Abstract:
The multicomponent reaction transport model is required to predict the concentrations of chlorine, organic content and biomass within a water distribution system. The estimated concentrations of these substances are largely affected by the various parameters used in the model. These parameters are grouped into those controlling chlorine decay and others governing the bacterial growth/substrate utilization. The parameter values used in the previous works show that the range of their values is wider. Further, the estimation of these parameters in the field is difficult. An efficient inverse model is essential to predict the correct set of these parameters so that the field measurements are reasonably well simulated. In the present study, the potential of Genetic Algorithm (GA) for the estimation of both these groups of parameters is explored. An inverse model, which includes Genetic Algorithm in its optimization module, is developed. The objective function used in the model compares the measured and simulated chlorine, substrate and biomass concentrations at the monitoring nodes in a weighted least square sense. The usefulness of the model developed is illustrated by applying it onto a large water distribution system under dynamic state. It is also attempted to identify the critical parameters governing the biomass growth and substrate utilization using this model. The GA based inverse model developed in this study is found to be a useful tool for estimating these parameters.
