Flight data driven quantification of the performance deterioration accumulated during a particular flight is still a challenging task. The evolution of the measured parameters during one flight is small compared to measurement uncertainty. Furthermore, it is a function of a multitude of parameters describing the operational environment of a particular flight mission.An approach to solve this challenge is investigated using a defined numerical experiment. This comprises multiple variations of a carefully chosen flight mission of a typical single aisle aircraft. The individual flights are simulated with a resolution of 1Hz. A set of non-dimensional groups has been derived collapsing the large number of parameter to a limited set of parameters of physical relevance. The non-dimensional loss of compressor leading edge material has been chosen as the parameter expressing flight mission severity.A Monte Carlo simulation has been carried out with the probability density functions of several operating conditions being fitted to real data. The result of the Monte Carlo simulation confirms the general observation that the spread of possible operational conditions experienced on one flight mission can be significant. This spread is dominated by a few non-dimensional groups with others being of lower significance. It is the non-dimensional groups of high significance, which need to be known well for any assessment of flight mission severity.The classification according to severity is carried out using Principal Component Analysis in combination with a Support Vector Machine algorithm. This combination is capable to classify the multitude of simulated flights into groups of different degree of severity. The required amount of training data sets as well as the optimum amount of classes of severity is discussed with the rate of classification and misclassification being the criteria.The robustness of the classification with regards to measurement uncertainties is discussed by adding several degrees of measurement uncertainty on top of the simulation data. It turns out, that measurement uncertainty lowers the rate of classification and increases the rate of misclassification. Still the general approach is viable.

Data-Driven Analysis of Engine Mission Severity Using Non-Dimensional Groups

Paper Proceedings: https://asmedigitalcollection.asme.org/GT/GT2021/volume/84898

Turbo Expo 2021

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.