Estimating NDMA Formation in a Distribution System Using a Hybrid Genetic Algorithm

A model was developed for estimation of N-nitrosodimethylamine (NDMA) concentration in a case study of a chloraminated water distribution system in Australia using influencing parameters that had an impact on NDMA formation via a hybrid genetic algorithm method. The model was based on hydraulic retention time, temperature, and monochloramine residual. The genetic algorithm and quasi-Newton optimization with a Broyden-Fletcher-Goldfarb-Shanno Hessian updating method were combined into a hybrid parameter optimization scheme for robust calibration of NDMA model parameters. In order to test the applicability and the accuracy of the proposed model, NDMA concentration and water quality characteristics for 220 post-disinfection water samples collected between 2006 and 2016 were used. Good agreement between measured and estimated NDMA concentrations was observed by performing statistical analysis on an independent set of water sample data (R-2 = 0.93). This modeling approach has the potential to be used as a decision support tool in order to manage and improve the chloramine disinfection process.