Determination of Water Content in Watered Anhydrous Coolant Liquid and Brake Fluid of Automobile Using Fourier Transform Near-Infrared Spectroscopy

被引:2
作者
Zhang Yu [1 ]
Tan Li-hong [1 ]
He Yong [2 ,3 ]
机构
[1] Zhejiang Tech Inst Econ, Hangzhou 310018, Zhejiang, Peoples R China
[2] Zhejiang Univ, Coll Biosyst Engn & Food Sci, Hangzhou 310058, Zhejiang, Peoples R China
[3] Minist Agr, Key Lab Spect Sensing, Hangzhou 310058, Zhejiang, Peoples R China
关键词
Near-infrared transmittance spectroscopy; Water content; Anhydrous coolant liquid; Brake fluid; Brands; LEAST-SQUARES REGRESSION;
D O I
10.3964/j.issn.1000-0593(2019)07-2128-08
中图分类号
O433 [光谱学];
学科分类号
0703 ; 070302 ;
摘要
Coolant liquid and brake fluid are important vehicle oils, and they are of great significance for the normal operation of vehicles. Adding water into the coolant liquid and brake fluid is the common adulteration method for coolant liquid and brake fluid. The active ingredients in adulterated coolant liquid and brake fluid will reduce, and function of coolant liquid and brake fluid will be influenced. This will result in the harm of vehicles, which will influence the normal operation of vehicles. Rapid and accurate detection of water content in coolant liquid and brake fluid is of importance for quality assurance. In this paper, Fourier transform near-infrared spectroscopy was used to determine water content in different brands of watered anhydrous coolant liquid and brake fluid. Three brands of anhydrous coolant liquid, and four brands of brake fluid were used. In addition, samples with water contents of 0%, 5% , 10%, 15%, 20% , 25%, 30% and 35% were prepared. Fourier transform near-infrared transmittance spectra of the samples were acquired, and the spectral range of 10 067 similar to 5 442 cm(-1) were used for analysis. There were differences on near-infrared transmittance spectra among samples with different water contents. Based on each brand of anhydrous coolant liquid and brake fluid, principal component analysis (PCA) indicated the obvious differences among samples with different water contents. Besides, second derivative spectra were used to select optimal wavenumbers for each brand of anhydrous coolant liquid and brake fluid, as well as the combination of all brands. The selected optimal wavenumbers were similar among different brands of anhydrous coolant liquid as well as the combination of brands, and the selected optimal wavenumbers were similar among different brands of brake fluid as well as the combination of brands. The number of wavenumbers reduced at least 98. 67% after selection. Based on the full spectra and the selected optimal wavenumbers, partial least squares (PLS) and least-squares support vector machine (LS-SVM) were built. All the models obtained quite good performances, with determination of coefficient (R-2) over 0. 9 and residual prediction deviation (RPD) over 3. These prediction models obtained good performances. The performances of models for single brands were similar to those for the combination of brands, indicating that it was feasible to build calibration models using the combination of brands which would benefit the practical application. The overall results indicated the feasibility of using Fourier transform near-infrared transmittance spectroscopy combined with chemometric methods could be used to determine water adulteration in different brands of anhydrous coolant liquid and brake fluid. The results would help to develop on-line detection systems for water adulteration in different brands of anhydrous coolant liquid and brake fluid, and would provide guidance for detecting water content other fluids for automobile.
引用
收藏
页码:2128 / 2135
页数:8
相关论文
共 13 条
  • [1] Near-infrared (NIR) spectroscopy for motor oil classification: From discriminant analysis to support vector machines
    Balabin, Roman M.
    Safieva, Ravilya Z.
    Lomakina, Ekaterina I.
    [J]. MICROCHEMICAL JOURNAL, 2011, 98 (01) : 121 - 128
  • [2] Measurement of Soluble Solid Contents and pH of White Vinegars Using VIS/NIR Spectroscopy and Least Squares Support Vector Machine
    Bao, Yidan
    Liu, Fei
    Kong, Wenwen
    Sun, Da-Wen
    He, Yong
    Qiu, Zhengjun
    [J]. FOOD AND BIOPROCESS TECHNOLOGY, 2014, 7 (01) : 54 - 61
  • [3] Partial Least Squares Regression (PLSR) Applied to NIR and HSI Spectral Data Modeling to Predict Chemical Properties of Fish Muscle
    Cheng, Jun-Hu
    Sun, Da-Wen
    [J]. FOOD ENGINEERING REVIEWS, 2017, 9 (01) : 36 - 49
  • [4] Comparison of partial least squares regression, least squares support vector machines, and Gaussian process regression for a near infrared calibration
    Cui, Chenhao
    Fearn, Tom
    [J]. JOURNAL OF NEAR INFRARED SPECTROSCOPY, 2017, 25 (01) : 5 - 14
  • [5] [胡碧霞 Hu Bixia], 2017, [南京理工大学学报. 自然科学版, Journal of Nanjing University of Science and Technology], V41, P660
  • [6] [蒋璐璐 Jiang Lulu], 2011, [石油学报. 石油加工, Acta Petrolei Sinica. Petroleum Processing Section], V27, P112
  • [7] Determination of lubricant base oil properties by near infrared spectroscopy using different sample and variable selection methods
    Lima, FSG
    Araújo, MAS
    Borges, LEP
    [J]. JOURNAL OF NEAR INFRARED SPECTROSCOPY, 2004, 12 (03) : 159 - 166
  • [8] Detection of fungal infections in strawberry fruit by VNIR/SWIR hyperspectral imaging
    Siedliska, Anna
    Baranowski, Piotr
    Zubik, Monika
    Mazurek, Wojciech
    Sosnowska, Bozena
    [J]. POSTHARVEST BIOLOGY AND TECHNOLOGY, 2018, 139 : 115 - 126
  • [9] Application of Near-Infrared Hyperspectral Imaging with Variable Selection Methods to Determine and Visualize Caffeine Content of Coffee Beans
    Zhang, Chu
    Jiang, Hao
    Liu, Fei
    He, Yong
    [J]. FOOD AND BIOPROCESS TECHNOLOGY, 2017, 10 (01) : 213 - 221
  • [10] Application of Visible and Near-Infrared Hyperspectral Imaging to Determine Soluble Protein Content in Oilseed Rape Leaves
    Zhang, Chu
    Liu, Fei
    Kong, Wenwen
    He, Yong
    [J]. SENSORS, 2015, 15 (07): : 16576 - 16588