Article 5416

Title of the article

ASSESSMENT OF DRINKING WATER QUALITY BY PHYSICAL
AND CHEMICAL PARAMETERS OF A WATER SOURCE 

Authors

Bubyr' Dmitriy Sergeevich, Postgraduate student, Ulyanovsk State Technical University (32 Severny Venets street, Ulyanovsk, Russia), bubir91@mail.ru
Bulyzhev Evgeniy Mikhaylovich, Doctor of engineering sciences, professor, sub-department of mechanical engineering, Ulyanovsk State Technical University (32 Severny Venets street, Ulyanovsk, Russia), ecovita05@mail.ru
Klyachkin Vladimir Nikolaevich, Doctor of engineering sciences, professor, sub-department of applied mathematics and informatics, Ulyanovsk State Technical University (32 Severny Venets street, Ulyanovsk, Russia), v_kl@mail.ru
Krasheninnikov Viktor Rostislavovich, Doctor of engineering sciences, professor, head of sub-department of applied mathematics and informatics, Ulyanovsk State Technical University (32 Severny Venets street, Ulyanovsk, Russia), kvrulstu@mail.ru

Index UDK

519.246.8:628.1

DOI

10.21685/2072-3059–2016-4-5

Abstract

Background. The object of the study is the water treatment system of the St Petersburg Vodokanal (water supply system), which is controlled by seven drinking water quality indicators depending on six physical-chemical parameters of water and two managed parameters (dose of a coagulant and a flocculant used in water purification). The subject of the research is the quality of drinking water. The aim of the study is to develop an early warning system for possible water quality loss due to deterioration of physico-chemical indicators of a water source (forecasting a possibility that one or more water quality indicators will be over the permissible limit) and to control activities (changing doses of a coagulant and a flocculant) to avoid emergency situations.
Materials and methods. The study of water quality was conducted by building regression models that best represent the dependence of drinking water quality indicators on physico-chemical parameters and controlled parameters, on the basis of monitoring of water treatment systems.
Results. The authors developed models for drinking water quality indicators that can be used to predict the future state of drinking water. 
Conclusions. As a result of regression modelling of drinking water quality indicators it has been found that global models do not provide the necessary accuracy of forecasting. In order to enhance the value of the determination coefficient it is suggested to use small samples applying piecewise-linear regressions taking into account autoregressions of the second order. To assess the effect of controlled agents (doses of a coagulant and a flocculant) on water quality indicators, the authors recommend to use piecewise quadratic models. The proposed approaches will allow to prediction the potential of emergency situations at which the selected indicators of drinking water quality may go over the permissible limit.

Key words

drinking water quality, water supply source state, regressive model, forecasting, piecewise linear regression, piecewise quadratic model

Download PDF
References

1. Valeev S. G., Bulyzhev E. M. Inzhenernyy zhurnal [Journall of engineering]. 2011, no. 10, pp. 39–42.
2. Valeev S. G. Regressionnoe modelirovanie pri obrabotke nablyudeniy [Regression modelling at processing of observations]. Moscow: Nauka, 1991, 272 p.
3. Klyachkin V. N. Statisticheskie metody v upravlenii kachestvom: komp'yuternye tekhnologii [Statistical methods in quality management: computer technologies]. Moscow: Finansy i statistika: INFRA-M, 2009, 304 p.
4. Khalafyan A. A. STATISTICA 6. Statisticheskiy analiz dannykh [STATISTICA 6. Statistical analysis of data]. 3rd ed. Moscow: Binom-Press, 2007, 512 p.
5. Statistica documentation. Available at: documentation.statsoft.com (accessed March 31,2014).
6. Joaquim P. Marques de Sá, Applied Statistics Using SPSS, STATISTICA, MATLAB and R. Berlin: Springer, 2007, p. 520.
7. Rasmussen C. E., Williams C. K. I. Gaussian Processes for Machine Learning. Massachusetts: The MIT Press, 2006, p. 248.
8. Seber G. A. F., Alan J. L. Linear Regression Analysis. 2nd edi. Wiley, 2003, p. 582.
9. Nathans L. L., Frederick L., Kim Nimon Practical assessment research & evaluation. 2012, vol. 17, no. 9. Available at: scholarship.rice.edu/ bitstream/handle/ 1911/71096/ 2012-Nathans- %20PARE-Regression Guidebook.pdf   (accessed April 30, 2015).
10. Casdagli M., Eubank S. (eds.) SFI Studies in the Sciences of Complexity, Proc. Addison- Wesley. 1992, vol. XII.
11. Palit A. K., Dobrivoje Popovic Computational Intelligence in Time Series Forecasting:Theory and En-gineering Applications (Advances in Industrial Control). London:Springer-Verlag London Limited, 2005, p. 372.
12. Box G. E. P., Jenkins G. M., Reinsel G. C. Time Series Analysis. Forecasting, and Control.3rd ed. Prentice-Hall, Englewood Cliffs. NJ, 1994, p. 406.
13. Krasheninnikov V. R., Bubyr' D. S. Mezhdistsiplinarnye issledovaniya v oblasti matematicheskogo modelirovaniya i informatiki: materialy 3-y nauch.-prakt. internet-konf. (20–21 fevralya 2014 g.) [Interdisciplinary research in the field of mathematical modelling and informatics: proceedings of III Scientific and practical online conference (20th-21st February 2014)]. Ulyanovsk: SIMJET, 2014, 233–236 p.й
14. Krasheninnikov V. R., Potapov M. A. Pattern Recognition and Image Analysis. 2011, vol. 21, no. 2, pp. 280–284.

 

Дата создания: 02.08.2017 14:56
Дата обновления: 03.08.2017 13:41