See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/3268366

Effect of Load Models in Distributed Generation Planning

Article in Power Systems, IEEE Transactions on · December 2007

DOI: 10.1109/TPWRS.2007.907582 · Source: IEEE Xplore

CITATIONS READS
396 1,880

2 authors:

D. Singh Rakesh Misra

Indian Institute of Technology BHU Indian Institute of Technology BHU
78 PUBLICATIONS 2,074 CITATIONS 37 PUBLICATIONS 1,189 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by D. Singh on 07 August 2015.

The user has requested enhancement of the downloaded file.

2204 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 22, NO. 4, NOVEMBER 2007

Effect of Load Models in Distributed

Generation Planning
Devender Singh, R. K. Misra, and Deependra Singh

Abstract—The effect of load models on distributed generation System real and reactive power losses
(DG) planning in distribution system is investigated in this work.
It is shown that load models can significantly affect the DG plan- , Real and reactive power losses with DG
ning. Normally a constant power (real and reactive) load model
is assumed in most of the studies. Such assumptions may lead to MVA flow in line
inconsistent and misleading results about deferral values, loss re- Voltage of th node
duction, payback period, and other subsequent calculations. It has
been demonstrated that DG planning based on such assumptions WDG System with DG.
would not be effective after implementation. It is shown that load
models can significantly affect the optimal location and sizing of WODG System without DG.
DG resources in distribution systems. A comparative study of real
and reactive power loss, real and reactive power intake at the main
substation and MVA support provided by installing DG resources
I. INTRODUCTION
for different type of loads models has been performed.
Index Terms—Distributed generation, distribution system, load
models. P OWER deregulation and restructuring have created in-
creased interest in distributed generation (DG), which is
expected to play an increasingly important role in the electric
power system infrastructure planning and market operations.
NOMENCLATURE Distributed generation is power source that can be connected to
a distribution network by a distribution company (DISCO) at
Voltage exponents of real and reactive any node or by the customer at the customer side of the meter.
loads. Distributed resources are strategically located and operated
in the system to defer or eliminate system upgrades, improve
Const., Res. Constant and residential load models. voltage profile, reduce system losses, reinforce grid, and to
Ind., Com. Industrial and commercial load models. improve system reliability and efficiency. Recent studies have
predicted that by year 2010, distributed generation will account
Mix. Mixed load model. for up to 25% of all new generation [1]. In the last few years
MVA capacity of line - . there has been significant contribution to research in the field
of DG resource planning. Normally, DGs are integrated in the
MVA intake at bus 1.
existing distribution system and the planning studies have to
Total system MVA intake by DISCO. be performed for optimal location and size of DGs to yield
NV, NL Number of voltage and line limit violations. maximum benefits.
Y. G. Hegazy et al. [1] evaluated the effect of DGs controlled
Real and reactive load at bus at nominal by the customer on system power capacity to satisfy the total
voltage. system load and predicted the average amount of unsupplied de-
Total system real and reactive power mand for a given year. The authors used Monte Carlo approach
demands. to model the operating histories of the installed distributed gen-
erators. El-Khattam et al. [2] proposed a method of solving dis-
Total size of DGs. tributed generation planning problem (location and size) in dif-
Real and reactive power injections at bus . ferent utility scenarios as an optimization problem. The objec-
tive function was based on supply-demand chain which aimed
Real and reactive power intake at bus 1.
to minimize the investment and operating costs of local candi-
date DGs, payments toward purchasing the required extra power
Manuscript received March 14, 2007; revised May 24, 2007. Paper no. by the DISCO, payments toward loss compensation services, as
TPWRS-00173-2007. well as the investment cost of other chosen new facilities for dif-
Devender Singh and R. K. Misra are with the Electrical Engineering Depart-
ment, Institute of Technology, Banaras Hindu University, Varanasi, UP, India
ferent market scenarios. Wang et al. [3] proposed an analytical
(e-mail: dsingh@bhu.ac.in; rkmisra@bhu.ac.in). method to determine the best location of candidate DGs for min-
Deependra Singh is Lecturer in Electrical Engineering Department, Kamla imum loss configuration. A rigorous analysis for uniformly dis-
Nehru Institute of Technology, Sultanpur, UP, India. tributed, centrally distributed, and increasingly distributed loads
Color versions of one or more of the figures in this paper are available online
at http://ieeexplore.ieee.org. has been carried out for constant and time varying loads on a
Digital Object Identifier 10.1109/TPWRS.2007.907582 feeder. The proposed approach was non-iterative unlike power
0885-8950/$25.00 © 2007 IEEE
SINGH et al.: EFFECT OF LOAD MODELS IN DISTRIBUTED GENERATION PLANNING 2205

