You are on page 1of 10

By Farid Katiraei

and Julio Romero Agero

Studies for Utility-Scale


Photovoltaic Distributed
Generation

Digital Object Identifier 10.1109/MPE.2011.940579


Date of publication: 21 April 2011

62

IEEE power & energy magazine

1540-7977/11/$26.002011 IEEE

may/june 2011

PHOTODISC, SOLAR PANELS COURTESY OF SUNEDISON

SOLAR PHOTOVOLTAIC DISTRIBUTED GENERATION (PV-DG) SYSTEMS ARE ONE


of the fastest-growing types of renewable energy sources being integrated worldwide onto distribution systems. Many North American utilities, governed by state or provincial incentives and/or
mandated by green-generation portfolio requirements, are facing installations of large PV plants
with capacities in the order of several megavoltamperes (MVAs) that are owned either by the utility
or by private power producers.
The large number of interconnection requests has spurred utilities to develop screening methods that can quickly identify cases with no (or minimal) impact on the distribution system versus
those applications requiring in-depth engineering studies. Typical study areas include investigating possible adverse impacts on the power quality, protection coordination, and operation of
distribution feeders.
As part of new PV-DG plant interconnection impact
studiesand in addition to typical steady-state voltage
and power flow analysessome utilities are requiring in-depth investigation of potential dynamic impacts
of inherently variable PV-DG units on transient feeder
voltages under various load and generation conditions.
These studies may also be tasked with determining
interactions with distribution equipment such as an
increase in the operation of line-voltage regulators and
substation tap changers and change in status (on/off) of
capacitor banks.
This article introduces new study tools and methodologies to help utility engineers investigate the potential impact of these new types of generation on the grid. In addition, it will help
engineers pursue remedial actions to reduce the barriers related to PV-DG interconnection
while preventing adverse impacts on the integrity and power quality of the grid.

may/june 2011

IEEE power & energy magazine

63

PV-DG in North America

Medium-Scale PV-DG

PV-DG plants dot the southwestern United States, where


natural conditions favor exploiting solar resources. The proliferation of PV-DG has not been limited to this region, however; states as far north as Vermont and Maine and some
provinces in Canada are also experiencing a rise in PV-DG.
It is due to economic and regulatory factors that the vast
majority of these plants are interconnected to the existing
distribution system in the form of PV-DG instead of to the
transmission grid. PV-DG may be broadly classified into
three types, as described below.

Capacities range from 101,000 kW and include installations on small or large buildings (e.g., residential complexes,
retail stores, government sites, and other buildings). Their
typical interconnection configuration depends on the capacity of the PV-DG system. Larger plants (those with capacities in the hundreds of kW) may typically have installations
similar to those of utility-size PV-DG plants, including
separate interconnection transformers, with the main difference being the nominal ratings of associated equipment
(e.g., transformers and switches). Smaller plants in which
the PV-DG capacity is comparable to the load may have
typical installations similar to those of small-scale PV-DG
units, using the existing customer transformers, possibly
with minor changes in the interconnection.

Utility-Scale PV-DG
Megawatt (MW)-size plantse.g., 110 MWare either
directly connected to conventional feeders or to distribution substations via dedicated (express) feeders. Utilityscale PV-DG has nominal capacities compatible with
substation ratings or manageable by medium-voltage
distribution feeders (e.g., 12.47-kV feeders). These types
of installations are three phase and typically require one
or more interconnection transformers. A MW-size PV-DG
plant generally includes several power-electronic inverter
modules connected in parallelusually called power conversion systems (PCSs)that vary in size depending on
the model and manufacturer (see Figure 1). Each PCS is
equipped with internal and external protection schemes
such as fast overcurrent protection and under and overvoltage and frequency safeguards, as well as active anti-island
protection schemes to prevent the PV-DG plant from feeding power to the grid in the event that the utility grid connection is lost.

