ISSUES WITH DISTRIBUTED GENERATION

If distributed generation is being talked about, but you want to know what it is and what
technical issues are involved, this course may cut through some of the mystery.

INTRODUCTION
If you find it really difficult to identify the differences in the new buzzword “distributed
generation” from the phrases peak-shaving, cogeneration, independent power producer
(IPP) and non-utility generation (NUG) don’t feel alone. In fact, distributed generation has a
lot in common with these other applications. It also has some unique characteristics in the
current usage of the phrase, compared with independent power production. In fact, distrib-
uted generation may be thought to be a subset of independent power production.

To focus in on the aspect of IPP that is new and different from the things we have been
doing under other names for some years, I offer this definition of the term “Distributed
Generation”:

Application of relatively small sources of electric power, placed at or near

the loads being served, and using a variety of energy resources with
focus on “clean and green” energy. Small sources are sources as small
as a generator for a single home, as small as 5 kW in output capacity, up
to generators capable of supplying multiple city blocks of domestic
housing loads. These sources will be tied directly to home or business
power lines at low voltage, or to utility distribution lines at distribution
circuit voltage.

Figure 1: Utility Network

Distributed generation (DGen) is not necessarily tied to a utility power grid by definition.
Many of the practical schemes being proposed or installed are intended to offset the sup-
1
ply of power by the electric utility, but will rely on the utility to supply power beyond the
capacity of the DGen generator. This configuration also provides the DGen with back-up
power in case of a shutdown of the local generation. If the DGen were only operated in an
isolated mode, with no connection back to the electric utility, this session would not really
be necessary. Setting up a small generating plant to serve a local load has been done in
home standby power, portable power, recreational vehicle power, farm power, construction
power, and many others. In other words, there is no really new technical problem pre-
sented by islanded small power generation serving a local load.

Figure 2: Isolated Generation

Only when the generating equipment will be tied in some way back to the electric utility
does the need arise to consider the potential technical challenges with the concept of
DGen.

Figure 3: Subtransmission Network

WHAT TECHNICAL CHALLENGES?

We can identify the challenges that affect both the DGen and the electric utility. These
challenges can be met in practical installations, but with the increasing interest in small
power generation, a major challenge is to scale the cost of the equipment that would be
used on a 50 MW plant down to a price practical for a 10 kW plant. That is not a small task!

2
 Figure 4: Distribution Network

To operate a generator, regardless of size, in a paralleled connection with a utility is further

complicated by the location of the connection on the distribution line or the consumer’s low
voltage power lines. Large power generating equipment is normally connected at trans-
mission or sub-transmission networks. These networks are designed with the awareness of
generator connections and can readily accept power sources tied to them. In the distribu-
tion circuit, the utility generally designs the line as a radial line, meaning that the source of
power is only from one end. When the line needs to be placed out of service, it is only
necessary to trip the distribution feeder breaker at the source of the power, the substation.
If generators become connected to a distribution feeder, opening the feeder breaker will
island any DGen equipment, leaving the DGen connected to other utility customers on the
same line. This possibility creates the technical challenges.

3
 Figure 5: Islanded Dgen

Safety Challenge
This possibility creates a major safety concern for the utility and for the other customers on
the radial distribution feeder. With no control of the generators connected to the feeder,
utility personnel cannot be certain that a distribution line is safe for maintenance or repair.
Safety concerns also exist in case of a downed line where the utility protection system is
designed to open the feeder breaker with a ground fault on the feeder. When the feeder
breaker trips, the DGen may keep the line hot and present danger of injury or death if
people should come in contact with the downed line. It is easy to see that safety must be
incorporated into the design of a DGen connected to an electric utility distribution system.

Figure 6: Hot Distribution Line

4
After safety, comes the protection of other customers of the electric utility who are depend-
ing on that utility to supply reliable high quality power to meet their needs. If the presence
of the DGen on the feeder causes operation at abnormal voltage or frequency, and dam-
age or loss occurs, the electric utility is at risk of liability. It is essential that the DGen make
no contribution to reduced power quality or reliability for other customers of the utility.

