can we rely
on the grid?
Before investing in new power transmission technologies, we must first examine what we need in terms of electric reliability.
By Lawrence T. Papay
What does
it take for most people to notice electricity? That question is something
of a mind-bender, like asking what is the color of the wind. That's because
it now takes some extraordinary absence of power—like the Northeast
blackout of August 2003—for most people to recognize the importance
of ubiquitous electricity in everyday lives.
That doesn't make electrification any less of a marvel. When the National
Academy of Engineering listed the top 20 engineering triumphs of the 20th
century, electrification was ranked as the No. 1 achievement. The electric
power grid is certainly one of the most—if not the most—complex
and marvelous machines ever created. But taking that as a given, it also
leads to certain problems that this marvelous machine can and will experience.
For example, it is a machine that was not created all at one time, but
on an as-needed basis. This means there are parts of the grid that are
very old (upward of 100 years). And it has certain Rube Goldberg attributes,
in which pieces have been and are being added to an already running system,
and may have an impact on the delicately balanced structure of the grid.
 |
 |
| Before and after: The 2003 blackout (seen below in an August 14 weather satellite image) turned off the lights in New York City, Toronto, Detroit, and Cleveland. |
In the wake of the Northeast blackout, there were widespread calls for
upgrading the power grid to replace aging and jury-rigged elements. But
before we as a nation embark on a crash program to invest in new power
transmission technologies, we must first step back and examine what we
need in terms of electric reliability.
Electric power is a "just-in-time" commodity. That is, electricity
is produced as it is needed. As a consequence, imbalances between supply
and demand are constantly adjusted for by the amount of generation being
provided. But a stable state can be maintained only if there is adequate
grid capacity—transmission—to deliver the power being produced
to the consumers.
The machine that is the electrical grid represents an enormous investment
over time. Historically, investments in transmission are made only to
add new assets to meet a growing population and a growing appetite for
electricity. This problem has been compounded because after deregulation
the question of who pays has gone unanswered to a large extent. Thus,
in recent times investments were not made in needed upgrading and expansion
of the nation's transmission grid.
To this end, as a result of two power outages in the western United States
in 1996, the U.S. Department of Energy formed a Task Force on Electric
System Reliability. In its final report, issued in September 1998, the
task force considered the question of reliability in light of the movement
toward a deregulated world.
Appropriate steps needed to be taken, the task force recommended, and
taken soon, as the transition to a deregulated system occurred. In fact,
the report stated that "the primary challenges to bulk-power system
reliability are presented by the transition itself, rather than by the
end state of competition. Failure to act will leave substantial parts
of North America at unacceptable risk." The August 14 blackout proved
that the task force was prophetic in its observations.
Competition has put an undue stress on the grid. Historically, a utility
grid was primarily intended to deliver electric power from a utility's
generation resources to its customers. Deregulation has increased the
flow of electricity into, out of, and across grids to a level that was
not considered in the original design of the system and the analysis of
its stability.
In his testimony to the House Energy and Water Committee shortly after
the August 14 blackout, Linn Draper, chairman and CEO of American Electric
Power, pointed out, "In the five-year period during which wholesale
electric competition first gained momentum, the number of wholesale transactions
in the U.S. went from 25,000 to 2 million—an 80-fold increase."
the 'weakest link' phenomenon
Deregulation also means that the reliability criteria that have been developed
are adhered to on a voluntary basis. This has led to a "weakest link"
phenomenon within a grid's structure. The recent outage in the Northeast
demonstrated that a weak link could bring down the grid. It is to the
credit of several neighboring utilities that their individual grids were
well instrumented and controlled so they could recognize the problem and
take appropriate action.
 |
| Severe weather can put stress on the grid. Here, ice from a winter storm has coated utility lines, causing them to sag and creating local blackouts. |
Obviously, an infrastructure that is aging and not designed to carry
out the transactions being demanded of it in today's world will require
a great deal of money to upgrade and enhance. Estimates are that more
than $60 billion will be required. But who should carry out this upgrading
and enhancement, and how will it be paid for? Should utilities implement
the upgrades? Or the independent system operators? Or the regional transmission
operators? Should the cost recovery be through standard utility rate structures
determined through state or federal rate proceedings or as service charges
on individual competitive transactions?
Such questions, all good, haven't even begun to be addressed.
But what specifically is to be done, and what technologies exist, to accomplish
improvements in the reliability and security of our national grids? Significant
work needs to be done in three major areas: transmission; substations
and their associated equipment; and system operations, including sensors,
controls, communications, SCADA, energy management systems, and the like.
Transmission lines can be improved in a number of ways. Restringing the
lines themselves with larger conductors or bundling a number of small
conductors can increase the current-carrying capability of the lines.
Likewise the voltage for a given transmission line can be increased, but
it may entail the installation of new towers or restringing the lines
to ensure that proper ground clearances are maintained. Similar upgrades
can be made on buried conductors, or cables. But unlike overhead lines,
in which heat loading is dissipated into the atmosphere, buried cables
have to sink excess heat into the ground, which may be a problem in certain
situations.
solid state control
The construction of new transmission lines is an attractive alternative.
However, the approval process for siting new transmission lines is often
more than twice as long as that for a new base-load coal-fired generator.
