Indexation of Bench Test and Flight Data

Cynthia Faure, Jérôme Lacaille, Jean-Marc Bardet, and Madalina Olteanu
Submission Type: 
Full Paper
phmec_16_054.pdf446.48 KBJune 28, 2016 - 9:23am

A big amount of data are provided every day in Snecma’s test benches. Produced by thousands of sensors, test and flight data represent a big interest for engineers but the manual analysis of particular information is to hard-working. This is why specific data must be extracted from the database. It is not unusual to miss interesting information when focusing on a precise study. Defining the data in a succession of labels where each label appears as transient or stabilized phase would be one way to solve the problem. The start and stop points of the different phases will be computed by an offline change-point detection algorithm. In order to detect potential crucial changes of characteristic variables, it is relevant to develop powerful statistical algorithms. The Pruned Exact Linear Time (PELT) method is a parametric change-point detection which searches the optimal partition of a monovariate signal with a linear complexity. Then on a multivariate aspect, patterns are built with parameters and initial conditions and classified in a specific category with a map/reduce scheme. This classification will allow different analysis: the comparison of different patterns with the definition of a distance and the research of a specific pattern in a large database. For example if an engine shows a specific engine’s temperature pattern after the test pilot change the shaft rotation speed from one level to another, the engineer may ask if this behavior is usual. If not, it should be very interesting to see if such pattern happens in the past on other engines or other tests and dig from the database the old documents related to those rare events and eventually the people concerned. The objective of this project is to progressively score and classify different patterns in an increasing database of labels. The first step was to implement the PELT algorithm. Then it is possible to identify the different transient phases extracted from small subsets of temporal measurements and compute models for each patterns. These codification of transient phase will lead to a classification into labels or topics. After defining enough patterns, a new bench test will be classified automatically in a database of patterns.

Publication Year: 
Publication Volume: 
Publication Control Number: 
Page Count: 
Submission Keywords: 
database; turbofan; bench test; flight data
Submission Topic Areas: 
Data-driven methods for fault detection, diagnosis, and prognosis
Submitted by: 

follow us

PHM Society on Facebook Follow PHM Society on Twitter PHM Society on LinkedIn PHM Society RSS News Feed