Application of response surface methodology and artificial neural network for optimizing phosphate removal from lagoon wastewater
| dc.contributor.author | Khordadi, Aidin | |
| dc.contributor.examiningcommittee | Lozecznik, Stanislaw (Civil Engineering) | |
| dc.contributor.examiningcommittee | Rout, Prangya (Fermibio Solutions Pvt. Ltd.) | |
| dc.contributor.supervisor | Yuan, Qiuyan | |
| dc.date.accessioned | 2024-07-31T19:29:45Z | |
| dc.date.available | 2024-07-31T19:29:45Z | |
| dc.date.issued | 2024-07-24 | |
| dc.date.submitted | 2024-07-26T16:39:52Z | en_US |
| dc.date.submitted | 2024-07-31T17:59:15Z | en_US |
| dc.degree.discipline | Civil Engineering | |
| dc.degree.level | Master of Science (M.Sc.) | |
| dc.description.abstract | Lagoons are primary wastewater treatment methods used in rural municipalities and small communities in Canada. This study aimed to optimize phosphate removal and reduce generated sludge in lagoon wastewater treatment under varying operational parameters. Two series of bench scale experiments were conducted to evaluate phosphate removal and sludge production using aluminum sulphate and ferric chloride as coagulants, and cationic polymers as coagulant aids. Two models were developed to predict optimal conditions for phosphate removal and sludge production. Response Surface Methodology (RSM) with optimal design was employed to assess the impact of pH, temperature, coagulant dosage and type, and flocculant type and dosage on the responses. Subsequently, a feedforward multilayer Artificial Neural Network (ANN) model was developed based on RSM inputs, along with floc perimeter and area, to forecast final phosphate levels in effluent and the amount of generated sludge. The results revealed that polymers with 40% cationic charge and higher molecular weight were more efficient compared to polymers with higher charge and lower molecular weights. Additionally, the R-squared (R2 ) values for the RSM models were 0.9264 and 0.9194 for phosphate removal and sludge production, respectively. The corresponding R2 values for the ANN models were 0.7994 and 0.7965, indicating good predictive performance. | |
| dc.description.note | October 2024 | |
| dc.identifier.uri | http://hdl.handle.net/1993/38347 | |
| dc.language.iso | eng | |
| dc.rights | open access | en_US |
| dc.subject | Lagoon wastewater | |
| dc.subject | Phosphorus removal | |
| dc.subject | RSM and ANN | |
| dc.title | Application of response surface methodology and artificial neural network for optimizing phosphate removal from lagoon wastewater | |
| dc.type | master thesis | en_US |
| local.subject.manitoba | yes |