Performing Diagnostics & Prognostics On Simulated Engine Failures Using Neural Networks

Abstract

Networksby Owen MACMANN Good prognostic health management (PHM) solutions for jet engines remain elusive, owing partially to lack of run-to-failure data sets. A good PHM solution has the potential to improve on unscheduled maintenance by offering an accurate, real-time estimation of the engine's current health state. Aero-engine simulations allow for generation of simulated data invaluable for data-driven PHM solutions. Simulated data can characteristically represent propagation of faults in an engine over time and present the results of that fault propagation in terms of realistically acquirable sensor data. A method of data set generation for jet engine degradation that incorporates multiple faults is described. The generated data sets can be used for training a combined diagnostic/prognostic solution. This work proposes a neural network-based prognostic system that uses diagnostic evaluations as additional tag data for a prognostic analysis.Self-organized maps are used to classify data. The classifications are added to the data as an additional input for a neural network designed to predict remaining usable life.The method exhibits totally autonomous learning of data and produces improvements over approaches that do not pre-classify data.iii The author of this work would like to thank several individuals for their assistance in realizing all that is laid here before you: Dr. Kelly Cohen, for his unfailing support and good attitude; Dr. Alireza Behbahani of the Air Force Research Laboratory, for his team's help in providing software, literature, and insights crucial to understanding C-MAPSS40K and the larger problems behind diagnostics and prognostics; and, Dr.Kristin Rozier, for her help in providing a foundational understanding of mathematical logic.ix

An epigenetic modeling approach for adaptive prognostics of engineered systems

“…In [8], the authors implement a dynamic GA to select the best set of features for prognosis of an unbalanced rotor system, while the authors of [10] use a GA to identify precursors to four different failure modes for electro-mechanical actuators, enabling the detection of faults prior to compromise of system performance. Finally, [9] describes the use of a self-organizing map, a type of neural network, for the basis of a data-driven PHM approach for engine failure.…”

A Preliminary Study for Aircraft Engine Health Management based on Multi-Scale Kalman Filters

“…An example result is given in Figure 4. In the data driven approach, Bayesian regression (Zaidan, 2013) and artificial neural network (Macmann, 2016) most representative algorithms. The Bayesian regression combines the two sources of information: historical in service data from the engine fleet population and once-per flight transmitted performance measurements.…”