Application of Selected Methods of Black Box for Modelling the Settleability Process in Wastewater Treatment Plant

Szeląg, Bartosz; Gawdzik, A.

doi:10.1515/eces-2017-0009

Cited by 8 publications

(9 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Of the two methods considered for modelling wastewater quality, lower forecast uncertainty of the probability of SVI lim exceedance was obtained using the SVM method than the k-NN method (Fig. 4), which is also indicated by the parameters of the fit of calculation results to measurements [3,20]. Fig.…”

Section: Variablementioning

confidence: 88%

“…It makes it possible to assess the reliability of the results of simulation of activated sludge sedimentation capacity using mathematical models in which the values of measured quality indicators are replaced with the results of calculations, with the parametric uncertainty of the model taken into account as well. This approach can be used to improve the operational efficiency of the bioreactor when continuous measurements of sewage quality are not available, but also takes into account the accuracy of the model's forecast of sludge sedimentation, which most previous studies [20][21][22] have failed to consider.…”

Section: Discussionmentioning

confidence: 99%

“…Based on this, simulators of their values were developed using the Monte Carlo method, omitting their correlation. In other cases, the correlation between variables should be taken into account [20]. Independent variables describing the quality of wastewater are forecasted using statistical models.…”

Section: Variablementioning

confidence: 99%

See 2 more Smart Citations

Effect of Parametric Uncertainty of Selected Classification Models and Simulations of Wastewater Quality Indicators on Predicting the Sludge Volume Index

Chmielowski¹,

Bedla²,

Dacewicz³

et al. 2020

Pol. J. Environ. Stud.

View full text Add to dashboard Cite

Commonly used models for predicting natural phenomena include classification models. These may be black box methods (neural networks, support vector machines, random forests, etc.) or models giving an explicit dependence enabling conclusive assessment of the influence of the explanatory variables on the result of calculations. However, because in this model the result is a specific discrete value, the result often proves unsatisfactory. In the case of classification models, the problem is of a different nature, because only allocation to the appropriate class is modelled, which in many cases makes it possible to identify technological parameters with a much smaller number of input variables than in

show abstract

Section: Variablementioning

confidence: 88%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Effect of Parametric Uncertainty of Selected Classification Models and Simulations of Wastewater Quality Indicators on Predicting the Sludge Volume Index

Chmielowski¹,

Bedla²,

Dacewicz³

et al. 2020

Pol. J. Environ. Stud.

View full text Add to dashboard Cite

show abstract

“…Therefore, it seems correct to use an alternative approach, based on the extensive quantity, which means that creating models for simulating the process parameter is justified. The literature survey indicates that the issue of predicting sludge settleability has so far been of interest only to [5,12], although the effect on the value of SE for filamentous microorganisms was not described in these reports.…”

mentioning

confidence: 99%

Using Data Mining to Predict Sludge and Filamentous Microorganism Sedimentation

Chmielowski¹,

Czekański²,

Leśniańska³

2019

Pol. J. Environ. Stud.

View full text Add to dashboard Cite

IntroductionConsidering the stochasticity of the volume and quality of the incoming sewage, the operating parameters of reactors need to be controlled within a certain range so that the performance of the respective units of a sewage treatment plant (STP) can be optimized and the desirable pollution reduction effect achieved [1,2]. This requires prediction of various biochemical processes taking place in the reactor, using physical or statistical method for the purpose. The former have the advantage that the qualitative variability of sewage at the discharge from the reactor and the biological reactor's operating parameters is based on differential equation systems. However, for the calibration of physical models, one needs detailed information about the reaction path in the respective units, which requires continuous high-resolution measurements of a number of qualitative parameters of the sewage at the inlet, discharge and inside the reactor, leading to considerable problems in the experimental phase. Moreover, due to the number of model parameters and strong interactions between them, their calibration may often be difficult, as confirmed in multiple Pol. AbstractThis study attempted to develop statistical regression models for predicting the settleability of activated sludge based on the quality of incoming sewage and on the identified dominant filamentous species. As part of the analyses conducted for the purpose, classification models are presented that enable identification of the respective filamentous microorganisms, based on the working parameters of the bioreactor and the quality of the influent. The study calculations demonstrated that the modeling methods based on artificial neural networks, random forests, and boost trees can be applied for the identification of filamentous microorganisms Microthrix parvicella, Nostocoida sp., and Thiotrix sp. in activated sludge chambers in the STP located in Sitkówka-Nowiny. The best predictive capacity, covering identification of the above-mentioned filamentous bacterial species in activated sludge chambers, was observed for statistical models obtained by the random forest method.

show abstract

“…Mean prediction error (ME), which is a measure of estimator bias, a mean value of differences between the measured and predicted values [18]. The value of this error should be as low as possible and should be expressed in the same units as the predicted values:…”

mentioning

confidence: 99%

Geostatistical Methods in Water Distribution Network Design - A Case Study

Potyralla

2019

Ecological Chemistry and Engineering S

View full text Add to dashboard Cite

Modeling of the loads of water supply networks and their subsequent forecasting is an element necessary for making optimum decisions in the process of planning the development and operation of the water supply networks. The results of this modeling are decisive for the selection of the diameters of the pipelines and their arrangement on the water demand area. This study presents the results of estimation of average values of loads for the selected investment variants. The aim of the article is to present the possibility of simulations and analyses of the geostatistical interpolation methods. Data input in the model regarded the fragment of the real water supply network administered by the Municipal Water and Sewerage Company in Warszawa. Results of the computer analyses for the presented investment variants were related to the operating data of the water supply network and the data on water demand for the years 2014-2017 and 2018-2025. The aim of this paper is to present the advantages of GIS for the water supply systems and to prove that using the appropriate IT system, with provision of proper data processing, may lead to decisions which are optimum in view of the established, often very complex criteria.

show abstract

Application of Selected Methods of Black Box for Modelling the Settleability Process in Wastewater Treatment Plant

Cited by 8 publications

References 21 publications

Effect of Parametric Uncertainty of Selected Classification Models and Simulations of Wastewater Quality Indicators on Predicting the Sludge Volume Index

Effect of Parametric Uncertainty of Selected Classification Models and Simulations of Wastewater Quality Indicators on Predicting the Sludge Volume Index

Using Data Mining to Predict Sludge and Filamentous Microorganism Sedimentation

Geostatistical Methods in Water Distribution Network Design - A Case Study

Contact Info

Product

Resources

About