Online recognition of drainage type based on UV-vis spectra and derivative neural network algorithm

Highlights

• UV-vis absorption analyzer was applied in drainage type online recognition.

• The UV-vis spectrum of four drainage types were collected and evaluated.

• A convolutional neural network with multiple derivative inputs was established.

• Effects of different network structures and input contents were compared.

Abstract

Optimizing sewage collection is important for water pollution control and wastewater treatment plants quality and efficiency improvement. Currently, the urban drainage pipeline network is upgrading to improve its classification and collection ability. However, there is a lack of efficient online monitoring and identification technology. UV-visible absorption spectrum probe is considered as a potential monitoring method due to its small size, reagent-free and fast detection. Because the performance parameters of probe like optic resolution, dynamic interval and signal-to-noise ratio are weak and high turbidity of sewage raises the noise level, it is necessary to extract shape features from the turbidity disturbed drainage spectrum for classification purposes. In this study, drainage network samples were online collected and tested, and four types were labeled according to sample sites and environment situation. Derivative spectrum were adopted to amplify the shape features, while convolutional neural network algorithm was established to conduct nonlinear spectrum classification. Influence of input and network structure on classification accuracy was compared. Original spectrum, first-order derivative spectrum and a combination of both were set to be three different inputs. Artificial neural network with or without convolutional layer were set be two different network structures. The results revealed a convolutional neural network combined with inputs of first and zero-order derivatives was proposed to have the best classification effect on domestic sewage, mixed rainwater, rainwater and industrial sewage. The recognition rate of industrial wastewater was 100%, and the recognition rate of domestic sewage and rainwater mixing system were over 90%.

Graphical abstract

Keywords

Drainage online recognition / UV-vis spectra / Derivative spectrum / Convolutional neural network

References

[1] Asheri-Arnon T, Ezra S, Fishbain B (2018). Contamination detection of water with varying routine backgrounds by UV-spectrophotometry. Journal of Water Resources Planning and Management, 144(9): 04018056-1

[2] Benvidi A, Abbasi S, Gharaghani S, Dehghan Tezerjani M, Masoum S (2017). Spectrophotometric determination of synthetic colorants using PSO-GA-ANN. Food Chemistry, 220: 377–384

[3] Broeke J V D, Langergraber G, Weingartner A (2006). On-line and in-situ UV/vis spectroscopy for multi-parameter measurements: a brief review. Spectroscopy Europe, 18(4): S3–S4

[4] Cantwell R E, Hofmann R (2011). Ultraviolet absorption properties of suspended particulate matter in untreated surface waters. Water Research, 45(3): 1322–1328

[5] Cao Y S, Tang J G, Henze M, Yang X P, Gan Y P, Li J, Kroiss H, Van Loosdrecht M C M, Zhang Y, Daigger G T (2019). The leakage of sewer systems and the impact on the ‘black and odorous water bodies’ and WWTPs in China. Water Science and Technology, 79(2): 334–341

[6] Carré E, Perot J, Jauzein V, Lin L, Lopez-Ferber M (2017). Estimation of water quality by UV/Vis spectrometry in the framework of treated wastewater reuse. Water Science and Technology, 76(3): 633–641

[7] De Luca M, Oliverio F, Ioele G, Ragno G (2009). Multivariate calibration techniques applied to derivative spectroscopy data for the analysis of pharmaceutical mixtures. Chemometrics and Intelligent Laboratory Systems, 96(1): 14–21

[8] Dong J L, Hong M J, Xu Y, Zheng X Q (2019). A practical convolutional neural network model for discriminating Raman spectra of human and animal blood. Journal of Chemometrics, 33(11): e3184

[9] Gu J, Wang Z, Kuen J, Ma L, Shahroudy A, Shuai B, Liu T, Wang X, Wang G (2015). Recent advances in convolutional neural networks. Computer Science, 77: 354–377

[10] Hu Y, Wen Y, Wang X (2016). Novel method of turbidity compensation for chemical oxygen demand measurements by using UV–vis spectrometry. Sensors and Actuators. B, Chemical, 227: 393–398

[11] Hu Y T, Wang X P (2017). Application of surrogate parameters in characteristic UV-vis absorption bands for rapid analysis of water contaminants. Sensors and Actuators. B, Chemical, 239: 718–726

[12] Ichimura K (2017). The reproducibility and reliability of UV–vis higher-order derivative spectroscopy for quantitative analysis of spectral changes. Bulletin of the Chemical Society of Japan, 90(4): 411–418

[13] Jacquemin D, Planchat A, Adamo C, Mennucci B (2012). TD-DFT assessment of functionals for optical 0–0 transitions in solvated dyes. Journal of Chemical Theory and Computation, 8(7): 2359–2372

[14] Langergraber G, Fleischmann N, Hofstaedter F, Weingartner A (2004). Monitoring of a paper mill wastewater treatment plant using UV/VIS spectroscopy. Water Science & Technology A Journal of the International Association on Water Pollution Research, 49(1): 9–14

