1Department of Civil Engineering, Indian Institute of Technology, Bombay, Mumbai, India
2Water Resources Engineering, Department of Civil Engineering, Indian Institute of Technology, Bombay, Mumbai, India
Efficient management of an aquifer depends primarily on adequate knowledge of its hydro-geologic parameters. Estimation of the spatially distributed aquifer parameters in large aquifer systems often involves considerable amount of time and financial resources. These parameters are required for the groundwater system simulation to predict head and concentration behaviour in the flow domain. Inverse modelling of the system helps in the adequate assessment of these parameters for a meaningful system simulation. A parameter estimation method is developed based on genetic algorithm (GA) a global optimization method. A finite element model (FEM) based on Galerkin's approach is developed for computing the head and concentration distribution in the aquifer domain. The developed numerical model is applied to a confined aquifer involving a set of boundary conditions, source and sink terms to compute the system parameters within the chosen three zones of the aquifer. The results show that the genetic algorithm can be successfully used to obtain optimal estimate of aquifer parameters. The chosen aquifer problem is also solved by Gauss-Newton-Marquardt (GNM) method to estimate aquifer parameters and the results are compared with those obtained by using genetic algorithm. Largely the GA results are better compared to GNM parameter estimates, except, when error free head and concentration data (practically not possible) is used.
