Technology and Economics of Smart Grids and Sustainable Energy (2021) 6:3

https://doi.org/10.1007/s40866-020-00101-8

ORIGINAL PAPER

Linear Distribution System State Estimation with Integration of DG

S. Bindu 1 & S. Ushakumari 2 & J. S. Savier 3

Received: 29 May 2020 / Accepted: 22 December 2020

# The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. part of Springer Nature 2021

Abstract
Monitoring of a power system is essential for maintaining it in a secure operating condition. Electric distribution management
system (DMS) should possess a monitoring mechanism addressing the entire operating conditions of the system. Distribution
System State Estimation (DSSE) is an integral building block of the DMS. Because of new technologies integrated into the power
distribution system, new opportunities, problems and challenges related to control and management are being explored. To make
efficient and reliable control actions, an accurate knowledge of the system states is required. This paper presents Linear State
Estimator based DSSE which can handle the challenges mentioned above. The estimator studied and reported in the paper allows
to include synchronized phasor measurements provided by Phasor Measurement Units (PMUs). With the increasing integration
of DG in distribution system, the state estimation scheme needs be reworked to address the complexities introduced by power
exchange of DG. To address this, Linear State Estimation (LSE) is reformulated and its performance is analysed with DG
integrated in the sample network under study.

Keywords Power system . Distribution system state estimation . Linear state estimator . Distributed generation . Phasor
measurement unit

Introduction and economic reasons. Transmission systems have sufficient

measurements and many of them are sufficiently redundant
Monitoring in real time is highly essential for implementing making systems always visible.
real time control and protection functions of a distribution In the case of distribution systems, real time measure-
electric network. Available measurements are to be used prop- ments are less and therefore the system is not observable.
erly so that the best approximate of the system state which fits Some data, known as pseudo measurements, are used in
the measurements can be obtained. DSSE is treated as key order make the system observable. The use of PMUs in
component of DMS. Transmission System State Estimation distribution systems give accurate information about volt-
(TSSE) is well developed over the years. Due to inherent and age magnitude and phase angle which in turn can improve
fundamental differences in transmission and distribution sys- the estimation accuracy of DSSE. In a modern-day distri-
tems, TSSE cannot be directly applied for functional require- bution grid, the performance requirements of DSSE are be-
ments of DSSE. coming stringent. This is due to the challenges introduced
The unique characteristics of distribution system include by integration of DG and use of advanced technologies.
load unbalances, high R/X ratios and less observability due Ref. [1] summarizes the major obstacles and challenges
to inadequate measurements in real time. The main limitation associated with DSSE.
is the lesser number of measurements due to various practical Various methods of formulating the mathematical prob-
lem, application of pseudo measurements, placement of
metering instruments etc. are investigated in [2]. The con-
* S. Bindu ventional method and modern data-driven and probabilistic
[email protected]
techniques are reviewed. The role and significance of pseu-
do measurements due to limited real-time measurements
1
Government Engineering College, BartonHill, and network observability is emphasized in [3–7]. A hybrid
Thiruvananthapuram, India state estimator based on SCADA and using PMU measure-
2
TREST Research Park, Thiruvananthapuram, India ments are proposed in [8], which results in an improved and
3
College of Engineering Trivandrum, Thiruvananthapuram, India robust estimation.
 3 Page 2 of 7 Technol Econ Smart Grids Sustain Energy (2021) 6:3