1 MW PCS
dc/ac
Inverter

480 V
#2

dc/ac
Inverter

PCS 2
(1 MW)

12.47 kV
#1
Y-

480 V
#2

12.47 kV
#1
PCC

Y-
.
.
.
PCS 10
(1 MW)

480 V
#2

12.47 kV
#1

.
.
.

Small-Scale PV-DG
Capacities range up to 10 kW. This category mainly includes
distributed rooftop PV-DG units installed at customer residences and connected to secondary lines (120/240 V). The
PV-DG system is usually single phase and can produce
more or less electricity than required by the customers load.
Typical installations do not require an interconnection transformer. To reduce complexity in utility studies, many North
American utilities lump the PV-DG units installed on a common circuit and/or connected to a pole-mounted transformer
together and represent them by an aggregated PV-DG unit.
Figure 2 shows such a typical aggregation area for small
rooftop PV-DGs.

Need for PV-DG Impact Studies


Distribution systems have traditionally been designed to
operate in a radial fashion, and it is well known that the
interconnection of distributed generation may cause impacts
that need to be studied and planned for. Since the passage
of the Public Utility Regulatory Policies Act (PURPA) in
1978, typical applications of distributed generation have
consisted of reciprocating engines or small hydro plants
whose ac power injection is relatively constant and require
only a transformer for interconnection to the distribution
system. As expected, the proliferation of PV-DGwhich
has intermittent resource characteristics that vary the power
output throughout the day and requires the conversion of
dc generation to ac power via invertersrepresents a less

PoleMounted
Transformer

Y-
10 MW PV-DG Plant

figure 1. Utility-scale PV-DG plant (PCC signifies point of


common connection).
64

IEEE power & energy magazine

2 x 120 Vac
Secondary Circuit

figure 2. Aggregation of small, rooftop PV-DG units.


may/june 2011

Voltage rise and fluctuations have a direct impact


on feeder voltage profiles, which can lead to frequent operation of
LTCs, line VRs, and voltage-controlled capacitor banks.
familiar challenge for distribution utilities and can give rise
to impacts requiring different responses than those of conventional DG.
Due to the variability caused by passing clouds, PV-DG
can significantly affect volt/var control, power quality, and
system operation. Some of these impacts can only be investigated through dynamic/transient studies that include the
time-varying behavior of fast-acting generation (inverters),
load, and automatic voltage-control devices on the feeders.
Given the complexity of such studies and the fact that the
proliferation of this type of DG is fairly recent, the body of
work in this field is sparse, impacts and mitigation measures
are more difficult to identify, and utilities are less prepared to
deal with them. Furthermore, for small-scale PV-DG, studying the problem is increasingly complex because of uncertainty about the location and timeline of market penetration
as well as the potential interactions with emerging active
loads such as electric vehicles (EVs).
Impact studies are generally intended for quantifying the
extent of the issues and providing utilities with guidelines,
tools, and processes with which to manage the expected
steady-state and dynamic transient impacts of PV-DG. Most
important, these studies can assess mitigation measures for
any problems discovered and determine the cost and effectiveness of alternative solutions. The proliferation of PV-DG
is expected to continue through the next decade and beyond;
therefore, it is critical for distribution utilities to understand
the associated impacts of integrating PV-DG plants on distribution system planning and operation. Depending on the
degree of PV-DG deployment, impacts can be local (e.g., at
individual feeder or substation level) or systemwide (e.g.,
affecting several feeders and substations across the utilitys
service territory and including subtransmission and transmission facilities).
PV-DG impacts on distribution systems can be either
steady-state or dynamic in nature, and they include:
changes in feeder voltage profiles, including voltage
rise and unbalance
changes in feeder loading, including potential equipment and component overload
frequent operation of voltage-control and regulation
devices, such as load tap changers (LTCs), line voltage
regulators (VRs), and capacitor banks
reactive-power flow fluctuations due to operation of
switched capacitor banks
power quality, PV-DG intermittency may lead to voltage fluctuation issues
may/june 2011