flow programs. Therefore, there is no convergence problem in- TABLE I

volved, and results could be obtained quickly. However, authors LOAD TYPES AND EXPONENT VALUES
have indicated that other constraints such as voltage and line
limits may affect the DG placement. Ochoa et al. [4] proposed
various indices to evaluate the impact of distributed generation
on distribution networks. The indices to measure the impacts
of DG introduction on loss reduction, voltage profile, current
carrying capacity of conductors, and short circuit currents for
three-phase and single-phase ground faults are proposed. Fi- the results. In the practical situation, loads are not explicitly res-
nally, a multi-objective function is derived for such planning idential, industrial, and commercial; rather, load class mix may
studies. Chiradeja and Ramakumar [5] derived indices to mea- be seen by distribution system depending on the nature of area
sure the technical benefits in terms of voltage profile improve- being supplied. Therefore, a load class mix of residential, indus-
ment indices (VPII), line loss reduction indices (LLRI), and trial and commercial load has also been investigated.
emission improvement. Finally the authors devised a composite
index by combining the various indices. II. LOAD MODEL-BASED TEST CASES
Load models are well known in stability studies [6]–[9]. To quantify the effect of various load models on distributed
These studies are normally aimed toward voltage or frequency generation planning, a 38-node distribution system is adopted
dependent representation of load especially for dynamic or (Fig. 9). The data for p.u. line impedances, load data and the
static stability studies of power systems. Exhaustive review line MVA limits are given in the Appendix (Table VIII). The ef-
of load model to be used for power flow and dynamic studies fects of selected voltage dependent load models are investigated
has been presented in [7]. The effect of load models in in different planning scenarios (test cases). Practical voltage de-
planning studies was demonstrated in optimal capacitor place- pendent load models, i.e., residential, industrial, and commer-
ment/switching by Rizy et al. [10] in a distribution system and cial, given in [7] have been adopted for investigations. The load
by Arnborg et al. [11] in under-voltage load shedding studies. models can be mathematically expressed as
The authors demonstrated that consistent with the modeling of
feeder load as constant power, it was expected that the switching (1)
of capacitor bank to improve the power factor would result in (2)
decreased real and reactive power injections at the substation.
This decrease in real and reactive power injections is due to In a constant power model, conventionally used in power flow
improved voltage profile. In fact the measured real and reactive studies, is assumed. The values of the real and reac-
power injections increase. In this case, contrary to the constant tive exponents used in the present work for industrial, residen-
power load modeling, a reduction in real and reactive power tial, and commercial loads are give in Table I [7].
injection reflecting the reduced line losses is not observed at During investigations the comparison of constant power load
the substation. An assumption of constant power load modeling model assumption with the practical load models are empha-
leads to general misunderstanding of reduced power injection sized. While investigating effect of residential load, the 38-node
due to reduced losses. Analysis of the experimental results system is assumed to be supplying residential consumers only
[10] using voltage sensitive load model showed that while the (all loads are residential type). Similarly, for industrial and com-
feeder losses are reduced following the capacitor placement, the mercial load, it is assumed that all the loads are industrial and
attendant improvement in voltage profile results in an increase commercial type, respectively. In a practical situation, a load
in loads that exceeds the amount of loss reduction. class mix may be seen; therefore, a load class mix of residen-
In context of optimization, problem of DG placement is sim- tial, commercial and industrial loads is also adopted. This is in-
ilar in nature to that of capacitor placement discussed above. dicated in Table VIII along with the load data. The following test
Most of the planning methods invariably use power flow pro- cases are developed in which DG locations and sizes for case 1
grams which normally utilize constant real and reactive power and case 2 have been chosen arbitrarily.
load model representation. It is observed from the literature re- Case 1) Single DG Case: In the given load scenario, cus-
view that load models are not included in planning the location tomer decides to embed a single distributed gener-
and size, and calculating the said indices except Gozel et al. [12] ation resource of 0.8 p.u. on bus 14. The effect of
who have used analytical approach considering the load model load models on the following studies is to be made:
to show that the location does not change as significantly as size a) number of voltage limit violation (NV);
of DG. However, the analysis did not take the constraints of b) number of line limit violations (NL);
voltage and line limits and therefore has only instructive value. c) MVA , real p.u. demand ,
In this paper, a detailed study of effects of load models in DG and reactive p.u. demand on the
planning is made to asses the technical impacts and feasibility main substation;
of DG planning. Extent up to which load models can affect the d) loss reduction in real and reactive power;
location and size of the DGs in such studies has also been in- e) saving of MVA capacity on the main substa-
vestigated. The voltage dependent load models for residential, tion.
industrial, and commercial loads are adopted from [7]. A 38-bus Case 2) Multiple DG Case: In the given load scenario cus-
radial distribution system from [13] is utilized for demonstrating tomers decide to embed three distributed generators
2206 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 22, NO. 4, NOVEMBER 2007