Sizing and placement of Distributed Generation (DG) for reported that such algorithms bring in the challenge of in-
improved performance of a distribution system is proposed creased execution time.
in [9]. This paper also proposes techniques used to improve Weighted Least Square (WLS) technique is widely used in
adaptive capability of the network to handle extreme DSSE. But the instability issue, sensitivity to initial point etc.
events. Ref [10] explains optimal placement of DG in radial are the problems in this technique. Several methods are pro-
distribution network using evolutionary algorithms like posed to tackle the problems associated with the WLS meth-
Genetic Algorithm and Particle Swam Optimisation. A od. As a prospective and promising solution, development of a
multi-objective performance index-based sizing and place- LSE method and its performance are addressed in this paper.
ment of DG in distribution systems is studied in [11]. Load The method accounts for PMU measured data and measure-
is considered to be varying by considering the effect of ments by line current sensors. To address the issues related to
industrial, residential and commercial loads. The issues in insufficient number of measurements in a distribution net-
the allocation of DG is presented along with an analytical work, pseudo measurements are introduced using the estimat-
method for the allocation of DG in radial distribution sys- ed values from a linear load flow. The linear load flow is opted
tems in [12]. since it is easy to implement and accurate. A linearization
Various aspects of State Estimation (SE) of a distribution method based on the Taylor’s series approximation is adopted
grid, which is observable, using Weighted Least Square and in the load flow. These measurements, along with the avail-
Extended Kalman Filter are investigated in [13]. The main able actual measurements are used in LSE which eventually
disadvantage of these methods is that they need some prior brings out the system states. LSE is non iterative and over-
knowledge about the state of the system for a satisfactory comes the issues associated with WLS method. From the re-
performance. sults, it can be seen that the LSE can be a good choice when
Methods like Branch-Current-based SE (BCSE) take infor- the electric network has a limited number of actual measure-
mation from network structure and the nature of the measure- ments and there are pseudo-measurements which are general-
ments. An efficient and robust branch current based DSSE ly inaccurate as compared to the actual measurements. The
with minimum storage requirements is presented in [14]. performance of the SE method proposed is also tested with
The method of including phasor measurement data in post DG integrated at specific nodes of the sample network under
processing mode in transmission systems is discussed in study.
[15] whereas [16] presents a DSSE method which uses only LSE methodology proposed is covered in detail in “Linear
few measurements and gives a SE with improved accuracy. Distribution State Estimation (LSE)” section of this paper.
Singular Value Decomposition (SVD), Principal Component The sizing and placement of DG and LSE methodology inte-
Analysis (PCA) algorithms for data compression in SE and grating DG are discussed in “LSE in the Presence of DG”
BCSE method to find the system states for the distribution section. In “Results” section, the simulation results of the pro-
network are addressed in [17]. posed method on an IEEE 33 bus is discussed. Fifth section
Various types of estimators are explained in [18] where the gives the conclusions of the work carried out.
importance of real time monitoring, accurate estimation of
network topology and timely identification of the topology
changes etc. are emphasized. SE with microgrids is studied Linear Distribution State Estimation (LSE)
in [19]. A recursive Bayesian estimator is developed and re-
ported in [20]. A normalized residual test for detection of To find out the true values in the network, a linear load flow
network topology change is studied in [21]. The Branch for three-phase power distribution systems is used. ZIP model
Current based SE (BCSE) and the challenges in DSSE are of the loads are considered along with balanced and unbal-
investigated in [22]. The SE algorithm explained in [23] refers anced loads. The method is very simple and accurate. The
to a compensation method to cope with the low redundancy of problem of high R/X ratio is addressed by using a linear
measurement sets. Even though this method prevents the ill- approximation.
conditioning of the gain matrix, the solution obtained suffers The admittance matrix relating nodal voltages and currents
from low accuracy. is as follows:
Ref. [24] addresses the problem of minimum number of



measurements and discusses semidefinite programming tech- IA Y AA Y AB VA
¼ ð1Þ
nique for DSSE to deal with this problem. Augmentation of IB Y BA Y BB VB
pseudo measurements using ANN based forecasting is used
for tackling challenges due to limited measurements in distri- The current in each node is related to the voltage as:
bution system as reported in [25]. The historical data used for S * Pk
training the ANN is highly critical as erroneous forecasted Ik ¼ þ m:S * Ik þ m2 :S * Zk :V k ð2Þ
V *k
values will eventually result in inaccurate SE. It is also
Technol Econ Smart Grids Sustain Energy (2021) 6:3 Page 3 of 7 3

where m = 1/Vnominal specific pattern. This also is considered as a difficulty in ap-

