Hannover Messe 2023

At the Hannover Messe 2023 the Fraunhofer Institute for Large Structures in Production Engineering IGP will present the Digital twin: fault detection on large machines using acoustic sensor data acquisition. Fraunhofer IGP presents the interaction between acoustic sensor system and digital twin. Our sensor system hears when a ship's engine is not running smoothly. On our model of a ship's engine, the trade fair guest can become the malfunction and manipulate it by flicking or knocking. The four sensors placed around the engine hear the problem and transmit the error message directly to the digital twin. Here it is analyzed which fault has triggered the malfunction.

When the motor sounds funny

Experienced machinists hear faults before they become acute. The Fraun-hofer Institute for Large Structures is presenting a holistic method for monitoring large engines at the Hannover Messe. Failures of main engines, for example of large ships, are not only annoying, but above all expensive due to spare parts, labor and delays. In the research project "Acoustic Sensor Network with Real-Time Data Evaluation" three EU-funded joint projects are currently being brought together in order to detect faults in ship engines in good time before critical failures occur in the future.

This includes an intelligent sensor network and fault location. Systems available on the market monitor individual critical points of the machine in order to detect local deteriorations. The ASEDA project (Acoustic Sensor Network with Real-Time Data Analysis) is developing modular sensors with integrated signal processing. Pre-processing and detection of machine defects can take place directly on the trusted embedded system. Interconnected and synchronized in a sensor network, localization of defect noise and thus global instead of local evaluation becomes possible.

Another project deals with pattern recognition with automated training Together with the University of Rostock and S.K.M. Informatik GmbH, the automated evaluation of data from the sensor network is implemented. Up to now, a manual definition of discrete properties in the sensor signals has been common practice in order to classify data based on them. By using intelligent algorithms for signal evaluation, the classification can be automated. Deviations from the normal state indicate, for example, an imbalance or wear. This makes it possible to detect faulty patterns at an early stage, before they are audible to humans.

In cooperation with Logic Way GmbH, a cloud-capable service platform called "IDaS-Sensor-Services" is being developed, which on the one hand secures the measurement data and at the same time has capacities for more complex long-term analyses. In this project, interfaces between the individual subsystems and the user are provided. An intuitive and feedback-driven interface on mobile devices ensures easy usability of the system. In addition, a long-term analysis shows trends within the measurement results. The combination of service-oriented architecture and suitable virtualization creates the basis for service availability across company boundaries and extreme scalability.

Together, a digital twin is created. Through the retrofit, the combination of centralized and decentralized evaluation with the inclusion of human-collected data, behavior patterns can be derived. The corresponding reaction then takes place in the form of maintenance and repair services. The combination of service-oriented architecture and suitable virtualization creates the basis for service availability across company boundaries and extreme scalability.

Visit us:

Hall 16 - Booth A12

30521 Hannover


Konrad Jagusch

Tel. +49 381 49682 - 51


Christoph Heinze

Tel.: +49 381 49682 - 21


Acoustic sensor system

Automatic sound classification with unsupervised training methods.

Fraunhofer IGP

Privacy warning

With the click on the play button an external video from www.youtube.com is loaded and started. Your data is possible transferred and stored to third party. Do not start the video if you disagree. Find more about the youtube privacy statement under the following link: https://policies.google.com/privacy