Fig. 1. Flowchart for creation of database.

Fig. 2. Flowchart for selecting the minimum loss configuration from the data-
base.

of 0.20, 0.30, and 0.20 p.u. on buses 14, 24, and 32,
respectively. The effect of load models on the fol- the flowchart given in Fig. 1. The bus voltages occurring outside
lowing studies is to be made: the range 0.95–1.03 p.u. are treated as voltage violation.
a) NV; The flowchart for selecting the minimum loss configuration,
b) NL; for selected load type from the database is provided in Fig. 2.
c) . and on the main sub- A similar algorithm is used for minimum MVA criterion also.
station; Solutions with line limits and voltage violations are filtered out.
d) loss reduction in real and reactive power;
e) saving of MVA capacity on the main substa-
tion.
IV. SIMULATION RESULTS
Case 3) In the given load scenario, DISCO decides to embed
a distributed generator of optimal size at optimal The summary of simulation results obtained for various test
location (without undergoing any system upgrades) cases is presented for different types of load models in this sec-
for tion. The total system MVA intake by the DISCO, , is
a) loss reduction or defined as
b) reduction of the main substation MVA intake
(system upgrade deferment).
The following studies are to be made. (3)
i) Optimal DG location and size planning: Effect of load
models on the optimal location and size of distributed The differences in all the relevant quantities for the system
generators are studied. with and without DG are indicated for different load types.
ii) Error in Load Model: Suppose planning for optimal size Numbers of voltage and line limit violations have also been
and location for cases (a) and (b) has been done assuming presented. Further, to compare the results, non-normalized
constant power load. Now, technical feasibility of already versions of indices based on [4] are also developed. These
planned location and size is studied if loads in the system indices are defined in the following paragraphs.
are actually non-constant type. Real and Reactive Power Loss Indices (ILP and ILQ): The
real and reactive power loss indices are defined as

III. DATA PREPARATION

(4)
Power flow solutions for the 38-node distribution system are
obtained in the following fashion. DG size is considered in a (5)
practical range (0-4.00 p.u), decided by the total system demand
which is 3.9093 p.u. The DG of 0.0 p.u. corresponds to system The lower the values, the better the benefits in terms of loss
without DG whereas 4.00 p.u corresponds to a case when all reduction accrued to DG location and size.
the real power requirements are met by DG. It is considered Voltage Profile Index (IVD): It is related to the maximum
that the DG is operated at unity p.f. Each bus of the system voltage drop which, in this case, considers the maximum drop
is considered for the placement of DG of given size from the between each node and the root node. This index could also
range considered. Then load flow program is run for each of the be used to find prohibitive locations for DG considering pre-
cases. The complete procedure to create database is presented in established voltage drop limits. In this way, the lower the index,
SINGH et al.: EFFECT OF LOAD MODELS IN DISTRIBUTED GENERATION PLANNING 2207

TABLE II
SUMMARY OF RESULTS OBTAINED FOR CASE 1: SINGLE DG CASE

Fig. 3. ILP and ILQ for different load models (Case 1).

Fig. 4. IVDs and dIVD for different load models (Case 1).

the better the network performance. The IVD can be defined as

follows:

Fig. 5. IC for different load models (Case 1).

