Please use this identifier to cite or link to this item:
|Title:||Performance of horizontal roughing filter using principal component regression and multiple linear regression treating informal settlement greywater||Authors:||Mtsweni, S.
Bakare, B. F.
|Keywords:||Principal component regression;Horizontal roughing filter;Principal component analyses;Greywater;Informal settlement||Issue Date:||2019||Publisher:||WCECS||Source:||Mtsweni, S., Bakare, B.F., Rathilal, S. 2019. Performance of horizontal roughing filter using principal component regression and multiple linear regression treating informal settlement greywater. In: Proceedings of the World Congress on Engineering and Computer Science. San Francisco, October 2019. 28-32.||Conference:||World Congress on Engineering and Computer Science 2019||Abstract:||Water scarcity remain a major challenge facing many countries around the world. These water challenges results to seeking other possible water alternatives techniques to save available water. Greywater reuse is one such alternative to save water if it can be treated successfully to meet greywater standards for reuse. However, within the usability of greywater at the core is the ability to monitor its quality during treatment that will meet the required standard of water for reuse particularly when water reuse alternatives are considered. The aim of the this paper is to demonstrate beyond the potential usability of the Horizontal Roughing Filter (HRF) which was identified as a possible greywater reuse treatment option/technology while monitoring its performance and effluent greywater quality using principal component approach and principal component regression techniques in HRF as a tool of provision of possible practical solution in community facing water challenges. The study was conducted using a pilot scale HRF to treat informal settlement greywater in the study area in Durban Umlazi, Southern Africa. Results showed that HRF was able to achieve removal efficiency of turbidity in greywater effluent above 90% at a filtration ate of 0.3 m/h and 60% chemical oxygen demand (COD). The predictor variables were temperature, COD, conductivity, total solids and pH. The principal component regression was more robust to identify performance relationship in the data better than multiple linear regression.||URI:||http://hdl.handle.net/10321/3394||ISBN:||9789881404879||ISSN:||2078-0958|
|Appears in Collections:||Research Publications (Engineering and Built Environment)|
Show full item record
Files in This Item:
|MTSWENI_WCECS2019_Pg28-32_2019.pdf||1.21 MB||Adobe PDF||View/Open|
checked on Aug 9, 2020
checked on Aug 9, 2020
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.