IEEE Power & Energy Magazine - January/February 2016 - 125

in my view (continued from p. 128)

and information systems." Thus the problem of the need to only use one vendor's
software, which had been around for a
while, was to be dealt with. EPrI also
included a reference to a comprehensive
information model, described in the T&D
World Magazine news release as providing "a common definition of the electric
system data, in object form, that is used by
the various power system applications."
The EPrI ccAPI project got off the
ground due to efforts of Herman Amelink, then of Energy control consultants [later part of Kema and now part
of Det norske Veritas (norway) and
germanischer Lloyd (germany)], and
gerry cauley of EPrI (now cEo of
the north American Electric reliability
corporation). cauley was the original
EPrI ccAPI project manager. Macro
corp. was the contractor to EPrI for the
project with James resek from Macro
as the manager. Later David Becker
took over the project at EPrI, and Terry
Saxton from KEMA consulting took
over as manager. Ed Dobrowolski, then
at PEco, was an advisor to EPrI. Jay
Britton, Jim resek, and Terry Saxton all
worked on the project with EPrI.
The transition of the EPrI ccAPI project to the now-familiar cIM
framework occurred during the years
of the EPrI project itself. To allow the "plug and play" capabilities
sought by both the EMS Architecture
Task Force and the ccAPI project,
the EPrI teams began to design an
application program interface (API).
Instead of concentrating on the API
details that have to do with a vendor's computers, operating system,
etc., the team began the effort to develop the framework for the meaning
and structure of the information exchanged between the EMS database
and various applications. According
to Jay Britton, "ccAPI migrated first
toward semantics driven by EMS requirements and then away from an
internal EMS architecture focus and
january/february 2016

toward exchange of information between control centers."
In the beginning, it was clear that
the power industry itself would have to
be the source to develop the semantic
model to describe the kind of data to be
used in EMS applications. Jay Britton
presents two questions:
✔ where do we to
get the semantic model?
✔ what do we do
with a semantic
model?
The first use of
the cIM was to allow
EMS operations staff
to exchange network
models between itself and neighboring
network operations
staff. But this is not
an API. cIM is not an
API that allows one
to build a power system EMS application
and "plug" it into any
compliant EMS.
A parallel effort
that also contributed
to the cIM development was the EPrI
operator Training Simulator (oTS)
project, which had the goal of developing a training simulator that could be
used by any EPrI member. The simulator was built by control Data corporation's EMS group in Plymouth, Minnesota, and PEco was the first EPrI
member for which the simulator would
be built and installed. Ed Dobrowolski
provided the back story on this parallel development and how it came to influence cIM. "The requirements from
EPrI were that the project be transportable to other utilities and not just
usable at PEco Energy which was the
host (I was project manager at PEco),"
he said. "Therefore the oTS team had
to develop a set of models that were not
proprietary and a database for the oTS

to run off of that could be reused by
other utilities."
Thus the EPrI oTS database became the first structure for the cIM.
At this stage, the cIM comprised an
MS Access database and entity relationship diagram that grew to cover
an entire wall. The effort now was
to build a semantic
model. A semantic
model, according to
Terry Saxton, is "a
vocabulary of basic
terms representing
real-world objects, a
precise specification
of what those terms
mean and how they
relate to each other."
The power of such
a model representing the entire electric utility operations
domain can scarcely
be overestimated,
paving the way for
the cIM to become
the key to achieving a model-driven
interoperability solution for integrating the many disparate
systems needed to operate and manage
the electric grid.
The Unified Modeling Language
(UML) was chosen as the language
to represent the cIM semantics. The
conversion of the cIM model to UML
was in many ways the birth of the
cIM model as we know it today-a
UML model that can be directly used
in electronic form by software tools
to generate artifacts (like cIM-based
XML message payloads and power
system network models) that vendors can download and incorporate
in their products.
Those who met at EPrI now realized
that a much bigger effort would be required
and that the end goal should be a series
of international standards; consequently

The transition
of the EPRI
CCAPI project
to the nowfamiliar CIM
framework
occurred
during the
years of the
EPRI project
itself.

ieee power & energy magazine

125

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - January/February 2016