This paper has presented the framework for a new
expert system based incipient failure detection and
predictive maintenance (FDPM) system for distribution
systems. The FDPM system is comprised of several
expert system related components and databases for use by
the mathematical and neural network models which predict
deterioration of distribution equipment. The distribution
system is monitored continuously for incipient failure
detection. classification, and location. and periodically for
predictive maintenance assessment. The FDPM system
unifies all equipment into a map which is used to locate
the cause of detected incipient behavior or equipment near
end-of-life.
providing efficient and accurate performance.
The initial work shows great promise in