Fraunhofer IDMT demonstrates its method for acoustic quality inspection at »Sensor+Test 2019«, the trade fair for measurement and sensor solutions.
From June 25th to 27th 2019, the Fraunhofer Institute for Digital Media Technology IDMT in Ilmenau will be presenting a new solution for acoustic quality inspection allowing contact-free, non-destructive testing of manufactured parts and components. The method which has reached Technology Readiness Level 6 already, is currently being successfully tested in practical use together with a number of industrial partners.
Machine downtimes, manufacturing defects, excessive scrap - often the quality inspection of products and processes reach its limits due to the conditions in production halls. One-hundred percent quality monitoring and inspection is very time-consuming and expensive.
At »Sensor+Test 2019«, the Fraunhofer researchers will be showcasing interactive demonstrators for optimizing existing testing methods with the help of acoustic condition monitoring of objects. The method developed by Fraunhofer IDMT combines audio signal processing and machine learning algorithms.
One of the demonstrators will be illustrating how the noise produced by different motors can be measured and analyzed in order to detect out whether a motor runs flawlessly or whether it is defective (or likely to become defective due to motor overload). The second exhibit of the Fraunhofer IDMT is more playful. While indulging in a game of air hockey, visitors will be shown how different prepared pucks – which look and sound pretty much the same to the human eye and ear – can be distinguished in real time by the sound they produce during the game.
Intelligent acoustic measurement technology
”What sounds playful is actually very complex and the result of many years of research and development work”, explains Sascha Grollmisch, expert in machine learning at Fraunhofer IDMT. “The method comprises several steps of measurement and analysis: 1) accurate recording of the sound produced, 2) prefiltering of signal and noise, and finally intelligent signal analysis using machine learning algorithms”.
At its current stage of development, the system still requires large amounts of training data to deliver reliable results. “Our aim is to keep reducing the amount of required training data step by step, while at the same time keeping the system’s recognition rate at a high level. The goal is to end up having a self-learning system that is capable of autonomously assessing the quality of products and processes on the basis of acoustic condition monitoring”, says Sascha Grollmisch.
Feel free to visit us at Fraunhofer’s booth #5-248 to learn more about the intelligent acoustic measurement technology made by Fraunhofer IDMT.