overcurrent and overvoltage protection, including

misoperation of overcurrent protection equipment and


temporary overvoltage (TOV)
change in electric losses, where relatively large reverse power flow may increase losses
variations in power factor of a feeder or system, which
may have economic impacts on local distribution companies purchasing power from larger utilities
reliability and operation of the system.
The severity of these impacts varies with the penetration level,
the location of the PV-DG, and the electrical characteristics
of the distribution systems. For instance, a high PV-DG penetration level could cause a feeder to become an active circuit
and inject power back to the transmission system. This highpenetration condition may affect voltage profiles, overcurrent protection, and capacitor-bank operation. Such a situation may occur on feeders selected for integrating multiple
utility-scale PV-DG plants. Because both PV-DG output and
feeder load vary throughout the day, it is necessary to investigate impacts for different degrees of penetration of PV-DG
and also to assess the effects under various conditions of
feeder loading. Study scenarios should account for seasonal
PV-DG fluctuations and such variations in feeder loading.

Impact Studies
The typical scope of work for PV-DG impact studies
includes:
1) identifying the local and/or systemwide impacts of
PV-DG on the power distribution grid
2) providing utility customers with guidelines regarding
the expected impacts as a function of the penetration
level of PV-DG
3) assessing potential mitigation measures for any problem discovered during the study.
Additional tasks may include:
1) verifying models, such as performing PV inverter
tests in a laboratory environment to develop a detailed
computational model representing the dynamic characteristics of vendor-specific devices
2) developing best-practice interconnection guidelines
for PV-DG readiness studies.
The main objectives of impact studies are to:
1) quantify steady-state impacts
2) quantify dynamic impacts
3) determine remedial measures.
Steady-state impacts are estimated via distribution software analysis, and they require running batch processes
IEEE power & energy magazine

65

length, and customer density).


Results for each characteristic
Conduct Steady-State
feeder can then be extrapolated to
Systemwide Analysis
its respective cluster to determine
for Small- and MediumScale PV-DG
expected impacts on the overall
Validate SteadyPower Flow
utility system for a variety of penState Feeder
Analyses
etration scenarios.
Models
Validate/Verify
Inverter Models
Conduct Steady-State
Finally, systemwide studies
Individual Analysis for
may investigate potential interacUtility-Scale PV-DG
tion effects with EVs and other
Develop
Guidelines and
technologies such as distributed
Validate/Verify
Extrapolate to
Dynamic Feeder
energy storage. Systemwide studRepresent Utility
Models
ies are used, for instance, when
System
utilities want to evaluate the widespread impact of the massive proDynamic/
Develop
Conduct Dynamic/
liferation of small-scale PV-DG
Transient
Conclusions and
Transient Analyses for
Analyses
Recommendations
Utility-Scale PV-DG
units. This is primarily because
the individual unit-level impacts
of small installations usually are
figure 3. Generalized approach for PV-DG impact studies on distribution systems.
not significant, whereas they do
for analyzing a variety of PV-DG outputs at correspond- become noticeable and of interest for moderate to large
ing feeder loading conditions at various times of the day, penetration levels. In that situation, the systemwide studmonth, or year. Such simulations may require analyzing ies can assess the aggregated effect of a multitude of these
only worst-case scenarios or more comprehensive analy- units with locations distributed along various feeders and
ses (e.g., 8,760 hours of the year). Dynamic impacts are substations across the utilitys territory.
estimated by means of detailed modeling and simulaAssessing PV-DG impacts requires the intensive use of
tions conducted for worst-case but feasible scenarios. The computer simulations. Addressing various types of impacts
dynamic studies incorporate control-system behavior and necessitates concurrent use of steady-state and dynamic
response times (second or minute based) for feeder appa- simulation tools. A new simulation study terminology,
ratus and generation.
quasi-static analysis, is emerging that involves time-series
Impact studies can be of local or systemwide scope. statistical studies and solving sequential power flows per
Localized studies address impacts of PV-DG on a specific time step of interest within a prespecified time frame. For
feeder or substation. These studies usually have the objective example, the objective of an impact study might be to
of identifying impacts and mitigation measures for intercon- determine the potential effect of adding a PV-DG plant to
nection of one or several utility-scale PV-DG plants onto a a feeder on the number of LTC operations. Therefore, one
specific feeder of a substation. They tend to be deterministic will need to run 480 min-based power flow simulations
in nature because first, it is usually feasible to obtain load to capture the effect of voltage variations caused by the
and PV data and evaluate impacts under a variety of realistic PV-DG plant on the tap changer position during eight hours
feeder loading and solar radiation conditions and second, the of daily light. The study results can be used to calculate
location and characteristics of large PV-DG installations are expected changes in the total number of LTC operations per
commonly known in advance based on the interconnection representative day by comparing two cases: no PV and with
request made to the utility.
PV. The study may need to be repeated per day for an entire
Systemwide studies address impacts on the overall util- year to incorporate monthly and seasonal variations in load
ity power distribution systems for a set of what if sce- and PV generation.
narios that consider growing market penetration levels.
Figure 3 presents an overall impact study methodolThese studies deal with uncertainties about the location, ogy that can be used in conducting both local and systemtimeline, and characteristics (e.g., installed capacity) of wide PV-DG studies. The red blocks represent steady-state
PV-DG plants. They typically include a mix of different analyses and the blue blocks represent dynamic/transient
types of PV-DG, with a considerable amount of small-scale studies; these studies are complementary and synergetic.
residential units. Generally, it may not be feasible from an Steady-state studies can be conducted using commercial
economic standpoint to study and evaluate the impacts on distribution software. They may require analyzing numerous
all utility feeders. Hence, study approaches may be targeted combinations of PV-DG output and feeder loading condiat identifying and studying a set of characteristic feed- tions, however. This can be a time-consuming task if autoers only. Each characteristic feeder represents a cluster of mated tools for batch-file processes are not available. On the
feeders with similar features (e.g., voltage levels, loading, other hand, dynamic transient studies involve specialized
Identify
Representative
Feeders

