Keywords: heavy metals, pollutant plume, catchment area, real genetic algorithm, finite difference method, Lattakia-Syria.

The purpose of this paper is to investigate the capability of a real genetic algorithm (GA) to remove the heavy metals pollutant plume from an aquifer. The real GA is used in conjunction with proven and accepted finite difference analysis to reach an efficient solution. It also illustrates the interaction between the real GA optimizer and the finite difference technique. Two scenarios regarding well positions, the number of wells, the rate of pumping at each well and total cost were determined for the purpose. A new system has been developed for removing heavy metals (mercury Hg and cadmium Cd) from groundwater in a real problem, and proved to be optimal. Real-coding GAs proved to be a practical means of optimizing engineering solutions to problems related to groundwater quality management, in particular those involving discontinuous functions. The real GA can be used with an accredited finite difference method towards an optimal solution. The developed application of a combined simulation-optimization method for the clean-up of contamination from a real landfill in a coastal area of Syria, can also be used in other water resources quality applications.

The major conclusions are as follows: according to Scenario 1, for a line of wells along the center line of plume, four wells are required to remove the pollutants in a maximum of fifteen time steps. The required pumping rate at each of the wells is 25m³/d. This solution was obtained after fifteen generations. By increasing the period to twenty time steps, only two wells will be required with a rate of 15m³/d. This solution was obtained after 21 generations. According to Scenario 2, four active wells are required to remove the pollutants in a maximum of fifteen time steps. The required pumping rate at each of the wells is 25m³/d. This solution was obtained after nineteen generations. By increasing the period to twenty time steps, only two wells will be required with a rate of 15 m3/d. This solution was obtained after 25 generations. Since Scenario 1 solution for 20 time steps proved to be identical to Scenario 2 solution for twenty time steps, and because Scenario 1 can be part of Scenario 2, it is possible to dispense with Scenario 1.

The run time of computer in the real coded GA for the real aquifer studied, proved to take much less time than that in the binary coded GA. Real-coding GAs proved to be a practical means of optimizing engineering solutions for problems related to groundwater quality management, in particular those involving discontinuous functions. The real GA can be used with accredited finite difference analysis programs towards an optimal solution.

The optimization process developed here for groundwater pollution removal including heavy metals can benefit from by applying its efficiency to other water engineering problems including those involving multiparameters, non-linear, and discontinuous functions and to complex systems as well. At the same time, it can be considered a potential as an alternative to standard traditional optimization techniques.

The application of a combined simulation-optimization method developed for the clean-up of contamination from a real landfill in a coastal area of Syria, can also be used in other water resources quality applications.

Finally, this research and its conclusions may encourage other researchers to apply the same methodology concerning other dangerous heavy metals existing in groundwater.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £70 +P&P)