Information Technology Reference
In-Depth Information
Identification of Groundwater Parameter Structure
Using Harmony Search Algorithm
M. Tamer Ayvaz
Department of Civil Engineering, Pamukkale University, Denizli, Turkey
tayvaz@pamukkale.edu.tr
Abstract.
Mathematical simulation models are widely used to predict the future response of
groundwater systems for different flow and mass transport conditions. These models are based
on the solution of governing partial differential equations which require some spatially distrib-
uted hydro-geological model parameters. However, these parameters are usually unknown due
to the complexity of groundwater systems. Therefore, identification of them is an important
task since they are the primary input of management models used in groundwater modeling.
This chapter provides a brief review dealing with the solution of parameter structure identifica-
tion problems based on the harmony search optimization algorithm. The results of this review
indicate that the harmony search algorithm yields nearly same or better solutions than those of a
genetic algorithm, which is another popular meta-heuristic optimization algorithm.
Keywords:
Groundwater Management, Parameter Structure Identification, Simulation-Optimization,
Harmony Search Algorithm.
1 Introduction
Groundwater is an important water resource throughout the world. Nowadays, the sig-
nificance of the groundwater resources continuously grows as society and economics de-
velop. However, uncontrolled and unplanned use of groundwater resources may cause
some serious geological problems, including groundwater depletion, seawater intrusion,
land subsidence and desertification [1]. Therefore, sustainable planning and management
strategies should be developed to optimally operate the groundwater resources.
Mathematical simulation models are the essential tools of developing sustainable
management strategies for groundwater systems. They are used to evaluate the aqui-
fer's response for different management strategies [2] by solving governing flow and
mass transport equations which require the spatial distributions of some hydro-
geological parameters (hydraulic conductivity, transmissivity, storativity, etc.). Note
that these parameters are usually determined by some laboratory or field studies, and
determination of them for large groundwater systems may require significant time and
cost. Therefore, much research has been conducted to determinate these parameter dis-
tributions based on field observations of hydraulic head values. This approach is identi-
fied as inverse modeling and usually solved using combined simulation/optimization
(S/O) models [3].
It should be noted that combined S/O models have been employed to the solution
of many groundwater modeling problems. These problems may be classified as three
main groups: Groundwater parameter estimation problems; Groundwater hydraulic
Search WWH ::
Custom Search