Assessment of the Effect of Wastewater Quantity and Quality, and Sludge Parameters on Predictive Abilities of Non-Linear Models for Activated Sludge Settleability Predictions

被引:11
作者
Szelag, Bartosz [1 ]
Gawdzik, Jaroslaw [1 ]
机构
[1] Kielce Univ Technol, Fac Environm Geomat & Energy Engn, Tysiaclecia Panstwa Polskiego 7, PL-25314 Kielce, Poland
来源
POLISH JOURNAL OF ENVIRONMENTAL STUDIES | 2017年 / 26卷 / 01期
关键词
activated sludge settleability; SVM; k-NN; MARS methods; ADAPTIVE REGRESSION SPLINES; MARS;
D O I
10.15244/pjoes/64810
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
This paper discusses the possibility of applying the three "black-box" methods to sludge settleability predictions. Additionally, the impact of the load of biogenic compounds and parameters of activated sludge on predictive abilities of the devised mathematical models is analysed in the paper. To conduct analyses we relied on the results of measurements of wastewater quantity and quality, and of the bioreactor operational parameters, taken on continuous basis at the Sitkowka-Nowiny treatment plant in 2012-16. The analyses conducted for the study indicate that the lowest values of errors in activated sludge settleability predictions for the wastewater treatment plant of concern were obtained for input data on the load of biogenic compounds at the inflow, microorganism culture environment, and activated sludge concentration.
引用
收藏
页码:315 / 321
页数:7
相关论文
共 17 条
[1]  
BURGES C., 2000, TUTORIAL SUPPORT VEC, V1
[2]   A COMPARISON OF 2 NONPARAMETRIC-ESTIMATION SCHEMES - MARS AND NEURAL NETWORKS [J].
DEVEAUX, RD ;
PSICHOGIOS, DC ;
UNGAR, LH .
COMPUTERS & CHEMICAL ENGINEERING, 1993, 17 (08) :819-837
[3]   MULTIVARIATE ADAPTIVE REGRESSION SPLINES [J].
FRIEDMAN, JH .
ANNALS OF STATISTICS, 1991, 19 (01) :1-67
[4]   Comparison and evaluation of empirical zone settling velocity parameters based on sludge volume index using a unified settling characteristics database [J].
Giokas, DL ;
Daigger, GT ;
von Sperling, M ;
Kim, Y ;
Paraskevas, PA .
WATER RESEARCH, 2003, 37 (16) :3821-3836
[5]   Using and comparing two nonparametric methods (CART and MARS) to model the potential distribution of gullies [J].
Gomez Gutierrez, Alvaro ;
Schnabel, Susanne ;
Lavado Contador, J. Francisco .
ECOLOGICAL MODELLING, 2009, 220 (24) :3630-3637
[6]   A soft computing method to predict sludge volume index based on a recurrent self-organizing neural network [J].
Han, Hong-Gui ;
Li, Ying ;
Guo, Ya-Nan ;
Qiao, Jun-Fei .
APPLIED SOFT COMPUTING, 2016, 38 :477-486
[7]   Maximising the usefulness of flood risk assessment for the River Vistula in Warsaw [J].
Kiczko, A. ;
Romanowicz, R. J. ;
Osuch, M. ;
Karamuz, E. .
NATURAL HAZARDS AND EARTH SYSTEM SCIENCES, 2013, 13 (12) :3443-3455
[8]   Modeling and analysis of pumps in a wastewater treatment plant: A data-mining approach [J].
Kusiak, Andrew ;
Zeng, Yaohui ;
Zhang, Zijun .
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2013, 26 (07) :1643-1651
[9]  
LOU I, 2012, MATH PROBL ENG, V2012, P1, DOI DOI 10.1155/2012/237693
[10]   Flash-flood forecasting by means of neural networks and nearest neighbour approach -: a comparative study [J].
Piotrowski, A. ;
Napiorkowski, J. J. ;
Rowinski, P. M. .
NONLINEAR PROCESSES IN GEOPHYSICS, 2006, 13 (04) :443-448