Modeling of temperature distribution of engineering objects on the basis of data obtained from a thermal imaging camera and geodetic instruments

Abstract

In the modern world, collections of various types of data have become the most valuable commodity. Data is obtained automatically or manually and comes from various measuring techniques, monitoring systems, sensors or methods of tracking human activity on the Internet. This combined with the idea of intelligent devices, which can automatically exchange data (Internet of Things), creates unimaginably large sets of data that seem to be unrelated to each other. Only the application of advanced Big data analysis can detect connections that are not visible at first glance. A common desire to integrate various types of measurement data also applies to engineering facilities. In the case of modeling temperature distribution, the integration of data from terrestrial laser scanning and thermovision, i.e. mapping of thermal images on point clouds, is very popular. This provides a three-dimensional representation of the tested object enriched with information about temperature. These types of models are used to identify damaged parts of the object, determine the size of leaks (e.g. heat and/or air leaks) or perform thermal analyses. In the classic approach, modeling of temperature distribution is performed in two-dimensional space using a set of quasi-continuous thermographic images with a colored scale or a set of point data with temperature values measured at a specific time. In the latter case, we can talk about the integration of spatial data (regarding the location of points, obtained by geodetic techniques) with the results of point measurements of temperature made by physical techniques (e.g. contactless pyrometer). The combined data sets can be further analyzed, e.g. using geostatistical methods, enabling continuous temperature distribution model to be obtained. This article presents the three-dimensional integration of data from two measurement techniques and methods to model the temperature distribution in two-dimensional space. Three-dimensional spatial integration concerned the point cloud acquired by terrestrial laser scanning and thermal imaging. The measurement object was a fragment of the ventilation system in the underground car park. Modeling of temperature distribution in two-dimensional space uses an unusual data source, which are temperature readings from a precision level, obtained during measurements of vertical displacements of a multi-level underground parking structure. These data, combined with images from a thermal imaging camera, enabled performing thematic maps of temperature distribution to be made using the IDW method and ordinary kriging. These maps can help in the future to interpret the values of vertical displacements of the parking structure.