Institutional Repository
Technical University of Crete
Technical University of Crete
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/14D1030B-76A0-40E5-AC12-696FA58AFF6A | ||
| Year | 2016 | ||
| Type of Item | Diploma Work | ||
| License |
|
||
| Bibliographic Citation | Maria Christoforidou, "Optimal management of salt water intrusion phenomenon of Katapola aquifer of Amorgos island using genetic algorithms", Diploma Work, School of Environmental Engineering, Technical University of Crete, Chania, Greece, 2016 https://doi.org/10.26233/heallink.tuc.66693 | ||
| Appears in Collections | |||
| Relations with other Items |
Salt water intrusion is considered as one of the most common issues in costal aquifers, while, in case of islands, the problem is magnified by the lack of alternative solution for water supply. Additional pressure is applied due to tourism and increased water demand. This is the case for many Greek islands. Amorgos is one of those islands and as a result, the island has to receive water quantities through shipping during summertime. The management of costal aquifers is crucial because of the underlying danger of permanent degradation of the aquifer. As a solution, four different plans are suggested and examined so that the salt water intrusion is constrained. These strategies were created through optimization process. In order to achieve global optimal, or near global optimal, a groundwater simulation model (Argus One – PTC) is being coupled with an optimization technique (Genetic Algorithm – MATLAB optimization tool). The initial groundwater model (Σιάκα, 2015) was adjusted so that it can predict the hydraulic heads, calculated for specific pumping rates, for the next five years (2016-2020). This simulation model is being used as an evaluation method of the fitness of various sets of pumping rates which are created from the Genetic Algorithm. The algorithm creates these sets of pumping rates through selection, crossover and mutation operators. The stopping criteria of the algorithm were set to be the maximum number of 100 generation. Each generation consists of 27 individuals in every population. As a result, the outcome of the optimization may not have yet reached a near optimal solution.
Copyright © DIAS 2013
