Advanced

Robotic Gas Source Localization in an Industrial Environment

Persson, Erik (2010) In MSc Theses
Department of Automatic Control
Abstract
Gas leaks are an important safety issue in oil and gas production. For example, natural gas often contains large portions of hydrogen sulfide, a gas that is lethal to humans in concentrations as low as 0.1%. In addition natural gas itself is explosive. During the past fifteen years, a considerable number of studies have been made into how to detect and localize gas leaks. Equipped with sensors measuring the point concentration of specific substances, a variety of mobile robots and algorithms have been looking for gas sources indoors and outdoors, underground and under water, in airless conditions and in windy dittos. Due to the complexity of turbulence and the limitations of gas sensors, robotic gas source localization has turned out to be... (More)
Gas leaks are an important safety issue in oil and gas production. For example, natural gas often contains large portions of hydrogen sulfide, a gas that is lethal to humans in concentrations as low as 0.1%. In addition natural gas itself is explosive. During the past fifteen years, a considerable number of studies have been made into how to detect and localize gas leaks. Equipped with sensors measuring the point concentration of specific substances, a variety of mobile robots and algorithms have been looking for gas sources indoors and outdoors, underground and under water, in airless conditions and in windy dittos. Due to the complexity of turbulence and the limitations of gas sensors, robotic gas source localization has turned out to be complicated and so far it has not made its way to large scale real world applications. This study is an attempt to bring robotic gas source localization a bit closer to that. Three algorithms, carefully chosen from the literature, are adapted to an industrial environment. In addition, two novel strategies are derived from the original ones through combination of them. A comparative study between the five algorithms is made where their performances are evaluated and compared. This study has been conducted within a project of ABB in Oslo that investigates how industrial robots can be used in an oil and gas-context. (Less)
Please use this url to cite or link to this publication:
author
Persson, Erik
supervisor
organization
year
type
H3 - Professional qualifications (4 Years - )
subject
publication/series
MSc Theses
report number
TFRT-5872
ISSN
0280-5316
language
English
id
8847505
date added to LUP
2016-03-16 12:41:52
date last changed
2016-03-16 12:41:52
@misc{8847505,
  abstract     = {Gas leaks are an important safety issue in oil and gas production. For example, natural gas often contains large portions of hydrogen sulfide, a gas that is lethal to humans in concentrations as low as 0.1%. In addition natural gas itself is explosive. During the past fifteen years, a considerable number of studies have been made into how to detect and localize gas leaks. Equipped with sensors measuring the point concentration of specific substances, a variety of mobile robots and algorithms have been looking for gas sources indoors and outdoors, underground and under water, in airless conditions and in windy dittos. Due to the complexity of turbulence and the limitations of gas sensors, robotic gas source localization has turned out to be complicated and so far it has not made its way to large scale real world applications. This study is an attempt to bring robotic gas source localization a bit closer to that. Three algorithms, carefully chosen from the literature, are adapted to an industrial environment. In addition, two novel strategies are derived from the original ones through combination of them. A comparative study between the five algorithms is made where their performances are evaluated and compared. This study has been conducted within a project of ABB in Oslo that investigates how industrial robots can be used in an oil and gas-context.},
  author       = {Persson, Erik},
  issn         = {0280-5316},
  language     = {eng},
  note         = {Student Paper},
  series       = {MSc Theses},
  title        = {Robotic Gas Source Localization in an Industrial Environment},
  year         = {2010},
}