66

IEEE power & energy magazine

may/june 2011

This article introduces new study tools and


methodologies to help utility engineers investigate the potential
impact of these new types of generation on the grid.
electromagnetic transient tools commonly used for electro- impacts. In this case, an express feeder may be built to conmagnetic transient analyses or protection system studies but nect the PV-DG plant directly to the substation. Figure 5
unfamiliar to distribution engineers. Figure 3 shows that shows a substation that includes an express feeder and sevstudies start with identifying the representative feeders (for eral conventional feeders. Typically, there is no load on
systemwide studies) or the feeder to be analyzed (for local- the express feeder. The PV generation is consumed by the
ized studies).
adjacent feeders of the substation. One of the conventional
Depending on the relative size of a PV-DG plant with feeders, which includes automatic voltage control devices
respect to the feeder load and ratings, the PV-DG may be (here, a voltage regulator and capacitor bank), is also repconnected to an existing conventional feeder or a dedicated resented to investigate the impact of changes in the LTC
(express) feeder. Figure 4 shows a schematic diagram of a tap position (due to PV plant output variations) on the voltconventional feeder that includes a 2-MW PV-DG plant at ages and operation of the adjacent feeder. The equivalent
the end of the feeder. There are two VRs (VR1 and VR2) load of the rest of the neighbor feeders connecting to the
on the path to the PV plant (along the feeder trunk) and same substation is also included in the boundary of the
a switched capacitor bank (Cap 1) on an adjacent branch study to maintain realistic substation transformer loading.
close to the PV-DG plant. An impact study for this feeder
should investigate, among other things, any potential effect Steady-State Analyses
of changes in the feeder voltages and power flow on the for Utility-Scale PV-DG
operations of VR1, VR2, and Cap 1. Blue arrows show the For utility-scale PV-DG, simulations are conducted for each
direction of active power flow based on the amount of load representative feeder or for a particular feeder based on a
and PV generation per feeder section. Bidirectional power PV interconnection request. For the latter case, there may
flow is expected for the majority of the feeder, including the be two or more interconnection requests in the queue for
power through VR1 and VR2. Immediately upstream of the the feeder in question; however, utility preference is norfirst recloser (RC1), there is sufficient load due to the loading mally to study the plants interconnection in the order that
of the adjacent branch to make power flow unidirectional applications are submitted by assuming addition of one PV
under all circumstances. VR controls based on cogeneration and/
or bidirectional operation modes
M4
C1
may be considered and studied
PV
for VR1 and VR2. Also, there
Cap1
is a potential for zero mismatch
PV2
PV1
between PV generation and load
Vr2s
Vr2p
downstream of RC1 that may
2 MW
affect anti-islanding protection of
C3
VR2
the PV-DG plant upon accidenM6
Cap3
tal switching of RC1. Therefore,
islanding studies may be required
C2
to determine disconnection time
M5
Cap2
Vr1s
Substation
of PV-DG and the maximum temM3
RC2
VR1
porary overvoltages (TOVs) that
Vr1p
the operation of RC1 may cause.
M2
M1
Utilities may suggest impleRC1
LTC
menting dedicated feeders to
Measurement Point
Adjacent
accom modate a la rge PV-DG
Feeders
Load
LTC or Voltage Regulator
plant when it is not feasible
Capacitor Banks
Recloser
to connect the plant onto the
conventional feeders due to an
expectation of numerous adverse figure 4. Conventional feeder with a PV-DG plant.
may/june 2011

