A Method for Estimating the Diameter of Water Pipes Using Artificial Neural Networks of the Multilayer Perceptron Type

DOI

10.2991/icaita-18.2018.13How to use a DOI?

Keywords

water distribution system; hydraulic calculations; diameter selection; artificial neural networks; classification

Abstract

Designing a water distribution system is inherently associated with hydraulic calculations, the primary purpose of which is to select the diameters of water pipelines. Computer programs may choose diameters, but most often, this task is up to the designer. It is necessary to control flow velocity; at the same time, however, it is also necessary to minimise pressure losses in the water mains network. In order to improve the above process, an artificial neural network was designed, which, after hydraulic calculations, evaluates the accuracy of the diameters selected on a classification basis. The output layer of the neural network consists of ten neurons corresponding to the nominal diameters of the water pipes. Classification of the correct pipeline diameter is done using "one-of-N" encoding, that is, only one neuron from the output layer is activated, thus selecting the correct diameter. Based on data from hydraulic calculations, the neural network diagnoses the diameter of the pipelines along individual sections of the water supply network and proposes appropriate values or accepts the existing ones.

Copyright

© 2018, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Jacek Dawidowicz
PY  - 2018/03
DA  - 2018/03
TI  - A Method for Estimating the Diameter of Water Pipes Using Artificial Neural Networks of the Multilayer Perceptron Type
BT  - Proceedings of the 2018 2nd International Conference on Artificial Intelligence: Technologies and Applications (ICAITA 2018)
PB  - Atlantis Press
SP  - 50
EP  - 53
SN  - 1951-6851
UR  - https://doi.org/10.2991/icaita-18.2018.13
DO  - 10.2991/icaita-18.2018.13
ID  - Dawidowicz2018/03
ER  -