| As world economy continues to globalize, increased competition has led to a reduction in profit margins. To remain profitable in such a competitive environment, an industry needs to reduce the downtime of critical components even for maintenance purposes. However, this should not jeopardize the genuine maintenance needs of the equipments. In other words, an optimized maintenance strategy is needed in which, maintenance is performed only when a need for it arises; unlike the traditional periodic maintenance. Condition based Maintenance (CBM) is one such strategy.;Induction machines are work horse of an industry. Their criticality to industry may be gauged by the fact that they account for more than 60% percent of the energy consumed in USA's manufacturing sector. Hence, CBM for induction machines makes perfect economic sense. Although CBM is a three step process, first step i.e. fault diagnostics or identifying a fault and determining its severity, has been studied most widely for induction machines. It is for this reason that literature is abounds with techniques for detecting various fault conditions at an incipient stage in induction machines. Rotor imbalance is one such fault condition. Several researchers have studied rotor imbalance fault in induction machines. They have identified the signatures to look for in the stator current to detect it. However, an accurate explanation for appearance of these signatures is lacking. Moreover, in majority of the papers, only one phase of stator current has been used for detecting and classifying the severity of rotor imbalance. Since current sensors are available in all the three phases for protection and control purposes, combining fault information from all of them may yield more accurate results.;Therefore, this thesis focuses on providing an accurate explanation for the appearance of rotor imbalance signatures in stator current and developing a comprehensive rotor imbalance diagnostics scheme that incorporates sensor fusion. |
