|
| Optimization of In-Situ Carbon Injection Bioremediation of Nitrate Contaminated Groundwater |
| Received:June 16, 2014 |
| View Full Text View/Add Comment Download reader |
| KeyWord:nitrate;RT3D;in situ bioremediation;optimization;genetic algorithm;simulated annealing |
| Author Name | Affiliation | E-mail | | JIANG Lie | Beijing Key Laboratory of Water Resources and Environment Engineering, China University of Geosciences(Beijing), Beijing 100083, China
Jiangxi Geological Environment Monitoring Station, Nanchang 330012, China | | | HE Jiang-tao | Beijing Key Laboratory of Water Resources and Environment Engineering, China University of Geosciences(Beijing), Beijing 100083, China | jthe@cugb.edu.cn | | XU Zhen | Beijing Key Laboratory of Water Resources and Environment Engineering, China University of Geosciences(Beijing), Beijing 100083, China | | | LIU Yu-mei | Beijing Key Laboratory of Water Resources and Environment Engineering, China University of Geosciences(Beijing), Beijing 100083, China | |
|
| Hits: 2768 |
| Download times: 3111 |
| Abstract: |
| In-situ carbon injection is a promising and economically effective technique for enhanced biodenitrification of groundwater contaminated by nitrate. Optimizing carbon injection can improve remediation efficiency. In this paper, a landfill site in Beijing was selected to perform an optimization study. A biological coupling Monod solute transport model(RT3D) based on in situ bioremediation technology was established. Genetic Algorithm(GA) and Simulated Annealing(SA) were used to optimize the layout of potential carbon(ethanol) injection wells in the control areas of groundwater nitrate plumes. The results show that, at 10 mg·L-1 of target nitrate concentration in the studied area, both GA and SA methods obtained the same three carbon injection wells, 15 wells fewer than the originally designed ones. These three injection wells should be arranged in triangular position. More carbon source should be injected in the upstream well than in two downstream wells. After 100 d of injection, the total removal of nitrate nitrogen was 90.78% and 84.51% for GA and SA, respectively. Compared to GA, SA cost 1.46% less, with stronger convergence and smaller variability but longer computing time. |
|
|
|