[15] Lepot M, Torres A, Hofer T, Caradot N, Gruber G, Aubin J B, Bertrand-Krajewski J L (2016). Calibration of UV/Vis spectrophotometers: A review and comparison of different methods to estimate TSS and total and dissolved COD concentrations in sewers, WWTPs and rivers. Water Research, 101: 519–534

[16] Liu J, Osadchy M, Ashton L, Foster M, Solomon C J, Gibson S J (2017). Deep convolutional neural networks for Raman spectrum recognition: a unified solution. Analyst (London), 142(21): 4067–4074

[17] Liu Y, Hou L G, Bian W, Zhou B L, Liang D B, Li J (2020). Turbidity in combined sewer sewage: an identification of stormwater detention tanks. International Journal of Environmental Research and Public Health, 17(9): 3053–3062

[18] Liudmil A (1997). Drawbacks of the present standards for processing absorption spectra recorded linearly as a function of wavelength. Trends in Analytical Chemistry, 16(9): 536–543

[19] Ma J J, Meng F S, Zhou Y X, Wang Y Y, Shi P (2018). Distributed water pollution source localization with mobile UV-visible spectrometer probes in wireless sensor networks. Sensors (Basel), 18(2): 606

[20] Makantasis K, Karantzalos K, Doulamis A, Doulamis N (2015). Deep supervised learning for hyperspectral data classification through convolutional neural networks. In: Proceedings of the IEEE Geoscience & Remote Sensing Symposium. Milan: IGARSS IEEE, 4959–4962

[21] Meriç S, Selçuk H, Belgiorno V (2005). Acute toxicity removal in textile finishing wastewater by Fenton’s oxidation, ozone and coagulation–flocculation processes. Water Research, 39(6): 1147–1153

[22] Monakhova Y B, Kolesnikova S S, Mushtakova S P (2013). Independent component analysis algorithms for spectral decomposition in UV/VIS analysis of metal-containing mixtures including multimineral food supplements and platinum concentrates. Analytical Methods, 5(11): 2761–2772

[23] Palacios-Morillo A, Alcazar A, De Pablos F, Jurado J M (2013). Differentiation of tea varieties using UV-Vis spectra and pattern recognition techniques. Spectrochimica Acta. Part A: Molecular and Biomolecular Spectroscopy, 103: 79–83

[24] Parmar A, Sharma S (2016). Derivative UV-vis absorption spectra as an invigorated spectrophotometric method for spectral resolution and quantitative analysis: Theoretical aspects and analytical applications: A review. Trends in Analytical Chemistry, 77: 44–53

[25] Qin X, Gao F, Chen G (2012). Wastewater quality monitoring system using sensor fusion and machine learning techniques. Water Research, 46(4): 1133–1144

[26] Şahin S, Demir C, Güçer Ş (2007). Simultaneous UV–vis spectrophotometric determination of disperse dyes in textile wastewater by partial least squares and principal component regression. Dyes and Pigments, 73(3): 368–376

[27] Sarraguça M C, Paulo A, Alves M M, Dias A M, Lopes J A, Ferreira E C (2009). Quantitative monitoring of an activated sludge reactor using on-line UV-visible and near-infrared spectroscopy. Analytical and Bioanalytical Chemistry, 395(4): 1159–1166

[28] Takahashi M B, Leme J, Caricati C P, Tonso A, Fernandez Nunez E G, Rocha J C (2015). Artificial neural network associated to UV/Vis spectroscopy for monitoring bioreactions in biopharmaceutical processes. Bioprocess and Biosystems Engineering, 38(6): 1045–1054

[29] Thomas O, Brogat M (2017). UV-Visible Spectrophotometry of Water and Wastewater. Rennes: Elsevier, 73–138

[30] Tomazzoli M M, Pai Neto R D, Moresco R, Westphal L, Zeggio A R, Specht L, Costa C, Rocha M, Maraschin M (2015). Discrimination of Brazilian propolis according to the seasoning using chemometrics and machine learning based on UV-Vis scanning data. Journal of Integrative Bioinformatics, 12(4): 15–26

[31] Vieira dos Santos A, Masini J C (2007). Evaluating the removal of Cd(II), Pb(II) and Cu(II) from a wastewater sample of a coating industry by adsorption onto vermiculite. Applied Clay Science, 37(1–2): 167–174

[32] Wang S, Wang B, Cao X, Jin W, Jia L (2007). Discussion on Chinese urban drainage system. China Water & Wastewater, 12: 16–21 (in Chinese)

[33] Ye Z, Gilman A, Peng Q, Levick K, Cosman P, Milstein L (2019). Comparison of Neural Network Architectures for Spectrum Sensing. Sanya: IEEE

[34] Yin X (2014). Discussion on pollution and control measures of rainwater system in rainwater and sewage separation drainage system. China Water & Wastewater, 30: 24–26 (in Chinese)

Acknowledgements

This work was supported by the National Key R&D Program of China (No. 2019YFD1100505), and the program of China Scholarship Council (No. 201806210101).

RIGHTS & PERMISSIONS

2021 Higher Education Press