基于紫外-可见光谱与深度学习CNN算法的水质COD预测模型研究

doi:10.3964/j.issn.1000-0593(2020)09-2981-08

Abstract
References
Related Citation (15)

Download: PDF (3276 KB)
Export: BibTeX | EndNote (RIS)

Abstract Water is vital for human life, and the quality of water is directly related to people’s quality of life. At present, research into chemical oxygen demand (COD) methods for determining water quality is mainly focused on spectral data preprocessing and spectral feature extraction, with few studies considering spectral data modeling methods. Convolutional neural networks (CNN) are known to have strong feature extraction and feature mapping abilities. Thus, in this study, a CNN is combined with UV-visible spectroscopy to establish a COD prediction model. The Savitzky-Golay smoothing filter is applied to remove noise interference, and the spectral data are then input to the CNN model. The features of the spectrum data are extracted through the convolution layer, the spatial dimensions are reduced in the pooling layer, and the global features are mapped in the fully connected layer. The model is trained using the ReLU activation function and the Adam optimizer. A series of experiments show that the CNN model has a strong ability to predict COD in water, with a high prediction accuracy and good fit to the regression curve. A comparison with other models indicates that the proposed CNN model gives the smallest RMSEP and MAE, the largest -R², and the best fitting effect. It is found that the model has strong generalization ability through the evaluation effect of the training samples. To counter the inaccuracy of the predicted results caused by the peak shift of the absorption spectrum, a regression model based on a strengthened CNN (CNNs) is also developed. After denoising, the spectral data can be divided into three categories according to the different characteristics of absorption peaks, and the corresponding CNN regression model is input respectively for prediction. When the corresponding regression model is applied, the experimental results show that the sectional CNNs model outperforms our original CNN model in terms of fitting, prediction precision, determination coefficient, and error. Not only does R² increase significantly, reaching 0.999 1, but also the MAE and RMSEP of the test samples also reduced to 2.314 3and 3.874 5, respectively, which were reduced by 25.9% and 21.33% compared with out original CNN. Performance testing of the prediction model, indicates that the detection limit is 0.28 mg·L^-1and the measurement range is 2.8~500 mg·L^-1. This paper describes an innovative combination of a CNN with spectral analysis and reports our pioneering ideas on the application of spectral analysis in the field of water quality detection.

Key words：Ultraviolet visible spectrum; Convolution neural network; Chemical oxygen demand; Prediction model

Received: 2019-02-20 Accepted: 2020-01-01

ZTFLH:

O657.32

Corresponding Authors: LIU Xin E-mail: liuxin_CN@qq.com

Cite this article:

URL:

https://www.gpxygpfx.com/EN/10.3964/j.issn.1000-0593(2020)09-2981-08 OR https://www.gpxygpfx.com/EN/Y2020/V40/I09/2981

[1] Skuras D，Tyllianakis E. Water Research，2018, 143：198.
[2] Li J，Luo G B，He L J，et al. Critical Reviews in Analytical Chemistry，2017, 48：47.
[3] Liu F，Zheng P C，Huang B C，et al. IFIP Advances in Information and Communication Technology，2017, 478：619.
[4] Alves F C G B S，Coqueiro, A，Marco, P H，et al. Food Chemistry 2018, 273: 124.
[5] Yao N，Liu Z A，Chen Y，et al. Sensors，2015, 15：20501.
[6] Chen J，Liu S，Qi X，et al. Sensors & Actuators B Chemical, 2018, 254: 778.
[7] Guan L，Tong Y F，Li J W，et al. Optik, 2018, 164: 277.
[8] Mason J D，Cone M T，Fry E S. Applied Optics，2016, 55：7163.
[9] He L，Kaoru O，Dong M X. IEEE Network，2018, 32：96.
[10] Kalinin A A，Higgins G A. Pharmacogenomics，2018, 19：629.
[11] Gholizadeh M H，Melesse A M，Reddi L. Sensors，2016, 16：1298.
[12] Al-Saffar A A M，Tao H, Talab M A. International Conference on Radar，2018.
[13] Schettino B M，Duque C A，Silveira P M. IEEE Transactions on Power Delivery，2016, 31：1400.
[14] Saing V，Vorasayan P, Suwanwela N C. Iet Image Processing，2018, 12：105.
[15] Liakos K G，Busato P，Moshou D，et al. Sensors，2018, 18：2674.
[16] Zhang M F. Modern Physics Letters B，2017, 31.
[17] He F，Zhang L Y. Journal of Process Control，2018, 66：51.
[18] Castro W，Prieto J M，Guerra R，et al. Journal of Food Engineering，2018, 238：95.
[19] Fernandez-Espinosa A J. Talanta，2016, 148：216.
[20] Khan S，Ullah R，Khan A. Applied Spectroscopy，2017, 71：2111.
[21] Alessio S M. Signals & Communication Technology，2016. 1.
[22] Alberto S G，Dora B H，Francisco A. Journal of Supercomputing，2018, 3：1.
[23] Chen L，Hu X M，Xu T，et al. IEEE Transactions on Intelligent Transportation Systems，2017, 18：3303.
[24] Ghorbani M A，Zadeh H A，Isazadeh M，et al. Environmental Earth Sciences 2016, 75: 476.
[25] Feng W W，Li D，Cai Z Q，et al. International Symposium on Advanced Optical Manufacturing & Testing Technologies: Optical Test，2016, 9684.
[26] Li A，Li Y X，Li X H. Tensor Flow and Keras-based Convolutional Neural Network in CAT Image Recognition. 2017 2nd International Conference on Computational Modeling, Simulation and Applied Mathematics (CMSAM)，2017. 529.
[27] Salehi N，Monsefi R，Yazdi H S. Knowledge-Based Systems，2018, 151：62.
[28] Zhao H Z，Liu F X，Li L Y，et al. Applied Intelligence，2017, 48：1707.
[29] Bako S，Vogels T，Mcwilliams B，et al. ACM Transactions on Graphics，2017, 36：1.
[30] Kingma D P，Ba J. Adam: A Method for Stochastic Optimization. Computer Science，2014.
[31] Chen B S，Wu H A，Li S F Y. Talanta，2014, 120：325.
[32] de Myttenaere A，Golden B，Le Grand B，et al. Neurocomputing，2016, 192：38.
[33] Khair U，Fahmi H，Al Hakim S，et al. Forecasting Error Calculation with Mean Absolute Deviation and Mean Absolute Percentage Error. International Conference on Information and Communication Technology (IconICT)，2017. 930.
[34] Wang X Z，Wang R，Xu C. IEEE Transactions on Cybernetics，2017, 48：703.
[35] Zhao H, Liu F, Li L, et al. Applied Intelligence, 2018, 48(7): 1707.
[36] LIU Fei, DONG Da-ming, ZHAO Xian-de, et al. Spectroscopy and Spectral Analysis，2017, 37(9)：2724.
[37] Jiang M Y，Liang Y C，Feng X Y, et al. Neural Computing & Applications, 2016, 29: 61.