Machine monitoring processes powered by Artificial Intelligence

Researchers at the Swiss Federal Institute of Technology in Zurich (ETH) have developed an Artificial Intelligence (AI) that is able to recognize the sound of a “healthy” machine, further details of which can be found in a press release issued by the university. The AI detects abnormal sounds and is then able to determine whether maintenance is required. Prompt detection of a defect can, for example, avoid a costly machine breakdown.

The machine learning process represents the true innovation of this development from the Greater Zurich Area. It allows a computer to fully learn the wavelet transformation process. This term essentially conveys the way in which sounds and noise can be mathematically depicted as waves. In specific terms, wavelet transformation “decomposes a function into a set of wavelets”, ETH Zurich writes in the press release. The aim is to work out how much of a wavelet is contained in a signal, which in turn helps the computer to extract the required information from such sounds.

However, the development can be used for other purposes aside from machine monitoring processes. For example, the researchers are already experimenting with using their AI to recognize bird calls. “With our research, we were able to demonstrate that our machine learning approach detects the anomalies among the sounds, and that it is flexible enough to be applied to different types of signals and different tasks”, concludes Olga Fink, Professor of Intelligent Maintenance Systems at ETH Zurich, in the press release.