with near infrared spectroscopy:Measurement of sucrose and colour[J].Food Chemistry,2016,208:103-110.
[10] NAWAR S,BUDDENBAUM H,HILL J,et al.Estimating the soil clay content and organic matter by means of different calibration methods of vis-NIR diffuse reflectance spectroscopy[J].Soil & Tillage Research,2016,155:510-522.
[11] STANIMIROVA I.Practical approaches to principal component analysis for simultaneously dealing with missing and censored elements in chemical data[J].Analytica Chimica Acta,2013,796(17):27-37.
[12] FREITAS M R,BARIGYE S J,DARé J K,et al.Aug-MIA-SPR/PLS-DA classification of carbonyl herbicides according to levels of soil sorption[J].Geoderma,2016,268:1-6.
[13] ESQUERRE C,GOWEN A A,Burger J,et al.Suppressing sample morphology effects in near infrared spectral imaging using chemometric data pre-treatments[J].Chemometrics & Intelligent Laboratory Systems,2012,117(6):129-137.
[14] MOES J J,RUIJKEN M M,GOUT E,et al.Application of process analytical technology in tablet process development using NIR spectroscopy:blend uniformity,content uniformity and coating thickness measurements[J].International Journal of Pharmaceutics,2008,357(1/2):108-118.
[15] MERCIER S,MONDOR M,MARCOS B,et al.Estimation of missing values in a food property database by matrix completion using PCA-based approaches[J].Chemometrics & Intelligent Laboratory Systems,2017,166:37-48.