IEEE power & energy magazine

67

plant at a time. When the next PV interconnection application is considered for review and study, the base case will
then include any previously studied and approved PV-DG
plants on the feeder.
The objective of steady-state simulations is to analyze
feeder voltage profiles, equipment loading, power flows,
and losses for the base case (without PV or with only previously connected PV-DG plants) and after interconnection
of a newly proposed PV-DG plant. Furthermore, simulations capture the number of operations for LTCs, line voltage regulators, and capacitor banks. Study results estimate
the annual operation increases and assess potential impacts
on equipment maintenance. Numerous simulations may be
required to model different scenarios involving variations in
load and PV generation. The inputs required for these analyses include:

PV
Plant

Four Miles
PV Express Feeder

6 MW

LTC
Conventional Feeder

M1

M2

C1

Substation
Adjacent
Feeders
Load
Measurement Point

LTC or Voltage Regulator


Recloser

Capacitor Banks

figure 5. System schematic for express feeder studies.

0.0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23

Normalized Apparent Power

1.5

0.5

0.5
1.0
1.5

0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%

Hour

figure 6. Reverse power flow for various penetration levels


of PV-DG.
68

IEEE power & energy magazine

Potential Steady-State
Impacts and Concerns
Some of the most common expected impacts of PV-DG on
the distribution system include the following.

Reverse Power Flow


Proliferation of PV-DG can lead to reverse power-flow conditions at section, feeder, and substation levels, as shown in
Figure 6. Reverse power flow can negatively affect protection
coordination and operation of line voltage regulators. Under
high-penetration scenarios, the total PV-DG output will
likely offset the feeder load. The power flow direction will be
reversed, and the feeder will start exporting power to neighbor feeders or to the transmission system. Because distribution feeders are typically designed for unidirectional power
flows, this situation may noticeably affect the overcurrent
protection coordination of the distribution system. Therefore,
to account for possible reverse power flow, specific studies
must be done on a feeder-by-feeder basis to select the most
adequate protection strategy and design if high penetration of
utility-scale PV-DG plants is expected. Reverse power flow
can also affect the operation of VRs, and they must be evaluated under control modes that allow bidirectional power flow
(e.g., cogeneration or bidirectional modes) to avoid potential voltage violations. Note that PV-DG may cause reverse
power flow during the daytime. Hence, any selected operation
mode for voltage regulators should be truly bidirectional and
assessed for no-PV cases with opposite power flow expected
at night or low-PV generation conditions.