(6)

Residential Load Model: A similar trend of reduction with

MVA Capacity Index (IC): As a consequence of supplying
DG placement is observed for , and ex-
power near to loads, MVA flows may diminish in some sections
cept for . As opposed to the case of constant load model,
of the network, thus releasing more capacity, but in other sec-
in this case, despite the addition of a DG, there is increase in
tions, they may also increase to levels beyond distribution line
from the main substation. All the above reductions are
limits. This index gives important information about the level
significantly smaller than those of constant load model. As far
of MVA flows/currents through the network regarding the max-
as at the main substation is concerned, it is seen that in
imum capacity of conductors. This gives the information about
case of residential load model, placement of DG raises
need of system line upgrades. Lower values of the index indicate
(by 0.0567 p.u.) whereas in case of constant load model, the
more amount of capacity available. Line overloads are indicated
is decreased by 0.0419 p.u. which is almost of the same
by index values above 100%
order.
There is one line limit violation (line-7), in this case when DG
is placed in the system as against the case of constant power load
(7)
model.
Industrial Load Model: The trend of reduction in
, and with DG placement, as observed
A. Case 1: Single DG Case: Distributed Generation Resource for constant and residential load model, is maintained. These
of 0.8 p.u. on Bus-14 reductions are significantly smaller than those of constant load
The summary of results obtained in this case for different load model and marginally larger than those of residential load
models is depicted in Table II. The various indices as discussed models. Further, the increase in is distinctly higher than
above are presented in Figs. 3–5. those observed for residential as well as constant load models.
Constant Power Load Model: From Table II, it is observed There are no voltage as well as line limit violations for the
that there is decrease in , and at the system with and without DG.
main substation due to placement of DG. Furthermore, and Commercial Load Model: In this case, the reductions in
also get reduced appreciably. The overall system MVA, all the quantities indicated in Table II except the are
, of the DISCO is reduced by 0.0751. However, there having their lowest values compared to all the other load
are no voltage as well as line limit violations for both the models. Further, the total real power demand of the DISCO
situations, i.e., with and without DG. ( p.u.) with DG is
2208 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 22, NO. 4, NOVEMBER 2007

marginally higher than that for system without DG, i.e., 3.7987 TABLE III
p.u. The and may in fact increase, despite the SUMMARY OF RESULTS OBTAINED FOR CASE 2: MULTIPLE DG CASE
reduction in and , taking advantage of loss reduction
void. This is due to improvement in voltage profile which
results in an increase in loads that exceeds the amount of loss
reduction. In such a case, if the DISCO happens to pay same
amount/MWh, it loses instead of gaining from DG placement.
This may not be detectable if proper load model is not taken
into account. There is one line limit violation (line-7) when DG
is placed.
Mixed Load Model: The reductions in , and
are less compared to constant load model. The increase
in is also observed as in case of other load models ex-
cept for constant load model. A line limit violation (line-7) is
observed in the system.
The real and reactive loss indices, ILP and ILQ, are depicted
in Fig. 3. It can be observed that the indices indicating the effect
of DG placement are low in case of constant power load model
assumption, whereas for non-constant power load model, these
are higher. Hence, planning studies for evaluating the advantage
of DG placement based on constant power load model may in-
dicate lower system losses.
The IVDs and the difference between IVDs (dIVD) of system
with and without DG are plotted in Fig. 4 indicating the im-
provement in voltage profile due to DG placement. In this case in from the main substation, unlike the case of constant
also, the constant power load model assumption gives a picture power load model.
of higher improvement as compared to actual system (non-con- The reductions in , and are signifi-
stant power load model). cantly smaller than those obtained for constant load model. As
The IC indices depicting the used line MVA capacity in the far as at the main substation is concerned, placement
system with and without DG for different load models are shown of DG raises reactive demand (by 0.0341 p.u.) unlike the case
in Fig. 5. For the system without DG, the used/available ca- of constant load model, where the is decreased almost
pacities for different load models are not much different. How- by same order (by 0.0353 p.u.). Also, unlike the constant power
ever, when DG is placed, IC values for non-constant power load load assumption, the system and do not directly reflect
model are significantly different from that of constant power into the and .
load model. In this case, in an actual system, constant power Industrial Load Model: It can be observed that the trend of re-
load model assumption gives false indication of high available duction in , and with placement of DGs
MVA capacity, whereas the fact is otherwise. is maintained in this case also. However, the reduction in
demand is more than the residential load model but less than
constant load model assumption. Further the increase in
B. Case 2: Multiple DG Case: Three Distributed Generators is almost doubled and is distinctly higher than those observed
of 0.20, 0.30, and 0.20 p.u. ( p.u.) on Buses 14, for residential as well as constant load model.
24, and 32, Respectively Commercial Load Model: The reductions in all the quan-
tities indicated in Table III except the and
The summary of results obtained in this case for different load are having their lowest values compared to all other load
models is depicted in Table III. The differences in all relevant models. Further, total real power demand of the DISCO
quantities for the system with DG and without DG are also in- ( p.u.) with DGs is
dicated for different load types. almost same to that of system without DG, i.e., 3.7987 p.u.
Constant Power Load Model: From Table III, it is observed The has increased marginally which is not
that there is decrease in , and due observed in previously studied load models.
to placement of DGs. Furthermore, MVA of the DISCO also Mixed Load Model: In this case, the reductions in
gets reduced. Hence, when constant load model is assumed, the , and are less compared to constant
reduction in and is directly translated as reduction in load model. The increase in is observed in this case
overall system MVA which gives erroneous indica- also as in case of other load non-constant load models.
tion of higher benefits of DG placement. This is because the The ILP and ILQ indices are depicted in Fig. 6. Similar to the
effect of voltage profile on loads is not taken into account. single DG case, the indices indicating the effect of DG place-
Residential Load Model: A similar trend of reduction in ment are low in case of constant power load model assumption,
, and except in is observed with whereas for non-constant power load model (except for the in-
DG placement. Despite the addition of DGs, there is increase dustrial load model), these are normally higher. The load models
SINGH et al.: EFFECT OF LOAD MODELS IN DISTRIBUTED GENERATION PLANNING 2209