From planning through construction, building a new transmission line can
take more than nine years.
The most important "new technology" for being able to increase
power flows over existing lines or cables lies in the use of so-called
FACTS (flexible ac transmission systems) equipment that can be installed
at existing substations. The various FACTS devices are based on the use
of solid state power electronic controllers and thyristors. They provide
fast-acting control capability to allow greater control of power flows
(eliminating parallel path or "loop flow"), loading of transmission
lines closer to their thermal limits, greater power transfer capability
(reducing reserve requirements), prevention of cascading outages (by limiting
failure consequences), and damping of power system oscillations.
FACTS devices are derived from technology developed in the 1960s for high-voltage/direct-current
(HVDC) applications, and have been introduced into ac systems on a limited
basis over the past 10 to 15 years. The devices can range from static
VAR (reactive volt-ampere) compensators, static synchronous compensators,
static synchronous series compensators, thyristor controlled braking resistors,
to series capacitors or reactors, thyristor controlled voltage regulators,
phase shifting transformers, and unified power flow controllers. In addition
to the control capabilities themselves, the devices are electronic in
nature and, as a consequence, can act much faster than the current state-of-the-art
electromechanical devices, such as circuit breakers and switches.
The various components needed to upgrade substation technology in this
way can be added through a replacement program. Piece by piece, the entire
system can be modernized.
In terms of control of power flows within the grid at a local level, advanced
SCADA (for "supervisory control and data acquisition") systems
are available and can be used at power plants' switchyards and substations
to provide better local control and more intelligent information to the
system operator. SCADA systems can handle local voltage and frequency
fluctuations caused by the failure of a network component—whether
generator, transformer, or transmission line—through active switching
to bring supply and demand back into balance. Generally, the higher-level
SCADA systems are programmed to automatically take steps to drop load
and sectionalize the grid to regain stability. SCADA systems can be installed
at the substation level of the grid or at a higher level.
But there is one more area where grid reliability can be improved: the
"intelligence" by which the grid is designed, monitored, and
controlled. By this, I mean the use of advanced models, sensors, control
systems, and communications to better know and understand the state of
the grid, its weakest links, and what contingencies are possible. Existing
models do not adequately allow for simulation of the more complex interactions—such
as interregional coordination, system planning, and congestion management—that
are involved in the modern, deregulated transmission environment of today.
In a post-9/11 world, such simulation capability would also allow for
better threat assessments and vulnerability analyses as well.
 |
| Much of the mechanical equipment at substations like this one can be replaced by more reliable solid state devices. |
There are a variety of ways this can be improved. Let's start with energy
management systems. Also known simply as EMS, they take the data provided
from the grid in real time and carry out computations of the state of
the system, including estimates of contingencies and potential problems
in management and control of that grid as well as economic dispatching
of generation and the handling of power flows across the grid. The problem
has been compounded by the fact that as individual electric utility grids
have been folded into regional grids, the number of data points involved
and the number of nodes modeled have increased enormously. So state estimation,
which once had been close to real time, now can take minutes to be calculated.
In December, the Pennsylvania-New Jersey-Maryland independent system operator
started up a new state-estimation program that can handle up to four times
as many nodes as it had modeled previously. This is a great start. As
part of the master plan for EMS programs, interface protocols for operators
must be developed so that an open exchange of data and information is
possible.
A second component of overall grid management is the data being transmitted
to the control center and used in the state estimation. Given where technology
is today and the complex machine we are controlling, we need to turn currently
dumb sensors into local intelligent agents that really become two-way
communications devices for quasi real-time information flow and control.
The state of the art in this area is moving very rapidly and making more
detailed information available to control system operators and their programs.
'self-healing' grid
Eventually, what will be needed is an intelligent, adaptive grid that
is "self-healing." By intelligent and adaptive, I mean that
the grid will recognize a series of events and control algorithms will
automatically cause the grid to fail gracefully by forming appropriate
power islands in which generation and load are matched. To accomplish
this kind of control requires that the energy management system run a
state-estimation program faster than real time using information coming
from intelligent agents, and that it have purely electronic control mechanisms
at its disposal. In this manner, given a catastrophic event (or simultaneous
terrorist attacks), the grid will make the optimal choice to keep as many
customers as it can online while mitigating the effects of failure.
In a post-9/11 world, this requirement will be a must if the goal is to
mitigate the effects of simultaneous attacks that have as their objective
to paralyze the electric power system.
Unmentioned in this discussion so far is the real need to ensure the cyber
safety of the grid diagnostic and management systems. The recent blackout,
at least in part, is traceable to the failure of an individual utility's
process computers. Any attempt at blind control has been and would be
disastrous. A system secure from intrusion is mandatory. And, given the
ramifications of a massive blackout, redundancy is a must.
Systems of the future must be secure from both external and internal cyber
attacks.
Seen as a whole, upgrading the national electricity grid is a huge undertaking.
But it must be done if we are going to be able to avoid widespread blackouts
like the one in August—or even more devastating outages on a scale
we have yet to see.
Lawrence T. Papay is vice president for integrated solutions at Science Applications International Corp. in San Diego, Calif. Papay is also a nuclear engineer who has served on numerous advisory boards, including the President's Council of Advisors on Science and Technology.