Heat-loss measurement using infrared thermography by multi-threshold analysis

Abstract

Thermal performance assessment of industrial systems is an essential process for optimal energy management, heat loss evaluation, and energy diagnosis applications. This work presents a method based on the advanced Otsu’s multi-threshold analysis of infrared thermographic images for industrial heat loss measurement. A boiler in a rum distillery was used as a case study. The novelty of the proposed approach lies in the application of a criterion to perform the frequency analysis of two-dimensional thermographic images. The thermic energy and heat frequency distribution are obtained from the infrared images of the target object, being integrated for calculating the total heat loss based on well-known heat transfer models. For determining the heat loss caused by radiation, the Stefan-Boltzmann Law has been used. The heat loss due to convection was assessed by applying different approximate empiric models: McAdam’s, F. Michijev’s, C. King‘s, and Pavlov’s, and the heat-transfer dedicated approach of Churchill-Chu which were evaluated by the infrared images parameters processed at Otsu’s threshold value of 15 as optimal. Convergence in the heat loss calculated between F. Michijev’s, C. King‘s, Pavlov’s, and Churchill-Chu’s models was found with an average value of 959 W (+/−36.5 W). However, the Mc Adam’s model presented the lowest value of 775 W being significantly different from the other methods. The presented image processing strategies, combined with the heat transfer equations, can be used as robust and accurate tools to evaluate the heat transport phenomena in the industrial sector using thermographic images.