TABLE IV
MINIMUM LOSS CONFIGURATION FOR DIFFERENT LOAD MODELS
FOR CASE 3(I): OPTIMAL LOCATION AND SIZE PLANNING

Fig. 6. ILP and ILQ for different load models (Case 2).

TABLE V
MINIMUM MVA CONFIGURATION FOR DIFFERENT LOAD MODELS
FOR CASE 3(I): OPTIMAL LOCATION AND SIZE PLANNING

Fig. 7. IVD for different load models (Case 2).

(0.1889 p.u., in Table III). The ,

and have also decreased consistent with constant
power load model assumption.
From Table V, it is seen that location of DG for minimum
(to defer substation upgrades) is same (bus 6) to that of
Fig. 8. IC for different load models (Case 2).
minimum loss criterion, of course, with different optimal
(3.6050 p.u.). It also establishes the fact that for a constant power
significantly affect the studies for evaluating the advantage of
load model, the reduced and are translated into reduced
DG placement.
, and therefore, the change in location may be minimal
The IVDs and the differences between the IVDs (dIVD) of
when optimality criterion is changed from minimum to min-
the system with and without DGs, indicating the improvement
imum .
in voltage profile due to DG placement, are plotted in Fig. 7.
Industrial Load Model: In this case, minimum configura-
Similar to the single DG case, the constant power load model
tion (Table IV) requires optimal to be 0.94 on bus 12. The
assumption gives a picture of higher improvement as compared
size and location are significantly different than that for constant
to actual system (non-constant power load model).
power and residential load models. The minimum which can
The IC indices depicting the used MVA capacity in the system
be achieved by placing DG (optimally) is 0.1093 p.u. as com-
with and without DGs for different load models are shown in
pared to 0.0973 p.u. for constant power load model. Also the
Fig. 8. Compared to the system with DGs, the utilized line ca-
is marginally higher than that of constant power load
pacities for different load models are not much different for the
model.
system without DG. However unlike single DG case, studies
In case of minimum configuration (Table V), optimal
based on constant power load model assumption give false in-
location is same to that of residential load but quite different
dication of low available MVA capacity.
from that of constant power load model assumption. Similarly,
optimal size of DG is close to that of residential load model and
C. Case 3 (i): Optimal DG Location and Size Planning
is significantly different than that of constant load model.
The summary of results obtained for optimal and Commercial and Mixed Load Models: For these load models,
location for minimum loss (minimum ) configuration is minimum configuration (Table IV) requires optimal to
given in Table IV and the same for minimum MVA (minimum be 0.365 p.u. (Com.) and 0.375 p.u. (Mix.) on bus 16.
) is given in Table V. For different load models, the It is observed that the is remarkably different from those
optimal (without system upgrade) and corresponding of load models considered so far. The location is having the dif-
, and are also indi- ference of two, four, and ten nodes when compared to that of res-
cated in Tables IV and V. idential, industrial, and constant load models, respectively. This
Constant Power Load Model: For minimum configu- location is farthest from the root node compared to the other load
ration (Table IV), optimal is 2.6 p.u. at bus 6. The models. The minimum is significantly higher compared to
is reduced to 0.0973 p.u. as compared to that without DG other load models. The and for commercial and
2210 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 22, NO. 4, NOVEMBER 2007

