Non-Invasive Visual Computing Modelling for Obstruction Inspection of Pipes in the Fertilizer Industry

  • Marta Duarte FURG
  • Victor Coch FURG
  • Jovania Dias FURG
  • Silvia Botelho FURG
  • Nelson Duarte FURG
  • Paulo Drews FURG

Resumo


In the context of industrial automation, one of the areas that have seen a lot of increased demand is the fertilizer industry, that has to keep up with agricultural needs around the world. These factories have several ducts where unwanted chemical reactions take place, resulting in congealment of product and clogging of the pipes over time. These obstructions compromise production and the whole machinery, causing a major financial loss, hence the need for a non-invasive method to determine to what extent a pipe or duct is obstructed. This paper presents a thermal imaging approach to model clogging and deposit distributions within pipes, rooted in visual computing, along with the use of laser based 3D scanning for calibration and result validation.
Palavras-chave: Fertilizers, Ducts, Three-dimensional displays, Cameras, Visualization, Laser modes
Publicado
23/10/2019
DUARTE, Marta; COCH, Victor; DIAS, Jovania; BOTELHO, Silvia; DUARTE, Nelson; DREWS, Paulo. Non-Invasive Visual Computing Modelling for Obstruction Inspection of Pipes in the Fertilizer Industry. In: SIMPÓSIO BRASILEIRO DE ROBÓTICA E SIMPÓSIO LATINO AMERICANO DE ROBÓTICA (SBR/LARS), 16. , 2019, Rio Grande. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2019 . p. 280-285.