Figure 7: Faulted Network

Next, the equipment of the DGen must be adequately protected from faults or abnormal
operating conditions existing on the utility distribution feeder from causing damage to the
DGen. The occurrence of a fault on the utility lines must be anticipated, and the DGen
equipment must be protected from damage exceeding its capability. This means that the
DGen equipment must be capable of withstanding high short circuit currents, high voltage
transients and surges, short term over and under voltage or frequency conditions, and
overloads, which may occur in service. The DGen equipment must be disconnected before
damage occurs under any of these scenarios.

Finally, the DGen equipment may malfunction in some way as to disturb the distribution
feeder power delivery. Action must be taken to disconnect the DGen from the distribution
circuit under any of these circumstances, limiting disruption of the distribution service.

5
Figure 8: DGen Malfunction

Figure 9: DGen Malfunction

6
The protection equipment supplied with the generator breaker and the tiebreaker normally
addresses all of these concerns. Technically, there is hardware available to satisfactorily
provide the necessary protection, and at a satisfactory price level for application to genera-
tors rated in multiple megawatts. The challenge is to find a way to provide equivalent
protection that is affordable for a small power plant. If the concept of small power genera-
tion located at the load is to succeed, it must be economically viable. The cost of the gener-
ating equipment is coming down in diverse fields such as gas turbines, fuel cells, and
solar cells. It will be necessary for the protection to be developed to meet this need for
distributed generation technology to flourish.

Control Challenge
In similar fashion to the safety challenge, controlling the load flow on a paralleled genset
must be addressed. It will probably be the most common desire of the DGen owner to be
able to operate the generator at maximum capacity, continuously, when fuel is available to
do so. To achieve best efficiency, most of the technologies being discussed for this appli-
cation must be operated at rated capacity. If the local load demand exceeds the generat-
ing capacity, parallel operation with the utility would allow for the utility to provide the
additional power requirements. This is not an uncommon arrangement. During periods
when the local load is less than generator capacity, the DGen owner wants to supply the
excess capacity back to the utility. And be PAID!

Figure 10: Typical DGen System

In the past, IPPs operating a plant for their own use negotiated to get the best rate for their
excess power from the utility. The rate was based on “avoided cost” of generating the
power. The interpretation of this term resulted in widely varying rates for IPP operators. In
today’s deregulated market, the price of energy is widely variable, and large IPP operators
are able to collect very high prices for power at peak energy prices. Applying this kind of
spot market pricing to a 10 kW generator would make generating a good return on invest-
ment a lot easier, but the cost to manage such a pricing scheme for such small generation

7
sources would be very difficult to justify. At a practical level, even the installation of meter-
ing to keep track of kilowatt-hours of electricity input and output is expensive for such a
small generator. The incoming kWh meter is, of course, already in place for billing pur-
poses. Adding an exported power meter or changing to a net metering system capable of
recording power consumed minus power generated, when the result of the on-site genera-
tor will be that the utility revenue will be reduced, is not an attractive business proposition
for the utility. And what will be the cost to add kiloVAr-hour metering? Thus, for small gen-
erators, adding revenue accuracy metering to keep track of power generated might be-
come an unbearable cost if the DGen owner must pay for the added metering.

Load Control Example

Generator 280 210 350 0.8

Utility Power
KW KVAr KVA PF

Utility 400 300 500 0.8

Local Gen 0 0 0
Load 400 300 500

Local Generator Power

KW KVAr KVA PF

Utility 0 0 0
Local Gen 400 300 500 0.80
Load 400 300 500 0.80

Reduce Real Power Usage

KW KVAr KVA PF

Utility 200 300 361 0.55

Local Gen 200 0 200 1.00
Load 400 300 500 0.80

Improve Power Factor

KW KVAr KVA PF

Utility 300 90 313 0.96

Local Gen 100 210 233 0.43
Load 400 300 500 0.80

Control Bus Voltage

KW KVAr KVA PF

Utility 400 450 602 0.66

Local Gen 0 -150 150 0.00
Load 400 300 500 0.80

8
In addition, the issues of power control extend beyond measurement but must include a
means of controlling the real and reactive power supplied by the DGen plant. Again, such
a control system is readily available for multiple megawatt generator systems. The chal-
lenge with DGen is the creation of a simplistic control system that allows the owner to set
his generator to run at rated real and (if it pays) reactive power continuously to obtain the
maximum return on the investment in plant and fuel. The development of such a control
system consistent with the size of the DGen plant is indeed a challenge.

