Electrical Power Systems Quality, Second Edition phần 9 potx

capability to electromechanical network relays. In the past, these sup￾plemental relays had minimum time delays of 1 s or more since their

mission was to wait for the elevator to descend. However, not all util￾ities endorse this low-current, time-delay technique. Some feel that

any time delay in opening the network protectors degrades the high

service quality that the network system is intended to provide.

The load-generation control and DG tripping schemes mentioned

above are intended to ensure that the network protectors are never

opened by exported power. As long as the schemes work properly, the

network protectors are never exposed to the out-of-phase voltage con￾ditions that may exceed the switch capability. However, because of the

potentially catastrophic consequences of causing a network protector

failure, it is prudent to provide a backup. An interlocking scheme that

trips the DG instantaneously when a certain number of network pro￾tectors have opened ensures that the network protectors will not be

exposed to out-of-phase voltages for more than a few cycles. The deci￾sion as to how many protectors must open before the DG is tripped

(one, two, or all) is a tradeoff between security of the protectors and

nuisance tripping of the DG. Note that this scheme does not relieve the

DG installer from the responsibility of providing stuck-breaker backup

protection for the DG’s switching device.

An even more secure approach to avoiding overstressing the network

protectors is to replace existing protectors with new designs that are

capable of interrupting fault currents from sources with higher X/R

422 Chapter Nine

Time

Adjustable

delay time

Time delay for

low currents

Adjustable

instantaneous trip

threshold Instantaneous trip

for higher currents

100

Current (% of transformer rating)

Figure 9.32 Adjustable reverse-power characteristic.

Distributed Generation and Power Quality

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)

Copyright © 2004 The McGraw-Hill Companies. All rights reserved.

Any use is subject to the Terms of Use as given at the website.

ratios and of withstanding out-of-phase voltages across the open

switch. One major U.S. manufacturer of network protector units has

recently introduced such high-capacity protectors in 800- to 2250-A rat￾ings and plans to introduce them in ratings up to 6000 A. These pro￾tectors are designed to be retrofitted in many existing types of network

units.

A possible DG interconnection problem exists that would involve net￾work protectors without a network bus interconnection. If a DG is

interconnected on a feeder that also supplies a network unit, then if its

feeder breaker is tripped and the DG is not rapidly isolated, it may

impact one or more of the network units as if it were isolated on the net￾work bus. For this type of event to occur, the DG output does not have

to be matched to the feeder load. For the excess generation case, it only

has to be momentarily greater than the load on the network bus. Under

this condition the power continues to flow to the network bus from the

feeder with the interconnected DG, which keeps that protector closed.

However, the excess power flows through the network back to the other

feeders, resulting in the opening of the protectors connected to those

feeders. Once open, these protectors will be separating two indepen￾dent systems. For the case of less generation than load, the protector

connecting to the feeder with the generation may trip. Again, such a

condition would have a protector separating two independent systems.

Therefore, such DG applications should be avoided unless the DG

breaker is interlocked with the feeder breaker with a direct transfer

trip scheme.

9.7 Siting DG

The value of DG to the power delivery system is very much dependent

on time and location. It must be available when needed and must be

where it is needed. This is an often neglected or misunderstood concept

in discussions about DG. Many publications on DG assume that if 1

MW of DG is added to the system, 1 MW of additional load can be

served. This is not always true.

Utility distribution engineers generally feel more comfortable with

DG installed on facilities they maintain and control. The obvious choice

for a location is a substation where there is sufficient space and com￾munications to control centers. This is an appropriate location if the

needs are capacity relief on the transmission system or the substation

transformer. It is also adequate for basic power supply issues, and one

will find many peaking units in substations. However, to provide sup￾port for distribution feeders, the DG must be sited out on the feeder

away from the substation. Such generation will also relieve capacity

constraints on transmission and power supply. In fact, it is more effec￾Distributed Generation and Power Quality 423

Distributed Generation and Power Quality

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)

Copyright © 2004 The McGraw-Hill Companies. All rights reserved.

Any use is subject to the Terms of Use as given at the website.

tive than the same amount of DG installed in the substation.

Unfortunately, this generation is usually customer-owned and distrib￾ution planners are reluctant to rely on it for capacity.

The application of DG to relieve feeder capacity constraints is illus￾trated in Fig. 9.33. The feeder load has grown to where it exceeds a

limit on the feeder. This limit could be imposed by either current rat￾ings on lines or switchgear. It could also be imposed by bus voltage lim￾its. There is DG on the feeder at a location where it can actually relieve

the constraint and is dispatched near the daily peak to help serve the

load. The straightforward message of the figure is that the load that

would otherwise have to be curtailed can now be served. Therefore, the

reliability has been improved.

This application is becoming more common as a means to defer

expansion of the wire-based power delivery infrastructure. The gener￾ation might be leased for a peak load period. However, it is more com￾mon to offer capacity credits to customers located in appropriate areas

to use their backup generation for the benefit of the utility system. If

there are no customers with DG in the area, utilities may lease space

to connect generation or, depending on regulatory rules, may provide

some incentives for customers to add backup generation.

