ENBIS Spring Meeting 2017

28 – 30 May 2017; Monastery of Schlägl in Upper Austria Abstract submission: 11 November 2016 – 5 March 2017

Predictive Maintenance of Turbofan Engines

29 May 2017, 13:45 – 14:45

Abstract

Submitted by
Alessandro Tarchini
Authors
Simone Lombardi (MathWorks)
Abstract
Engineering and IT teams are using MATLAB to build today’s advanced Big Data Analytics systems ranging from predictive maintenance and telematics to advanced driver assistance systems and sensor analytics. Teams select MATLAB because it offers essential capabilities not easily accessible in business intelligence systems or open source languages.

Through a series of examples, we will demonstrate several MATLAB capabilities in various domains:
- Physical-world data: MATLAB has native support for sensor, image, video, telemetry, binary, and other real-time formats. Explore this data using MATLAB MapReduce functionality for Hadoop, and by connecting interfaces to ODBC/JDBC databases.
- Machine learning, neural networks, statistics, and beyond: MATLAB offers a full set of statistics and machine learning functionality, plus advanced methods such as nonlinear optimization, system identification, and thousands of prebuilt algorithms for image and video processing, financial modeling, control system design.
- High speed processing of large data sets. MATLAB’s numeric routines scale directly to parallel processing on clusters and cloud.
- Online and real-time deployment: MATLAB integrates into enterprise systems, clusters, and clouds, and can be targeted to real-time embedded hardware.

A case-study about how to use MATLAB to build a predictive maintenance system for turbofan engine for aviation applications will be shown. Particularly, this case-study will be focused on the development of a predictive maintenance system to predict when engine failures will occur on engines based on their live sensor data. The data used for the development of the predictive model are provided by NASA PCoE and acquired from 100 different engines of the same model. The predictive model developed can classify the engine status in urgency classes of maintenance, which are defined in accordance with the expected number of flights before failure.

Return to programme