Diagnosis of a drum motor using mathematical modeling and signal analyzing techniques

In this paper, the development of a condition monitoring method for a drum motor is presented. A mathematical model of the motor is developed. This model was designed for description of the motor's dynamic behavior and for implementation of several wearout and defect phenomena. Using this model, a good understanding of such phenomena was achieved by carrying out simulation runs using different operational and wearing conditions. Several experiments at a physical demonstrator device were carried out additionally measuring mainly the motor's phase currents under different conditions. Using this information, we were able to extract several statistical features by performing proper orthogonal decomposition. These features were used for classification of different wearout and defect states by p erforming appropriate subspace operations which yield adequate data for a Fuzzy Pattern classification algorithm. As a result we were able to identify different reference states as well as previously unknown states.