Smart acoustic sensors minimize machine downtime

March 28, 2019 //By Christoph Hammerschmidt
Smart acoustic sensors minimize machine downtime
Interference noise in industrial plants is usually a sign that something is not running smoothly on a machine. The Fraunhofer Institute for Digital Media Technology IDMT (Oldenburg, Germany) is transforming such noise into a useful application: Using sophisticated algorithms, they analyze the noise and can thus avoid machine failures by predicting them - long before they really happen.

In industrial conveying technology, axial piston pumps are often used to lift heavy loads. An unforeseen failure of these widely used systems is usually associated with high costs. Researchers at Fraunhofer IDMT have therefore chosen the axial piston pump as a practical example for the acoustic monitoring of machines. If it no longer runs completely trouble-free, the experts can recognize this from their noises.

Locally installed, battery-powered sensor systems pick up pump noise via the air or as structure-borne noise, compare it with reference audio data and forward the collected information as metadata. This retrofittable solution allows not only a basic statement about the condition of the machine, but also the identification of various types of problems, such as bearing play or problems with the hydraulics, before major damage occurs. The maintenance technician can use the information to start maintenance work in a targeted manner, which saves a great deal of time and money.

At the upcoming Hanover Fair, the institute will present its development using a demonstrator. This allows the wireless sensor nodes to be configured on a tablet. Recognized acoustic events are visualized on a dashboard. In practice, the system can be easily retrofitted and integrated into existing communication and sensor infrastructures.


Vous êtes certain ?

Si vous désactivez les cookies, vous ne pouvez plus naviguer sur le site.

Vous allez être rediriger vers Google.