eResearch Report 2017 - 2018 - Page 29



Supporting the research endeavour
this against what the conditions
of the power plant theoretically
should be,” explains Fuls. “It gives
the operator and engineers an
indication of the condition of
the plant and early warnings if it
looks like things may go awry.”
EtaPRO also facilitates pattern
recognition, using data collected
over years to identify trends.
Giving students access to such
data on power plants will be
very valuable for research and
training: it means they can
start playing with the data and
developing useful algorithms to
streamline processes while still
in the university environment.
UCT eResearch is providing valuable
support in the form of a server to
house the EtaPRO platform and
data. Once licensing is finalised,
students will be able to access the
platform by simply logging in.
“It is not necessarily a bigdata environment, but the
server will be critical for the
EtaPRO platform and display,”
Fuls explains. Looking
forward, Fuls says the group
will likely be moving into the
big-data space as they work
more on pattern recognition
and machine learning.
This may well mean great
collaboration with eResearch
down the road.
27





Powered by


Full screen Click to read
Paperturn flip book system
Search
Overview
Download as PDF
Print
Shopping cart
Full screen
Exit full screen