The aim of this research was to assess the treatment efficiencies for concentrated stormwater runoff (gully pot liquor) of experimental vertical-flow constructed wetland filters containing common reed and different aggregates. For two years, six out of twelve filters received inflow water spiked with hydrated nickel and copper nitrate to simulate contaminated primary treated storm runoff. For those six constructed filters, an obvious breakthrough of dissolved nickel was recorded after road salting during the first winter. However, a breakthrough of nickel was not observed since the inflow pH was raised to eight after the first year of operation. During the second year, reduction efficiencies of heavy metals, five-day biochemical oxygen demand (BOD) and suspended solids (SS) improved considerably. Concentrations of BOD were frequently <20 mg/l, an international threshold for secondary wastewater treatment. This is likely due to biomass maturation and the increase of pH. Machine learning techniques such as K-nearest neighbours, support vector machine and self-organizing map were applied to predict BOD and SS, and to demonstrate an alternative method of analyzing water quality performance indicators. The results suggest that BOD and SS can be efficiently estimated by applying machine learning tools with cost-effective input variables such as redox potential and conductivity, which can be monitored in real time. Their performances are encouraging and support the potential for future use of these models as management tools for the day-to-day process control of constructed wetlands and other ‘black box’ systems.
Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:653760 |
Date | January 2006 |
Creators | Lee, Byoung-Hwa |
Publisher | University of Edinburgh |
Source Sets | Ethos UK |
Detected Language | English |
Type | Electronic Thesis or Dissertation |
Source | http://hdl.handle.net/1842/15197 |
Page generated in 0.0298 seconds