Automated detection of rotor faults for inverter-fed induction machines under standstill conditions

It is difficult to apply conventional online motor current signature analysis techniques for diagnosis of rotor faults for closed-loop induction motor drives for many applications due to the masking effect of the feedback current controller and/or variable frequency or load operation. Relying solely on traditional offline inspection techniques during regular maintenance does not allow frequent monitoring of rotor problems and is inconvenient since it requires rotor disassembly and/or manual rotor rotation. In this paper, a new automated technique for testing voltage source inverter-fed squirrel-cage induction machines at a standstill for rotor faults is proposed. The main concept is to use the inverter to excite the machine with a pulsating field at a number of angular positions to observe the variation in the impedance pattern due to broken rotor bars, whenever the motor is stopped. An experimental study on a 7.5-hp induction motor verifies that broken bars can be detected with high sensitivity and reliability. It will be shown that the proposed method can provide automated and reliable assessment of rotor condition frequently without motor disassembly, manual rotation, or additional instrumentation. The proposed test can also provide rotor quality assessment independent of variations in motor or load operating conditions since it is a standstill test.