mixed load models are significantly higher than that of constant TABLE VI
power load model. ACTUAL LOAD MODEL SCENARIO FOR MINIMUM LOSS CONFIGURATION
(DG OF 2.6 P.U. AT BUS 6) FOR CASE 3(II): ERROR IN LOAD MODEL
The minimum configuration (Table V) requires op-
timal to be 2.3250 (Com.) and 2.4350 (Mix) on bus 2. This
shows that the DG locations are entirely different for minimum
and minimum criteria for non-constant power load
models. Also the minimum criterion gives same bus lo-
cation (bus 2) for non-constant power load models.
This is explained as follows. Assuming constant power load
model for system without DG z

(8) TABLE VII

ACTUAL LOAD MODEL SCENARIO FOR MINIMUM MVA CONFIGURATION
and (DG OF 3.605 P.U. AT BUS 6) FOR CASE 3 (II): ERROR IN LOAD MODEL
(9)

Now assuming that DG is placed which is reducing the losses

and ,

(10)
and
(11)

Since constant power load assumption makes and in- estimate as far as losses are concerned. However, more impor-
sensitive to DG placement; for and to be min- tant is the fact that and from the main substa-
imum, and are to be minimized, which is same as tion increase. This nullifies the advantage of loss reduction. Fur-
minimum criterion. However, in case of non-constant power ther getting conservative estimate of losses for actual scenario
load models, and also become sensitive to DG place- obtained assuming constant power load model is also rendered
ment (different DG placement gives different voltage profiles), meaningless. Compared to the case of planned scenario, in case
apart from and . In such cases, the line limit con- of non-constant load models, , total system real power
straints become more important than the losses because values intake , and increase which is due to
of losses are negligible as compared to and . Thus, when increased and .
non-constant load models are used, the DG locations in gen- Considering the optimal solution of DG placement for min-
eral are near to the main substation where the line limits are imum ( of 3.605 p.u. at bus-6) obtained using con-
maximum compared to other lines of the distribution system. stant power load model assumption and implementing it on ac-
In order to minimize the maintaining the line limits, it tual system exhibiting non-constant power load behavior, the re-
is suitable to place the DG at that node where maximum line sults obtained are summarized in Table VII. In this case also, the
MVA limit is found, i.e., the line from main substation to the given location and size is not suitable as far as line limits are con-
next node, i.e., bus 2. Hence, in all the non-constant power load cerned. Also the optimal for non-constant load models
models, the optimum location for minimum comes out is higher than that obtained using constant load model. Imple-
to be bus-2. Since in this study, the bus-1 was not considered to menting the solution based on the constant power load model
be a candidate location, the next bus, the bus-2, was selected. assumption on the network exhibiting non-constant power load
behavior, in general, reduces the assumed benefits accrued to
D. Case 3 (ii): Error in Load Model DG placement.
This part of the work shows that DG location and size for
a system, evaluated assuming constant power load model will V. DISCUSSION
not reflect the planned benefits, when put into actual system. In this work, only voltage dependent load models are
The optimum location and size, for constant power load model, considered for DG planning studies. Frequency dependent
under minimum configuration (Table IV) were bus 6 and 2.6 models have not been considered due to the following reason.
p.u., respectively. It is depicted in Table VI that placing a DG System voltages are local phenomena whereas frequency is a
of given location and size in an actual load model scenario may pan-system phenomenon. This means that the bus voltage, up
lead to line violations. This holds for all the non-constant power to an extent, can be controlled by real and reactive load and
load models. generation at a particular bus. However, the frequency does not
In fact, and are lower than those obtained during plan- show such a characteristic and therefore cannot be controlled
ning study performed assuming constant load model. This may locally and remains same for whole of the system. Thus, the
indicate that the constant power load model gives conservative change in frequency dependent loads is not due to presence or
SINGH et al.: EFFECT OF LOAD MODELS IN DISTRIBUTED GENERATION PLANNING 2211

