Industrial facilities that plan the shutdown of equipment for service have large financial savings compared to those managing unplanned shutdowns. To this end, a variety
of algorithms have been developed and published in the literature that can monitor
a machine's health and indicate when the machine starts to develop a fault. In order
for such algorithms to be effective, they require raw data collected from machines.
Often this involves the placement of accelerometers and other sensory devices for
measurements of mechanical behavior. It is possible to extract much of the required
information from the electrical signals of the equipment. This is normally a less expensive installation since one only needs access to the lines supplying electric power
to the equipment. If these data acqusistion modules are accessible wirelessly, then
one can monitor all the interfaced equipment from a central location. To successfully
monitor such electrical equipment, a data acquisition unit is required that can sample
on five or six channels simultaneously, depending on the switch gear configuration.
This thesis details the development of an "endpoint" device that samples the
required number of channels to monitor the electrical signals of industrial equipment,
and interfaces to a wireless network. The hardware and software design of the "endpoint" is discussed in detail. Also, the software design of the server that receives the
data from the "endpoint" is presented.
The designed "endpoint" samples up to six channels simultaneously, at a rate of
at least 8 kHz per channel, and a data resolution of 16 bits. The data are then transmitted wirelessly to a central server for processing. The system has been tested both
in a laboratory environment and at an industrial environment. The desired specifications of the "endpoint" have been verified in both environments. Several "endpoints"
have been assembled to form a network and have been tested in a laboratory setting.
This work has resulted in the demonstration that an "endpoint" can be constructed using of the shelf components that is suitable for the continuous health
monitoring of multi-phase electric machines through a wireless network.
| Identifer | oai:union.ndltd.org:tamu.edu/oai:repository.tamu.edu:1969.1/ETD-TAMU-2391 |
| Date | 2008 May 1900 |
| Creators | Goodsell, Douglas Andreas |
| Contributors | Parlos, Alexander G. |
| Source Sets | Texas A and M University |
| Language | en_US |
| Detected Language | English |
| Type | thesis, text |
| Format | electronic, application/pdf, born digital |
Page generated in 0.002 seconds