Fault Prediction in Aircraft Engines Using Self-Organizing Maps

Abstract

Abstract. Aircraft engines are designed to be used during several tens of years. Their maintenance is a challenging and costly task, for obvious security reasons. The goal is to ensure a proper operation of the engines, in all conditions, with a zero probability of failure, while taking into account aging. The fact that the same engine is sometimes used on several aircrafts has to be taken into account too.The maintenance can be improved if an efficient procedure for the prediction of failures is implemented. The primary source of information on the health of the engines comes from measurement during flights. Several variables such as the core speed, the oil pressure and quantity, the fan speed, etc. are measured, together with environmental variables such as the outside temperature, altitude, aircraft speed, etc.In this paper, we describe the design of a procedure aiming at visualizing successive data measured on aircraft engines. The data are multi-dimensional measurements on the engines, which are projected on a self-organizing map in order to allow us to follow the trajectories of these data over time. The trajectories consist in a succession of points on the map, each of them corresponding to the two-dimensional projection of the multi-dimensional vector of engine measurements. Analyzing the trajectories aims at visualizing any deviation from a normal behavior, making it possible to anticipate an operation failure.However rough engine measurements are inappropriate for such an analysis; they are indeed influenced by external conditions, and may in addition vary between engines. In this work, we first process the data by a General Linear Model (GLM), to eliminate the effect of engines and of measured environmental conditions. The residuals are then used as inputs to a SelfOrganizing Map for the easy visualization of trajectories.

Aircraft Engine Fleet Monitoring Using Self-Organizing Maps and Edit Distance

“…It is not valuable to use the rough engine measurements: they are inappropriate for direct analysis by Self-Organizing Maps, because they are strongly dependent on environment conditions and also on the characteristics of the engine (its past, its age, ...). The first idea is to use a linear regression for each engine variable: the environmental variables (real-valued variables) and the number of the engine (categorical variable) are the predictors and the residuals of these regressions can be used as standardized variables (see [3] for details). For each engine variable r = 1, .…”

“…This article follows another WSOM paper [3] but contains necessary material (and possibly redundant) to be self-contained. It is organized as follows : first, in Section 2, the data and the notations used throughout the paper are presented.…”

Advances in Data Mining. Applications and Theoretical Aspects

“…It is not valuable to use the rough engine measurements: they are inappropriate for direct analysis by Self-Organizing Maps, because they are strongly dependent on environment conditions and also on the characteristics of the engine (its past, its age, ...). The first idea is to use a linear regression for each engine variable: the environmental variables (real-valued variables) and the number of the engine (categorical variable) are the predictors and the residuals of these regressions can be used as standardized variables (see [2] for details). For each engine variable r = 1, .…”

Dynamic Data-Driven Prognostics and Condition Monitoring of On-board Electronics

“…They built a topological model of the network and used their ANN algorithm successfully for fault detection and isolation [33]. Cottrell et al (2009) used self organizing maps for fault prediction in aircraft engines. The authors designed a procedure to visualize successive data measured on an aircraft engine, and use self organizing maps to project multi-dimensional data and track the measurements over time.…”