Voltage Rise
and Fluctuations

Reverse Power Flow


Versus PV-DG Penetration Level
1.0

1) a detailed distribution feeder model, including settings of voltage control and regulation equipment
2) a typical PV-DG injection profile (monthly and annual
average)
3) PV-DG capacity in kW
4) 8,760-hour feeder load data.
If available, additional information, such as the status of
capacitor banks and typical customer power factors, can
improve the accuracy of the results.

Some of the most notorious impacts of PV-DGs are voltage


rise and voltage variations due to output intermittency. Both
issues worsen as the penetration level of PV-DG increases.
The effects are particularly evident and problematic when
large PV-DG plants are connected near the end of long,
lightly loaded feeders. Figure 7 shows an example of the
PV-DG impact on a feeder voltage profile. The magnitude of
the voltage rise depends on the configuration of each feeder
and the location of the PV-DG and capacitor banks. Some
solutions for mitigating this impact are to:
1) modify the control settings of capacitor banks to
ensure that they are off during maximum PV-DG
output
may/june 2011

Dynamic studies for PV-DG integration typically


analyze the effects of fast-varying phenomena caused by
the PV-DG or initiated on the system.

Reactive Power Fluctuations


Frequent on-off switching of voltage-controlled capacitor
banks and frequent operation of LTCs and line VRs lead to
reactive power flow fluctuations. If the penetration level of
PV-DG plants is large and widespread, this may also affect
subtransmission and transmission systems. The disconnection of capacitor banks implies that this reactive power has
to be supplied by the transmission system. This can have
important economic impacts for large penetration levels
of PV-DG, given that transmitting reactive power is more
expensive than supplying it locally. It also has various implications for distribution substations and transmission lines,
such as increasing losses and substation/transmission line
loading. Figure 8 shows an example of reactive power fluctuations on a distribution feeder caused by the operation of
a voltage-controlled capacitor bank due to a PV-DGdriven
voltage rise.

Modification of Feeder
Section Loading
The location of the PV-DG can significantly affect the
loading of feeder sections. Therefore, before installing
utility-scale PV-DG plants, it is necessary to verify that
the feeder sections located between the PV-DG plants and
substation have enough available capacity to distribute the
may/june 2011

Increase in Power Losses


The impact of PV-DG on losses is similar to that on section loading. For low to moderate penetration levels, line
losses tend to decrease until they reach a minimum. For

Voltage Profile (PU)

1.05
1.04
1.03
1.02
1.00
1.00
0.99
0.98
0.97
0.96
0.95

No Caps
Caps
PV-DG
0

1
2
3
4
5
Distance from Substation (Miles)

figure 7. Impact of PV-DG on feeder voltage profile.

Reactive Power Fluctuation


0.8
0.6
0.4
0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4

Without PV-DG
With PV-DG
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

Voltage rise and fluctuations have a direct impact on feeder


voltage profiles, which can lead to frequent operation of
LTCs, line VRs, and voltage-controlled capacitor banks.
This may cause additional step-voltage changes. Due to typical delays associated with the control methodology of these
devices (e.g., 3090 s), minute-based step-voltage variations may be experienced. In addition, more frequent operation shortens the expected life cycle of these devices and
increases maintenance requirements. In the special case of
using line-drop compensation (LDC) for LTCs or line VRs,
the voltage impact may become more significant because
voltage regulation is a function of line current, which is offset by the PV-DG plants.

Voltage (PU)

Interaction with Capacitor Banks,


LTCs, and Line VRs

surplus power of the PV-DG plants (after subtracting local


and downstream load). Similarly, it is necessary to verify that
the ratings of distribution switchgears and equipment are not
exceeded. In some cases, particularly for large penetration
levels, reconductoring of some sections may be necessary.
Figure 9 shows the loading of the initial section of 15 different feeders as a function of the penetration level of PV-DG.
Here, for low to moderate penetration levels, PV-DG offsets
the load and decreases section loading. At high penetrations,
however, the section loading is increased, as the PV-DG contribution is larger than the base load.

