As the need for on-line monitoring methods has grown, numerous techniques have been developed for monitoring equipment using accelerometers. When using these techniques to monitor on-line, however, most methods require the sensing device to remain aligned in a specific direction. This is not practical in most industrial settings, especially in situations where the accelerometer must be removed and re-attached frequently. Therefore, a monitoring technique must be developed that is able to function independent of the orientation of the sensor. In previous work by the author, it was theoretically demonstrated that the eigenvalues of the modal energy matrix that is derived from a multivariate autoregressive model can be used to monitor acceleration signals such that the index is independent of the orientation of the sensor. This paper further develops this technique to use the eigenvalues of the corrected spectral matrix and verifies the abilities of the monitoring algorithm through the use of experiments involving the rotation of the sensor and the rotation of the signal source. The tests demonstrate the capabilities and adaptability of the technique, thus allowing the sensor to be removed and reattached quickly and effortlessly. The results also indicate the possibility of using the technique for accelerometers that are used within handheld probes.
|Original language||English (US)|
|Number of pages||5|
|Journal||Proceedings of the International Modal Analysis Conference - IMAC|
|Publication status||Published - 2001|
All Science Journal Classification (ASJC) codes