IEEE Power & Energy Magazine - May/June 2017 - 91

may/june 2017

ing a digital computer, or by any other
means, for that matter. in his opinion,
the main value of our effort was not the
replication of the blackout events but a
validation of the computer software and
study techniques that were then coming into widespread use. i was quite
surprised. While we were conducting
our study, neither i nor the committee

members knew that we were attempting something that had never before
been accomplished. Perhaps that was
just as well!

Consequences of
the 1965 Blackout
blame for the blackout should not
be attributed to poor planning or

making life visibly safer

SpanLite

Self-Illuminated
Power Line Marker
Meets FAA Advisory
Circular 70/7460-1L
(Dec. 2015)
Installs directly on
live lines up to 500 kV

Bird
Diverters

LED Obstruction
Lights

Low Line
Flags & Markers

making life visibly safer
800-722-8078 * pr-tech.com

ieee power & energy magazine

6709

Charles Concordia of the general
electric Company (ge), one of the giants in the science of electric power systems, was a member of this committee.
Concordia, on the behalf of ge, offered
the pro bono use of a newly developed
transient stability computer program.
Consistent with the technologies of the
mid-1960s, the software used punched
card input and paper output. the maximum number of nodes or buses was
500, which was considered state of the
art at the time. Since the program was
brand new, the subsequent study uncovered a fair number of bugs; as a side
benefit, the analysis served a kind of
beta testing for the program.
i was appointed chair of the modestly named Computer Committee to conduct this study. We were to do our work
without reference to what had actually
happened, relying only on the status of
the system at the beginning ^t = 0 -h,
the trip times of the five lines, and the
relay characteristics of all the highvoltage transmission lines in the northeast. it was a very tedious effort. the
automated features common in today's
stability programs were conspicuous by
their absence. each simulation had to be
run for a given time interval. the apparent impedances of each transmission
line were compared to their relay characteristics, and any line where tripping was
indicated identified. then the case was
rerun, that line removed at the appropriate time, and the simulation carried a bit
further. this process had to be repeated numerous times. Once completed,
though, the result was an almost exact
replication of the disturbance. virtually
every step in the complex sequence of
events had been reproduced. figure 4 is a
plot of generator angle versus time from
the 9 november 1965 computer simulation study. note how two generators separate from the others by well over 180°
as time goes on. When generator angles
separate by 180°, system instability is
virtually guaranteed.
When we reported our results at a
meeting of the blue ribbon committee,
one of its members pointed out that no
one had ever successfully duplicated
a large power system disturbance us-

http://www.pr-tech.com

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - May/June 2017