Job details

Smart Maintenance

Flanders Make

Flanders Make is the strategic research centre for the manufacturing industry. Our mission is to strengthen the long-term international competitiveness of the Flemish manufacturing industry. That’s why we work together with SMEs and large companies on pre-competitive, industry-driven technological research, resulting in concrete product and production innovation in the vehicle industry, the manufacturing industry, and production environments.

Goal of the internship

Manufacturing companies continuously try to increase their productivity by avoiding unplanned machine downtime. All too often, machine operators still base their maintenance actions on data acquired during periodic manual inspections, thereby missing to detect faults as early as possible.
The cost decrease of sensors (e.g. MEMS sensors) and computing hardware (e.g. Raspberry pie) allows for continuous monitoring of machines, enabling to act immediately when anomalies happen. Cloud connectivity of these machines also increases the availability and accessibility of the machine responses. Furthermore, combining model based signal processing and data-driven processing methods enables the development of potentially more robust and more accurate fault diagnosis in a family of rotating machines.
All of these technologies offer great opportunities for the industry to optimize machine maintenance strategies. It is, however, very challenging to decide in which technologies to invest to get maximal return on investment. To clarify the potential, a small-scale plant of 7 rotating machines is built by Flanders Make, allowing to apply these technologies. The following figure gives an overview of the small-scale plant. The plant is focused on the monitoring of bearings as machine components, because they are widely used and very critical for the good operation of rotating machines.

The goal of your internship is threefold :

  1. Perform your own accelerated lifetime test measurements on the 7 rotating machines. You can vary the initial condition of the bearing, speed, load and lubrication flow rate.
  2. Gain insights by analysing and interpreting the measured signals.
  3. Develop an approach to optimally transform raw data into actionable information in order to:
    - Detect bearing failures as early as possible;
    - Obtain a minimum of false positives/negatives;
    - Operate at a minimum cost: computational cost, data transfer, memory usage,

We propose to use a hybrid approach, where model-based algorithms are applied on the measurement data in a first step, and in a second step a data-driven approach is applied to determine :

  • The optimal combination of model-based algorithms;
  • The optimal threshold values for robust fault detection.

Learning targets :

  • Develop hands-on experience on rotating machinery by carrying out measurement in an automated way;
  • Signal processing: model-based and data-driven approach;
  • Develop a sense of industrial relevance of innovative technologies.

Profile student

  • Bachelor degree in mechanical, electrical or mechatronic engineering;
  • Knowledge of mechanical vibrations or experience with Matlab or Python is highly recommended;
  • Passionate by research and new technologies with focus on applications for industrial machines;
  • Result oriented, responsible and proactive;
  • A good communicator, able to communicate in English;
  • Eager to learn and a team player.

Practical data

This assignment is an internship.

The assignment is for minimum 3 months to maximum 6 months and takes place at the offices of Flanders Make located in Leuven, Belgium.
This assignment requires two interns that work simultaneous.
All software and hardware needed for the execution of the project will be provided by Flanders Make.

File attachments

  • Location: Leuven (Belgium), Gaston Geenslaan 8


For more information:
at the number: +32 (0)11 790 590
Please waiting during procesing your request
Mail to friends
Email(s) successfully sent
An error is occured please contact the system administrator