Solar-powered electrocoagulation unit with ANN-based prediction for sustainable treatment of oily wastewater
Więcej
Ukryj
1
Environmental Research Center, University of Technology-Iraq, Al-Sina'a Street, 10066 2 Baghdad, Iraq
2
College of chemical engineering, University of Technology-Iraq, Al-Sina'a Street, 10066 Baghdad, Iraq
3
Higher Institute of Artificial Intelligence, University of Technology-Iraq, Al-Sina'a Street, 10066 Baghdad, Iraq
4
Institute of Nanotechnology and Advanced Materials Research for Postgraduate Studies, University of Technology-Iraq, Al-Sina'a Street, 10066 Baghdad, Iraq
5
Petroleum Research and Development Center, Ministry of Oil, Bub Al-Sham, Baghdad, Iraq
Zaznaczeni autorzy mieli równy wkład w przygotowanie tego artykułu
SŁOWA KLUCZOWE
DZIEDZINY
STRESZCZENIE
Oily refinery effluent is still a major environmental concern, because it is so difficult to treat as well as the large amounts of energy used in treating it using traditional technology. This study evaluated a low-cost and renewable source of energy system known as a solar-powered electro-coagulation unit (SPEU), of treating refinery wastewater while providing an opportunity for energy savings. Real oily effluent collected from a refinery in Baghdad (Al-Daura Refinery, Iraq) has been treated to electrocoagulation using aluminum sacrificial electrodes with variable operating parameters. The impacts of electrical current density and time for electrolysis on removal efficiencies for turbidity, oil and grease, TSS, sludge generation and energy consumption have also been examined. An ANN model was established for predicting the removal efficiency based upon the operational parameters that are considered the model input values. Experimental data were collected which illustrated how current impacts pollutant removal; the optimal operating conditions in terms of current and treatment time (i.e., 5A, 45 minutes) resulted in 87% turbidity removal, 97% suspended solids removal, and up to 88% oil removal using pre-treatment while consuming an average of 15 kW-h/m3. In addition, the optimized ANN was capable of accurately predicting both the performance of the system and experimental data that had not been used during the development of the model by generating a minimum mean square error of 9.05 × 10-17 and a training correlation coefficient of R = 1. these results show a potential opportunity for electrochemical treatment systems driven by renewable energy to provide sustainable, intelligent alternatives to industrial wastewater management.