Reactive Power (MVAR)

2) avoid using fixed capacitor banks


3) lower the voltage reference on LTCs and line voltage
regulators.
A more effective solution is to operate the PV-DG plants at
nonunity leading power factors (absorbing VArs).

Hour

figure 8. Reactive power fluctuations due to interconnection of PV-DG.


IEEE power & energy magazine

69

Solar photovoltaic distributed generation systems are one of the


fastest-growing types of renewable energy sources being integrated
worldwide onto conventional distribution systems

Feeder
Loading (% of Themal Rating)

high PV-DG penetration levels, line losses tend to increase


for several reasons. For instance, the loading of distribution lines under high PV-DG penetration may be greater
than the normal feeder loading conditions. Another reason
may be the lack of local reactive power supply via capacitor
banks (if they have been switched off due to voltage rise).
Moreover, the nodal voltage increase caused by high PV-DG
penetration will increase the no-load losses of distribution
transformers. Figure 10 shows a plot of feeder losses as a
function of the penetration level of PV-DG. The results show
the initial decreasing trend of losses until they reach a minimum and then an increase until they exceed the losses for the
base case (no PV-DG).

Feeder Loading Versus PV-DG


Penetration Level

120

F1
F2
F3
F4
F5

100
80
60

F6
F7
F8
F9
F10

F11
F12
F13
F14
F15

40
20
0
0

10

20 30 40 50 60 70 80 90 100
PV-DG Penetration (%)

90
80
70
60
50
40
30
20
10
0

Plss (kW)
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

kW

figure 9. Feeder section loading as a function of PV-DG


penetration level.

Hour

figure 10. Feeder losses as a function of PV-DG penetration level.


70

IEEE power & energy magazine

Dynamic Analysis
and Effects
Dynamic studies for PV-DG integration typically analyze the
effects of fast-varying (transient) phenomena caused by the
PV-DG (e.g., generation intermittency due to cloud movements) or initiated on the system (e.g., following faults and
subsequent switching). The time frame of interest may vary
from subminute to several hours with very fine time steps
that can capture the dynamic behavior and response time of
the feeder equipment, including PV inverters (typically with
subsecond steps), automatic voltage control devices (with
subminute or minute steps), and loads.
Dynamic analysis generally covers study cases for light
and heavy load conditions such as sudden connection and
disconnection (tripping) of a PV-DG plant; quick, large
fluctuations of the PV-DG output due to intermittency; and
accidental islanding of part of a feeder downstream of an
automatic switching device (a recloser or remote-controlled
switch). The main study objectives are usually to determine
the impact of PV-DG integration on voltage transients and
power quality (e.g., voltage sags, swells, and flicker) and/
or PV-DG behavior during faults and system dynamics.
Such studies typically include potential interactions of conventional and nonconventional voltage control devices on
a feeder.
Figure 11 shows a proposed study methodology for
dynamic analysis. Color coding is used to relate overall
process flow blocks (on the right) to the specific steps (on
the left) for each stage of the study. The approach starts with
developing and verifying a base-case dynamic feeder model
including control aspects of automatic voltage regulation
devices and generic or vendor-specific PV inverter models.
No-PV study cases refer to the base-case situation of the
feeder prior to adding a newly proposed PV-DG plant. The
base-case model may include any existing generation plant
(conventional or nonconventional) and is used to determine
initial feeder conditions such as the status of capacitor
banks and tap changer positions for LTCs or VRs. In the
next step, prespecified PV study scenarios are simulated.
Two main groups of study cases are normally defined:
studies involving PV-DG output variations and
investigating their impact on feeder voltages and
operation
case studies intended to assess the effects of transients and subsequent switching initiated on the
feeder on the PV-DG operation to determine any adverse power-quality consequences.
may/june 2011

