Desarrollo de geotecnologías aplicadas a la inspección y monitorización de entornos industriales

Resumen

The technological development of the last two decades has brought about a radical change that is leading to a new paradigm in which the physical and digital worlds are intertwined. These changes have had a great impact on society, modifying communication methods, access to information, leisure, work, etc. In addition, the industry has adopted these disruptive technologies, which are contributing to achieving greater control and automation of the production process. In the industrial sector, maintenance tasks are critical to ensuring the proper operation of a plant or facility, as they directly influence productivity and can involve high additional costs. New technologies are making continuous monitoring and automated inspection possible, providing auxiliary tools to inspectors that improve fault detection and allow for the anticipation and optimization of maintenance task planning. With the aim of developing tools that provide improvements in maintenance tasks in industry, this doctoral thesis is based on the study of how geotechnologies can provide optimal solutions in monitoring and inspection. Due to the great variety of industrial environments, maintenance support tools must adapt to each specific case. In this regard, and in order to demonstrate the adaptability of geomatics and geotechnologies, industrial installations from very diverse areas have been studied, such as a machine room (indoor scenario), photovoltaic plants (outdoor scenario), and welding (indoor and outdoor scenarios). The scale of the studied scenarios has been very varied, ranging from smaller scales for the study of welds and machine rooms, to larger scales in the studies of vegetation evolution and the presence of bodies of water in photovoltaic plants. Geotechnologies demonstrate their versatility to work at different scales, with solutions that allow for great detail and precision, such as close-range photogrammetry and the Portable Mobile Mapping System (PMMS), as well as others that can cover larger areas of the territory, such as remote sensing or photogrammetry with drones. The focus of the thesis has been the study of industrial elements or installations at different scales. In the first case, a tool was developed for external quality control of welding, using close-range photogrammetry and algorithms for automatic defect detection. In the second case, the use of a PMMS is proposed to optimize data collection in fluid-mechanical installation inspection tasks. In the third case, drone photogrammetry and the combination of RGB and thermal images with computer vision algorithms were used for the detection of pathologies in photovoltaic panels. Finally, for the monitoring of vegetation and the detection of water masses in the environment of photovoltaic plants, remote sensing was employed through the calculation of spectral indices.