Political Challenge
For a healthy relationship to exist, the owner of a DGen and the electric utility must find
mutual benefit. For the DGen owner, the issues are mostly financial. For an investment in
capital, can the DGen plant produce positive net income sufficient to repay the investment
and improve reliability and cost of the electric supply? For the electric utility, can the added
capacity curtail the need to invest in increased generating capacity? If the utility could tap
into a source of additional generation capacity, only when needed, and pay less than spot
power rates, the issues of safety and control of the DGen become manageable. On the
other hand, the utility with more than adequate reserves of power would only see the DGen
owner as competition for the available load with a negative effect on utility revenue. A
successful DGen application will offer benefit to both parties.

TECHNOLOGY SOLUTIONS
Although control and protection products that will meet the demands of this application are
conspicuous by their absence, the manufacturers of these types of products now being
offered for larger IPP applications have the tools in place to get this job done. In Basler
Electric's product line, for example, digital technology using the microprocessor is being
used to increase complexity of the control and protection functions while lowering the cost
and size of product compared to older analog technology products. By taking our experi-
ence in the fields of engine control, generator control, automatic synchronizing, metering,
and system protection, we can see that the means of satisfying the demands of this appli-
cation are available to us. If the capability to communicate is also made available in this
type of product, the DGen owner operating a network of power generating plants may
centrally monitor and control the operation of the network from a single computer in his
office.

Define the Solution

Based on our assumptions about the operation of the DGen power plant, we can begin to
define the solution functionally. In order to minimize the cost of our solution, it is necessary
to try to simplify the solution as much as possible. Room is created to allow for innovative
designs to improve upon our stereotypical solutions derived from large power station
applications. In large power stations, the ability to customize and to meet a broad range of
requirements through programmability is highly desirable. For our small power plant, it
may be more desirable to standardize and reduce programming tasks to help keep the
cost in line.

One area of flexibility will be the need to adapt to a wide range of prime movers and gen-
erators. A gas turbine will require speed control. A solar power supply will require control of

9
an inverter and the charging of a storage battery. A fuel cell will require some kind of oper-
ating controls. Each technology may require a unique solution to meet its needs for protec-
tion and control of the power supply.

Controls for Reciprocating Engine Generator

Taking the most familiar case first, the recip-based genset requires familiar control and
protection. For isolated operation, we need only consider frequency and voltage control,
engine cranking and shutdowns, transfer switches to connect the load, and a circuit
breaker with minimal protective relaying. This simple, non-paralleling application can
benefit from further integration of the electronics into a single microprocessor control with
all of these functions, metering, and communications built in. Additional integration will
bring the costs in line with the needs of this small end of the power generation market.

Figure 11: DGen Recip

The next level of complexity is to impose the requirement to parallel and operate with the
electric utility. This level will add the requirement for programmable regulated loading
control. Added to the isolated operation needs comes the need to identify the control
strategy. One example is to extract as much power from the recip-based genset as pos-
sible and to run the genset based on maximizing the avoided cost of energy or on selling
power when the utility is willing to pay a high-demand price. The controls will need some
means of initiating a start of the genset, automatically synchronizing to the utility system,
and ramping the load up to the desired set point. Adjustment of the operating load levels

10
probably will come from an external source, such as an energy control center. One set
point for real power will control the fuel supply to the reciprocating engine. A second set
point for reactive power will control the excitation power to the synchronous generator.

Figure 12: DGen Recip Genset Controller

One simplification that becomes more attractive as the size of the genset is reduced is to
switch to an induction generator. This type of generator is unable to generate any power
unless it is connected to an external supply of ac power. The induction generator used
reactive power from the external supply for its source of excitation. It can generate real
power back to the power source but cannot generate any reactive power. Today, there may
be no revenue realized by a DGen for supplying reactive power, but there is a growing
trend under deregulation to revise energy rate structures to make reactive power a billable
commodity. When reactive power produces revenue like real power, the use of synchro-
nous generators becomes more advantageous.

One of the interesting possibilities for this type of generation, using the ability to communi-
cate with a genset controller, is the control of the generator, privately owned, by the electric
utility. Some utilities are already writing rate agreements with IPPs, giving discounted rates
in exchange for control privileges. Being able to tap into a source of generation in addition
to the utility generating stations gives the utility a resource that may be exploited in times of
energy shortage. Energy during a peak demand period can be worth hundreds of times
the normal value of energy on the “spot market”, thus making rate concessions to IPPs
justifiable.

