Because of the strong competition in electric motors manufacturing, manufacturers put more and more effort in noise control. Besides, noise might be a very strong indicator of possible motor errors, if observed properly.
Company is already measuring many properties of electric motor although during the mounting process of electric motors, many mistakes can be done and some of them can not be detected with those measurements. Existing control devices are based on absolute values of noise and vibration (classic approach), which in many cases, doesn't seem to be effective, so use of a psycho-acoustic approach in failure detection, might be worth trying.
In this work, the research was done using 20 sample electric motors. The goal of the research was to determine, whether we can distinguish between good and bad samples, by extracting and evaluating sound features.
|