absence of DG. Therefore, even if we model the load to be fre- TABLE VIII
quency dependent, the differences obtained in , SYSTEM AND LOAD DATA FOR 38-NODE SYSTEM
etc. cannot be attributed to DG placement.
Generally, the line limits are reached due to counter line
flows which are normally observed in systems with DGs. In this
work, no line over-loadings due to counter flows are observed
in case of constant as well as non-constant load models. The
overloading of line-7 (Tables II, VI, and VII) for non-constant
power load models is explained as follows. In case of non-con-
stant power load models, when DG is placed, the voltage profile
of the system changes (same is the case for constant power
load model also) and the system real and reactive loads change
due to their voltage dependence (unlike constant power load
model). Therefore, the flows in the lines normally increase to
supply the increased bus loadings. In fact, the overload of line-7
is due to increased flow (in the same direction) and not due to
counter flow.
It may appear that assuming low (10%–15%) DG penetra-
tions, the differences reported in this work, may not be signifi-
cant. The following investigations are carried out to clarify that
it is not so.
Let us assume that location for DG of 0.3750 p.u. ( % of
the ) is to be obtained for the following cases.
a) To Minimize MVA Requirement : The location
for constant power load model is bus-16 (as in case of minimum
configuration), but for mixed load, it is bus-2. The location
and the relevant quantities such as and
are significantly different.
b) To Minimize the System Losses: The location for con-
stant power load model and mixed load model is found to be
same, i.e., bus-16 (Table IV). However, the for constant load
model comes out to be 0.1500 p.u. (reduction of 0.0389 p.u.)
whereas, for the mixed load model, it is 0.1337 p.u. (reduction
of 0.0325 p.u.) which is a significant difference (around 19%).
Similarly, for constant load model is 0.0.0997 p.u. (showing
loss reduction of 0.0268 p.u.), whereas for the mixed load it is
0.0879 p.u. (reduction of 0.0225 p.u.) which is also significantly
different (around 19%). Similarly, the reductions in are
0.3612 and 0.3164 for constant load and mixed load, respec-
tively.
In the present work, high values of DG penetration appear due
to optimization of certain parameters (minimum /minimum
). These high values suggest that planning studies show
a significantly higher possible optimal DG penetration levels
when the constant power load models are assumed instead of
actual load models.
Though the investigations are performed on radial system, the
findings are relevant to all topologies. Fig. 9. The 38-node test system.

and size of DG resources in distribution system. A rise in reac-

VI. CONCLUSION
tive demand is observed when DG is placed in an actual system
Load models significantly affect the DG planning. It is estab- having non-constant power load model, which is contrary to the
lished that DG planning based on constant power load models observations in case of constant power load model assumption.
is not effective after implementation on actual systems. Deci- It is shown that though the losses get reduced, the total power
sions and results based on constant power load model assump- intake of the system increases, which makes loss reduction re-
tions are not technically feasible if employed on system having dundant. It is established in this work that the quantities affected
non-constant power load behavior. Also in the findings, a sig- are 1) number of voltage limit violation; 2) number of line limit
nificant effect of load models is observed on optimal location violations; 3) MVA, real, and reactive p.u. demand on the main
2212 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 22, NO. 4, NOVEMBER 2007

substation; 4) loss reduction in real and reactive power; 5) saving [12] T. Gozel, M. H. Hocaoglu, U. Eminoglu, and A. Balikci, “Optimal
of MVA capacity on the main substation; and 6) optimal location placement and sizing of distributed generation on radial feeder with
different static load models,” in Proc. Int. Conf. Future Power Systems,
and size of DG. It was observed that most or all of the quantities Nov. 2005, pp. 1–6.
are affected. [13] M. E. Baran and F. F. Wu, “Network reconfiguration in distribution
systems for loss reduction,” IEEE Trans. Power Del., vol. 4, no. 2, pp.
APPENDIX 1401–1407, Apr. 1989.