11
Instead of running the recip-based genset at maximum capacity any time it is operating,
the utility source may be maximized with the recip picking up any load demand in excess
of some utility limit. This strategy requires more complex controls, becoming more practical
with gensets of large size. One of the complexities is the need to limit the genset to load
within its capacity. Additionally, if the utility power is maximized and the genset power is
also maximized, what can the control system do? Only two choices remain. Allow the load
to suffer by disconnecting some load, by reduction of voltage, or by purchasing high cost
power from the utility in excess of the maximum demand limit.

Controls for Battery Storage-based Generation

Typical of “GREEN” energy supplies such as solar electric cells, fuel cells, and small wind
turbines, electrical power must be generated when the “fuel” is available, and often the
power can be supplied as dc power able to be stored chemically in storage batteries. With
this type of generation, the power is converted to alternating current so it may be conve-
niently used, or it is supplied to loads capable of using the direct current without any con-
version. Storing the energy has the advantage of allowing the generation and the con-
sumption to occur without being linked together. Solar cells, for example, generate electric
power during periods of daylight and at higher levels during bright sunshine. If the load
only needs power at night, if we use electric lights as an example, the battery permits the
generated power to be held until the load demands the energy is used.

Figure 13: Green Energy Supply System

If the load will only use the DGen power by transferring from utility to the DGen source, our
control system need only control the quality of the power as delivered to the load. Protec-
tion may be required of an elementary nature similar to the recip-based genset.

If the DGen is to be operated in parallel with the electric utility, the power must be con-
verted to alternating current, and the voltage and frequency of the power must be con-
trolled to enable real and reactive power supply to be controlled. The efficiency of the

12
battery-based DGen may be relatively constant at widely ranging levels of load. If so, the
control strategy, which seemed advantageous for the recip-based genset of running at full
rated capacity continuously or shutting down, may not be suitable. The battery-based
power may lend itself well to being controlled based on the economic value of the power
being generated.

For our control system design, therefore, I would propose making the operating power
levels, both real and reactive, able to be controlled by external signals. Such signals could
come from a local source of direction via a raise-lower contact able to set the operating
load levels, one setting kW and the other setting kVAr. Another alternative is to use a serial
communications port to allow external commands from local area network or a remote
control point to set the operating point. In either case, some external knowledge of electric-
ity rates is necessary to make the decisions.

Protection for Paralleled DGen Sources

To protect electric utility personnel and their customers from an unexpected hot circuit, it is
necessary first to be able to cut the tie to the utility line any time the utility line is shut down.
By monitoring the voltage and frequency of the DGen output, any loss of utility power is
most likely going to cause a change in real power and/or reactive power supplied by the
DGen. For reciprocating gensets, the natural characteristics of the genset will translate any
change in load into a frequency or voltage change requiring some time for recovery
through the action of the control systems. Therefore, the first form of protection for detecting
loss of utility power is over and under voltage and frequency. These functions can be
helped to detect the loss of utility power if the control systems force a change in voltage or
current as soon as the tie is lost. As long as there is some power flow from or to the utility,
and this power flow goes to zero, a load regulator will recognize the change in load and try
to change voltage or speed to correct for the change, causing a corresponding change in
voltage or frequency.

Figure 14: Minimal Protection Scheme for Parallel Operation with Utility Network

13
These same relaying functions can provide protection for poor power quality by tripping
the DGen system when voltage or frequency falls outside of acceptable limits. By adding
an automatic synchronizing function, the DGen plant operation begins to become auto-
mated, and the protection of the DGen equipment from improper synchronizing is attained.
Other utility customers are also protected from the ill effects of transients from the DGen
being synchronized incorrectly. Protection from faults on either side of the DGen breaker
can be provided by the breaker’s built-in tripping elements or by external Overcurrent
relays wired to a breaker trip coil. And last, the reverse power relay can protect the DGen
from loss of power supply.

CONCLUSION
If the new multifunctional technology of numeric control and protection can incorporate
these functions into an economical package, a contribution can be made to the feasibility
of DGen power sources even small enough to supply a single household with electric
power at an affordable price. The last challenge, the political challenge, remains to be
solved so that DGen can take its place in the bag of tricks used to supply the magic of
electric power for everyone to use.

14