Health Monitoring Techniques for High Availability Drives

The growing number and need for high availability drives in the modern
world places high importance on reliability and operational availability. Part of the
solution to the increased requirements on availability involves monitoring
degradation associated with machine failure, diagnosing the mode of failure and
predicting into the future to a possible time to failure.
Broad multi-disciplinary knowledge is required to understand drive ageing
and implement monitoring systems. This research reviews the failure modes
associated with high availability drives, in particular with regards to low voltage
random wound permanent magnet inverter driven machines. The main failure
modes are presented, with analysis of available literature regarding machine
failure. Gaps in literature are identified and the most common failure modes of
the low voltage machine, bearing and winding damage are presented.
Currently the area of least understanding in the field of drive health
monitoring involves the measurement and analysis of winding insulation failure.
The modes of insulation failure are investigated in detail and a new model for
ground-wall insulation is developed.
Methods to measure insulation ground-wall insulation have been
reviewed, and a new method has been developed to monitor the ground-wall
insulation health, during machine operation, using common-mode current
inherent in inverter driven machines during operation.
To expand the knowledge of insulation failure, the novel insulation health
monitoring method is used to monitor four stators during an accelerated ageing
test, representing monitoring of insulation of real machine lifetimes. The results
are analysed and additional testing is presented in order to develop the
monitoring system for standard industry drives.
The major conclusion from results indicates that insulation degradation
progression follows a deterministic trend, with clearly defined threshold of
failure, allowing prognosis of lifetime based on monitored parameters. Future
work recommendations are made to expand on the findings during the course of
research.