Experiments that simulate bearing failures and the presence of cavitation are performed on a pump loop. During the experiments several types of data are acquired such as vibrations, system pressure, flow rate and motor current. A multivariate analysis is then applied to these data. Abnormalities are detected by evaluating the residual between the measured value and the estimated value that is determined from the model in a normal state. Because the influence of bearing failures is limited to a very loc...
英字タイトル:
Evaluation of a Multi-Sensors Condition Monitoring Method