Advances in Engineering Software 32 (2001) 467±475
GAnet: genetic algorithm platform for pipe network optimisation
Â M.S. Morley, R.M. Atkinson, D.A. Savic*, G.A. Walters
Centre for Water Systems, School of Engineering and Computer Science, Department of Engineering, University of Exeter, Exeter, Devon EX4 4QF, UK Accepted 17 October 2000
Abstract The marriage of object-oriented programming techniques and genetic algorithms (GAs) provides a uniquely ¯exible environment for the development of evolution programs with technical applications. Computerised representations of hydraulic network models are used in various guises throughout the water industry. As well as being commonly linked to relational databases to form asset management systems and to geographic information systems (GIS) to provide visualisation and mapping facilities, common applications include modelling the hydraulic performance of the network and use in network optimisation applications. This paper describes an architecture for an integrated optimisation application, GAnet, which comprises a GA application, a GIS and a hydraulic network solver. q 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Evolution; Genetic algorithm; Optimisation; Object-oriented; Water network
1. Introduction Evolution programs , of which GAs are probably the best-known types, are general, arti®cial evolution search methods based on natural selection and mechanisms of population genetics. They emulate nature's effective optimisation techniques of evolution Ð being based on preferential survival and reproduction of the ®ttest members of the population, the maintenance of a population with diverse members, the inheritance of genetic information from parents and the occasional mutation of genes. These algorithms are best suited to solving combinatorial optimisation problems that cannot be solved practicably using more conventional operational-research methods....