Análise estatística do coeficiente de desempenho de um sistema de refrigeração convencional e modificado

Abstract

Statistical methods perform an important role in the field of refrigeration, as they enable engineers to identify variables that contribute to improving energy efficiency, resulting in more economical and sustainable systems. Thus, the objective of this work is to demonstrate the application of different statistical methods to analyze a experimental dataset of COP (coefficient of performance) obtained from a comparative study between a conventional and a modified refrigeration system. The Mann-Kendall and Sen’s slope tests, as well as descriptive and exploratory statistics, were used for data preprocessing. For a significance level of 5%, an analysis of variance (ANOVA) was employed to identify the factors influencing COP. In order to ensure the validity of the ANOVA results, assumptions of residual normality and variance homogeneity were checked. Based on the results, the mean was considered an adequate measure of central tendency, with a maximum percentage difference of 0.1% compared to the median, which is smaller than the average uncertainty of the COP values (± 0.94). Then, using the average COP data, in the first case, involving the type of device and setup temperature, it was found that the type of device significantly impacts the COP. In the scenario with the pump on, the COP was 34.2% higher in the modified system compared to the conventional one. In the scenario with the pump off, the COP was 22.9% higher in the modified system compared to the conventional one. In the second case, studying the type of convection and setup temperature in the modified system, it was found that forced convection increased the COP by 13.1% compared to natural convection. Considering the foregoing, it is recommended to adopt the modified refrigeration system and use forced convection to improve energy efficiency.