Turbofan main performance characteristics is its thrust. The engine is sold for a given thrust and cannot be deliver under a minimum threshold. Hence it is fundamental to evaluate thrust with precision. However, reception tests when all engine functions are verified before delivering to each airline company are realized in different bench test cells, under different weather conditions, with different pilots, and so on. All those context variations imply that the measurement is far to be normalized. Moreover, the certification process proposes computation of a marginal thrust which is the relative difference between standard thrust and a reference value. The standard thrust is obtained from the measurement in the bench test reported to normal standard ISO conditions.
Even after releasing the acquisition context constraints we still observe that the thrust margin is widely dispersed due to the test conditions. The dispersion of the thrust margin results from the complexity of the test conditions such as: complementary technical adaptations, benches and sites. Thus, it has been found that the measurement of the thrust margin is particularly influenced by certain components such as secondary nacelle and the bench itself.
One of our objectives is to make the thrust margin independent of the test conditions and to reduce its dispersion. This task is complicated by the fact that engine parts comes from different providers and we also try to follow the production trends of each part supplier independently.
This paper describes a fast solution to normalize engine thrust computation and identify specific suppliers production drifts.