Global optimisation techniques in water resources management

Efstratiadis, A; Koutsoyiannis, D

dc.contributor.author	Efstratiadis, A	en
dc.contributor.author	Koutsoyiannis, D	en
dc.date.accessioned	2014-03-01T02:54:18Z
dc.date.available	2014-03-01T02:54:18Z
dc.date.issued	2001	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/36816
dc.title	Global optimisation techniques in water resources management	en
heal.type	conferenceItem	en
heal.publicationDate	2001	en
heal.abstract	Optimisation has become a valuable tool in most of hydroinformatics applications, such as calibration of hydrological models, optimal control of hydrosystems, water quantity and quality management, water supply and sewage networks design, etc. Given that these problems are intrinsically nonlinear and multimodal, they do not exist deterministic optimisation methods that can locate the globally optimal solution. During the last two decades, probabilistic schemes have been developed for solving global optimisation problems. These methods use a combination of random and deterministic steps, without generally requiring restrictive conditions on the nature of the objective function. The scope of this study is the investigation of the features of these techniques, focusing on three of them, which are presented and compared by means of both mathematical applications and real-world problems. The first two are the most popular in applications related with hydrology and water resources, i.e. genetic algorithms and the shuffled complex evolution algorithm. The third one is a new simplex-annealing scheme, which incorporates the principles of simulated annealing in the well-known downhill simplex method. This scheme is very simple to implement and extended analysis proved that it is very effective in locating the global optimum as well as very efficient, in terms of convergence speed.	en