Table VIII has the system and load data for the 38-node
Devender Singh received the B.E. degree in elec-
system. Fig. 9 shows the 38-node test system. trical engineering from Sardar Vallabhbhai Regional
College of Engineering and Technolgy, Surat, India,
REFERENCES in 1993, the M.E. degree in electrical engineering
[1] Y. G. Hegazy, M. M. A. Salama, and A. Y. Chikhani, “Adequacy as- from Motilal Nehru Regional Engineering College,
sessment of distributed generation systems using Monte Carlo simula- Allahabad, India, in 1999, and Ph.D. in electrical
tion,” IEEE Trans. Power Syst., vol. 18, no. 1, pp. 48–52, Feb. 2003. engineering from Institute of Technology (IT),
[2] W. El-Khattam, Y. G. Hegazy, and M. M. A. Salama, “An integrated Banaras Hindu University (BHU), Varanasi, India.
distributed generation optimization model for distribution system plan- Presently, he is Reader in the Department of Elec-
ning,” IEEE Trans. Power Syst., vol. 20, no. 2, pp. 1158–1165, May trical Engineering, IT, BHU. His research interests
2005. are distribution generation planning, state estimation,
[3] C. Wang and M. H. Nehrir, “Analytical approaches for optimal place- short-term load forecasting, and AI applications in power systems.
ment of distributed generation sources in power systems,” IEEE Trans.
Power Syst., vol. 19, no. 4, pp. 2068–2076, Nov. 2004.
[4] L. F. Ochoa, A. Padilha-Feltrin, and G. P. Harrison, “Evaluating dis-
tributed generation impacts with a multiobjective index,” IEEE Trans. R. K. Misra received the B.Sc. (Engg.) degree in
Power Del., vol. 21, no. 3, pp. 1452–1458, Jul. 2006. electrical engineering and the M.Tech. degree in en-
[5] P. Chiradeja and R. Ramakumar, “An approach to quantify the tech- gineering systems from Dayalbagh Educational In-
nical benefits of distributed generation,” IEEE Trans. Energy Convers., stitute, Agra, India, in 1995 and 1997, respectively,
vol. 19, no. 4, pp. 764–773, Dec. 2004. and the Ph.D. degree in electrical engineering from
[6] C. Concordia and S. Ihara, “Load representation in power systems sta- Institute of Technology (IT), Banaras Hindu Univer-
bility studies,” IEEE Trans. Power App. Syst., vol. PAS-101, no. 4, pp. sity (BHU), Varanasi, India.
969–977, Apr. 1982. Presently, he is Reader in the Department of Elec-
[7] IEEE Task Force on Load Representation for Dynamic Performance, trical Engineering, IT, BHU. His research interests
“Bibliography on load models for power flow and dynamic per- are distribution generation planning, power system
formance simulation,” IEEE Trans. Power Syst., vol. 10, no. 1, pp. security, and AI applications in power systems.
523–538, Feb. 1995.
[8] IEEE Task Force on Load Representation for Dynamic Performance,
“Load representation for dynamic performance analysis,” IEEE Trans.
Power Syst., vol. 8, no. 2, pp. 472–482, May 1993. Deependra Singh received the B.Tech. degree in
[9] IEEE Task Force on Load Representation for Dynamic Performance, electrical engineering from Harcourt Butler Techno-
“Standard load models for power flow and dynamic performance sim- logical Institute, Kanpur, India, in 1997 and the M.E.
ulation,” IEEE Trans. Power Syst., vol. 10, no. 3, pp. 1302–1313, Aug. degree in electrical engineering from the University
1995. of Roorkee, Roorkee, India, in 1999. Currently, he
[10] D. T. Rizy, J. S. Lawler, J. B. Patten, and W. R. Nelson, “Measuring and is pursuing the Ph.D. degree from UP Technical
analyzing the impact of voltage and capacitor control with high speed University, Lucknow, India.
data acquisition,” IEEE Trans. Power Del., vol. 4, no. 1, pp. 704–714, He is a Lecturer in the Department of Electrical
Jan. 1989. Engineering, Kamla Nehru Institute of Technology,
[11] S. Arnborg, G. Anderson, D. J. Hill, and I. A. Hiskens, “On influence of Sultanpur (UP), India. His research interests are dis-
load modelling for undervoltage load shedding studies,” IEEE Trans. tributed generation planning and distribution system
Power Syst., vol. 13, no. 2, pp. 395–400, May 1998. analysis.

View publication stats