Study of energy consumption in a wastewater treatment plant using logistic regression

Abstract

The aim of this paper is to predict the energy consumption of a wastewater treatment plant from Romania, taking into account the flowrate, concentration of BOD, TSS, COD and the energy consumption. For the mathematical model the logistic regression was applied. The input data used were from a waste treatment plant in Romania, for a period of 2 years 2015 and 2017, a total of random 403 dataset. The treatment technologies of WWTP consist of advanced biological treatment SBR (nitrification, denitrification, and phosphorus removal), aerobic sludge stabilization, dewatering, storage and chemical disinfection. Octave software was used to build the model. The answer of the model refers to the fact that for a given situation there will be high energy consumption or low energy consumption in the wastewater treatment plant (WWTP). Performance of the model was compared with real value.

Machine Learning-Based Energy Consumption Estimation of Wastewater Treatment Plants in Greece

“…Bonescu et al [23] considered the logistic regression method (LRM). This approach involved the use of input variables such as flow rate and wastewater properties to predict the EC of a WWTP in Romania.…”

Influence of Effluent Quality from Sludge Dewatering on Electricity Consumption

“…Excess sludge is thickened in the mechanical thickener (8) and mixed with the preliminary sludge before being sent to the mesophilic fermentation chamber (9). The fermented sludge is dewatered in the decantation centrifuge (10). The sludge dewatering process is assisted by fl occulant.…”

“…al. found that the concentration of pollutants among others total suspended solids (TSS) in the sewage was a signifi cant predictor of electricity consumption in the downstream biological treatment process 10 . Research by Mininni et.…”

“…Phosphorus may be recovered from the stream by precipitation as struvite (magnesium ammonium phosphate) and next may be used as a fertilizer. At the moment, high-effi ciency recovery technologies have been Polish Journal of Chemical Technology, 25, 3, 50-55, 10.2478/pjct-2023-0024 developed 26 . Some of them allow the recovery of more than 90% of phosphorus.…”

Predicting Energy Consumption in Wastewater Treatment Plants through Light Gradient Boosting Machine: A Comparative Study

“…Results demonstrated that in terms of the Root mean squared error (RMSE) and Mean Absolute Error (MAE) values, the GBU had the lowest values in the test phase, 33.9 and 26.9, respectively. In [12], logistic regression was employed to estimate the energy consumption of a WWTP. The effectiveness of this approach has been assessed using data with 403 records collected from Romania between 2015 and 2017.…”