There is by no means universal agreement that this is a permanent

solution to the reliability problem. When utility planners are shown

Fig. 9.33, most will concede the obvious, but not necessarily agree that

this situation represents an improvement in reliability. Three of the

stronger arguments are

1. If the feeder goes out, only the customer with the DG sees an

improvement in reliability. There is no noticeable change in the ser￾vice reliability indices.

424 Chapter Nine

Feeder Limit DG Dispatched

ON

Daily Load Profile

DG Sited to Provide Feeder Relief

Figure 9.33 DG sited to relieve feeder overload constraint.

Distributed Generation and Power Quality

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)

Copyright © 2004 The McGraw-Hill Companies. All rights reserved.

Any use is subject to the Terms of Use as given at the website.

2. Customer generation cannot be relied upon to start when needed.

Thus, the reliability cannot be expected to improve.

3. Using customer-owned generation in this fashion masks the true

load growth. Investment in wire facilities lags behind demand,

increasing the risk that the distribution system will eventually not

be able to serve the load.

It should also be noted that the capacity relief benefit is nullified

when the distribution system is upgraded and no longer has a con￾straint. Thus, capacity credits offered for this application generally

have a short term ranging from 6 months to 1 year.

If one had to choose a location on the distribution feeder, where

should the DG be located? The optimal DG siting problem is similar to

the optimal siting problem for shunt capacitor banks. Many of the same

algorithms can be used with the chief difference being that the object

being added produces watts in addition to vars. Some of the same rules

of thumb also apply. For example, if the load is uniformly distributed

along the feeder, the optimal point for loss reduction and capacity relief

is approximately two-thirds of the way down the main feeder. When

there are more generators to consider, the problem requires computer

programs for analysis.

The utility does not generally have a choice in the location of feeder￾connected DG. The location is given for customer-owned generation,

and the problem is to determine if the location has any capacity-related

value to the power delivery system. Optimal siting algorithms can be

employed to evaluate the relative value of alternative sites.

One measure of the value of DG in a location is the additional

amount of load that can be served relative to the size of the DG.

Transmission networks are very complex systems that are sometimes

constrained by one small area that affects a large geographical area. A

relatively small amount of load reduction in the constrained area

allows several times that amount of load to be served by the system.

This effect can also be seen on distribution feeders. Because of the

simple, radial structure of most feeders, there is generally not a con￾straint so severe that DG application will allow the serving of addi￾tional load several times greater than the size of the generator.

However, there can be a multiplying effect as illustrated in Fig. 9.34.

This example assumes that the constraint is on the feeder rather

than on the substation. If 1 MW of generation were placed in the sub￾station, no additional load could be served on the feeder because no

feeder relief has been achieved. However, if there is a good site on the

feeder, the total feeder load often can grow by as much as 1.4 MW. This

is a typical maximum value for this measure of DG benefit on radial

distribution feeders.

Distributed Generation and Power Quality 425

Distributed Generation and Power Quality

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)

Copyright © 2004 The McGraw-Hill Companies. All rights reserved.

Any use is subject to the Terms of Use as given at the website.

Another application that is becoming common is the use of DG to

cover contingencies. Traditionally, utilities have built sufficient wire￾based delivery capacity to serve the peak load assuming one major fail￾ure (the so-called N-1 contingency design criterion). At the distribution

feeder level, this involves adding sufficient ties to other feeders so that

the load can be conveniently switched to an alternate feeder when a

failure occurs. There must also be sufficient substation capacity to

serve the normal load and the additional load expected to be switched

over during a failure. This results in substantial overcapacity when the

system is in its normal state with no failures.

One potentially good economic application of DG is to provide sup￾port for feeders when it is necessary to switch them to an alternate

source while repairs are made. Figure 9.35 depicts the use of DG

located on the feeder for this purpose. This will be substantially less

costly than building a new feeder or upgrading a substation to cover

this contingency.

The DG in this case is located near the tie-point between two feeders.

It is not necessarily used for feeder support during normal conditions

although there would often be some benefits to be gained by operating

the DG at peak load. When a failure occurs on either side of the tie, the

open tie switch is closed to pick up load from the opposite side. The DG

is dispatched on and connected to help support the backup feeder.

Locating the DG in this manner gives the utility additional flexibil￾ity and more reconfiguration options. Currently, the most common DG

technology used for this application is currently diesel gensets. The

gensets may be mounted on portable trailers and leased only for the

peak load season when a particular contingency leaves the system vul￾nerable. One or more units may be interconnected through a pad￾426 Chapter Nine

Pload

Pgen

= 0

Pload

Pgen

= 1.4

Figure 9.34 Ability of DG to increase the capacity of a distribution feeder is

dependent on DG location.

Distributed Generation and Power Quality

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)

Copyright © 2004 The McGraw-Hill Companies. All rights reserved.

Any use is subject to the Terms of Use as given at the website.