Danger Level Ranking of Possible Dam Failures in Turkey by Grey Relational Analysis


Akdemir H., Bayındır C.

4th International Conference on Intelligent and Fuzzy Systems (INFUS), Bornova, Turkey, 19 - 21 July 2022, vol.504, pp.237-243 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 504
  • Doi Number: 10.1007/978-3-031-09173-5_30
  • City: Bornova
  • Country: Turkey
  • Page Numbers: pp.237-243
  • Keywords: Grey relational analysis, Dam failures, Danger level ranking, Turkey's dams, SYSTEM-THEORY
  • Yıldız Technical University Affiliated: No

Abstract

Although dam failures are not very common in the course of engineering history, they can be catastrophic disasters causing many life losses, so they need to be investigated. Dam failures are difficult problems to analyze due to the complexity of the associated parameters which can be very hard to determine. With this motivation, in this study, grey relational analysis (GRA) was used to rank the danger levels of Turkey's aging 15 dams in case of a possible collapse. The failure mode of each dam was assumed to be a sudden collapse, thus breach development is not considered. This type of failure mode is more commonly observed during major earthquakes. The dams chosen for this study have been selected from the engineering point of view as they have the highest hazard potential in case of failure across the country. Accordingly, the important involved attributes of the model were determined as follows: surrounding population, distance from that population, elevation relative to that population, and reservoir size. The weights of these involved attributes were preferred as 0.40, 0.30, 0.10, and 0.20, respectively. In conjunction with the literature on the subject, the distinguished coefficient was selected as 0.5. The risk assessment based on the GRA results is performed for Turkey's 15 dams involved in the study. The output of this study will contribute to the disaster and risk management policies of Turkey's dams and will have similar applications worldwide.