Common examples of the latter category are fault studies and


Prepare and
No PV Studies: Initial Conditions
Verify
islanding studies.
Basic
Feeder
With
PV
Studies:
Use
Solar
Radiation
Profiles
As show n i n Fig u re 11,
Model
and
Load
Changes
(light/heavy
loads)
dynamic studies are also used to
assess the effectiveness of comRun No-PV Case and
Switching/
a) Effect on feeder voltages and
monly known mitigation meaCompare: Voltages
Islanding
expected
range
of
variations
sures or determine case-specific
(IEEE 1547 Taps and Caps Status
b) Interactions with cap bank controllers
solutions. Examples of commonly
Protection)
and/or tap changers
discussed mitigation solutions
c) Impact of reverse power flow on
Simulate
operation mode of voltage regulators
include:
PV
Study
changing the power factor
a) Identifying possible feeder Islanding scenarios
Scenarios
(islanded
areas
and
local/generation
mismatch)
of a PV-DG plant accordb) Temporary over voltage (TOV) issues based on
ing to the load and/or volttypical PV plant disconnection time
age measurements at the
Check
a) Identifying at various power factors
Mitigation
point of PV-DG connection
b) Change in VR control modes and settings
Methods
(the power-factor adjustc) Select optimum power factor to minimize losses
ments may be prescheduled
or performed in a dynamic
fashion to provide reactive figure 11. Proposed approach for PV-DG dynamic studies.
power compensation)
changing the set points or control modes of volt- tection) targeted to enhance the interaction of utility-scale
age regulators and capacitor banks to accommodate PV-DG plants with the grid and provide the means for coorbidirectional power flow and the effects of volt- dinated control and operation through localized or utilityage and current fluctuations introduced by PV-DG wide supervisory control systems.
plants
Presently, another major obstacle to performing PV-DG
adding a ground bank or fast ground overvoltage pro- dynamic analysis is the lack of high-resolution PV and load
tection scheme at the PV-DG site to prevent feeding data specific to the study area and for an extended period of
the load on an ungrounded system subsequent to faults time. Utilities typically have good load measurement data
and accidental switching.
at the substation level based on 5- or 10-min averages. One
For specific dynamic impacts, some customized and/or rarely finds subminute (1-, 5-, or 10-s) data resolutions for
complex mitigation methods may need to be developed and load and solar radiation in various regions of a utility distriexamined to ensure applicability.
bution territory, however. The need for high-resolution data
has been realized by many North American utilities involved
Need and
with PV-DG integration, and work is currently under way to
Future Direction
install and/or enhance the monitoring equipment on distriThere is an urgent need for systematic interconnection stud- bution systems so that it can collect data with high preciies due to the proliferation of utility-scale PV-DG and high- sion. Eventually, utilities will likely build data repositories
penetration PV integration scenarios. These studies need to of actual field measurement data for selected (or represencover various steady-state and dynamic aspects of distribu- tative) areas in their systems for which a high penetration
tion feeders operation under the new generation and load of emerging loads (e.g., EV) or generation (e.g., PV and/or
regime. Several North American and international organi- wind) may be expected.
zations have established task forces and working groups
(e.g., the IEEE 1547.7 and 1547.8 groups) to define study For Further Reading
methodologies and recommend modeling approaches for S. Wilcox and C. A. Gueymard. (2010). Spatial and temthe design and analysis of distributed generation integra- poral variability of the solar resource in the United States.
tion. Part of the collaborative effort among utilities, PV [Online]. Available: http://rredc.nrel.gov/solar/new_data/
inverter manufacturers, and industry experts involved in variability/Documentation/ASES_47760_final.pdf
this area also focuses on investigating and suggesting new
control capabilities through PV inverters. The next genera- Biographies
tion of PV inverters is intended to provide a variety of new Farid Katiraei is with Quanta Technology.
Julio Romero Agero is with Quanta Technology.
control features (e.g., voltage regulation, power curtailment, ramp-rate control, and communication-assisted prop&e

may/june 2011

IEEE power & energy magazine

71

You might also like