Impact of post-processing methods on accuracy of Darcian and Forchheimer permeabilities determination
Résumé
Getting the Darcian and Forchheimer permeabilities is required by a number of applications to characterize porous materials. In the meantime, ensuring the reliability of these numerical values is essential. While a number of studies aim at developing appropriate test benches, the quantitative importance of post-processing methods is often underestimated. Yet, they play a major role in terms of accuracy and reliability, in addition to the test benches and test methods themselves. The present work uses experimental data obtained in the framework of transpiration cooling of high-speed cooled structures. It presents a comparison between six methodologies which are found in the literature to post-process the experimental raw data. Their accuracy is found to be influenced by several parameters (Darcian and Forchheimer permeabilities orders of magnitude, operating range, and flow nature). This work intends to quantify these methods and to suggest which method to choose to minimize the uncertainty. Above 10−17 m2, the Darcian description should be replaced by the Brinkman one and the ISO 4022 method is generally the best processing method (best correlation factor) over the entire range of test. The Multiple Linear regression method shows a clear advantage over the Simple Linear Regression one without rebutting computation cost.