A novel approach for water quality management in water distribution systems by multi-objective booster chlorination
International Journal of Civil Engineering
Iran University of Science and Technology (IUST) and Iran Society of Civil Engineers
Compared to conventional chlorination methods which apply chlorine at water treatment plant, booster chlorination has almost solved the problems of high dosages of chlorine residuals near water sources and lack of chlorine residuals in the remote points of a water distribution system (WDS). However, control of trihalomethane (THM) formation as a potentially carcinogenic disinfection by-product (DBP) within a WDS has still remained as a water quality problem. This paper presents a two-phase approach of multi-objective booster disinfection in which both chlorine residuals and THM formation are concurrently optimized in a WDS. In the first phase, a booster disinfection system is formulated as a multi-objective optimization problem in which the location of booster stations is determined. The objectives are defined as to maximize the volumetric discharge with appropriate levels of disinfectant residuals throughout all demand nodes and to minimize the total mass of disinfectant applied with a specified number of booster stations. The most frequently selected locations for installing booster disinfection stations are selected for the second phase, in which another two-objective optimization problem is defined. The objectives in the second problem are to minimize the volumetric discharge avoiding THM maximum levels and to maximize the volumetric discharge with standard levels of disinfectant residuals. For each point on the resulted trade-off curve between the water quality objectives optimal scheduling of chlorination injected at each booster station is obtained. Both optimization problems used NSGA-II algorithm as a multi-objective genetic algorithm, coupled with EPANET as a hydraulic simulation model. The optimization problems are tested for different numbers of booster chlorination stations in a real case WDS. As a result, this type of multi-objective optimization model can explicitly give the decision makers the optimal location and scheduling of booster disinfection systems with respect to the trade-off between maximum safe drinking water with allowable chlorine residual levels and minimum adverse DBP levels.
Copyright © 2012 International Journal of Civil Engineering
Vol. 10 (1), pp. 51 - 60