Application of Multivariate Analysis to FIA Combustibility Test on Heavy Fuel Oils for Marine Diesels

Abstract

FIA combustion chambers have been used broadly for the testing of the combustibility of various bunker fuel oils. Their ultimate goal is, of course, discriminating good fuels from trouble inducing fuels, but so far not very successfully. There is a lack of a quantitative evaluation method for the combustion process inside them. In this study, multivariate analysis was applied to the heat release curves. For the first time we can characterize the minute differences between the trouble inducing fuel and the normal one. As a result, with 10 variables, much more precise discrimination of good fuels was realized. The reliability high was 95%. However, this figure cannot be trusted fully owing to insufficient samples.