Probabilistic Building Block Identification for the Multi-Objective Design and Rehabilitation of Water Distribution Systems
Publication: Water Distribution Systems Analysis 2008
Abstract
The multi-objective design and rehabilitation of water distribution systems (WDS) is defined as the search for the set of system designs which offer the best trade-off between a set of competing design objectives. Typically these objectives will consist of the cost of implementing a system design and a measure of the performance of that design. A recent development in the field of evolutionary algorithms is to use probabilistic methods to identify key building blocks (short, highly fit groups of genes). Probabilistic Model Building Genetic Algorithms replace the traditional crossover and mutation methods for the generation of offspring with the building and sampling of a probabilistic model describing the genomes in a set of promising solutions. Other methodologies use similar probabilistic methods to identify building blocks, but continue to use the traditional techniques of crossover and mutation with the condition that building blocks cannot be split. This paper compares the performance of three evolutionary algorithms which employ probabilistic building block identification to speed up convergence with that of the well known multi-objective genetic algorithm NSGAII for the multi-objective optimisation of a real world WDS. In particular the Univariate Marginal Distribution Algorithm, the hierarchical Bayesian Optimisation Algorithm and the Chi-square matrix method for building block identification are considered.
Get full access to this chapter
View all available purchase options and get full access to this chapter.
Information & Authors
Information
Published In
Copyright
© 2008 American Society of Civil Engineers.
History
Published online: Apr 26, 2012
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.