A linear approximation for V is considered as plying WLS in DSSE problems.
These problems are handled in LSE. The solution of LSE
V ¼ f ðΔVÞ ¼ ð1−ΔVÞ can be obtained without iterations since the measurement
functions are linear.
when ΔV is very small. A linear form is obtained as in (3) by The measurement functions include node voltage, current
evaluating Taylor series around zero where higher order terms and complex power.
are neglected and V= (1- ΔV) is used. The node voltage is defined as V = a+ib where a is the
1 1 vector defining the real part and b is the vector defining the
¼ ≈1 þ ΔV ¼ 2−V ð3Þ imaginary part of the node voltage. Some of the node voltages
V 1−ΔV
are directly measured by PMUs which are included in SE to
An expression for the nodal current (2) in linear form is enhance the performance of state estimation.
obtained as below Line current sensors are used to measure the current vector.
The measurement model for line current measurement is,
M þ N :V * B þ Q:V B ¼ 0 ð4Þ  
I ij ¼ V i –V j Y ij ð11Þ
with, The complex current vector is represented as
M ¼ Y BA :V A −2m:S PB −m:S
* *
ð5Þ      
IB Re Iij ¼ ai −a j Gij − bi −b j Bij
 
N ¼ m2 :diag S * PB ð6Þ      
Im I ij ¼ ai −a j Bij þ bi −b j Gij ð12Þ
Q ¼ Y BB −m :S 2 *
ZB ð7Þ
where Gij and Bij are the real part and imaginary part of Yij, the
(4) can be solved using rectangular representation as fol- admittance of line connecting the ith and jth nodes in the
lows: network.



 Load measurements are usually obtained through load fore-
−M r N r þ Qr N i −Qi Vr casting or historical data regarding consumption.
¼ : ð8Þ
−M i N i þ Qi −N r þ Qr Vi The equation for power injection is,
where r and i indicate the real and imaginary parts respectively V i I *i ¼ Si ð13Þ
of the parameters in rectangular form.
Equation (8) is used to calculate the true values of voltage. where Ii* is the conjugate of the current at node i and Si is the
The magnitude of true value of current and power are derived apparent power at node i which is a non-linear function.
from this. Since the real measurements are limited in number, Taylor series based expansion is used in this work to make
pseudo measurements are substituted. The pseudo measure- the above equation linear.
ments are derived from true values by adding a random error Let the voltage at ith bus is denoted in per unit as,
to the actual values. These values are then used in LSE.
V i ¼ 1−ΔV i ð14Þ
The measurement equation for weighted least square esti-
mation is, By Taylor series expansion and approximation, it can be
z ¼ h1 ðxÞ þ e ð9Þ written from (3) that,
1
where x is the state vector and the relationship between mea- ¼ 2−V i ð15Þ
Vi
surements and state variables is denoted by h1(x). e is the
vector of measurement errors having a Gaussian distribution The error introduced by this approximation is very small.
with zero mean and covariance R. Substituting (15) in (13)
The WLS problem is then defined as,
I * i ¼ S i ð2V i Þ ð16Þ
Min J ðxÞ ¼ ½z−h1 ðxÞT R−1 ½z−h1 ðxÞ ð10Þ
Re writing Ii based on the node voltages, (16) can be writ-
Since h1(x) is a non-linear function, iterative techniques are ten as
required to solve (10). But these methods greatly depend on    
the initial guess and are prone to divergence, particularly when ∑ V i −V j Y ij ¼ S *i 2−V *i ð17Þ
j∈N i
the errors of various measurements vary significantly.
Moreover, loads in the distribution system, do not follow a where Ni is the set of all nodes linked to the node i.
 3 Page 4 of 7 Technol Econ Smart Grids Sustain Energy (2021) 6:3

By re-organising (17) the complex power can be separated network results in the decrease of the total power loss and
out as below: improvement of voltage profile of the system. Optimal place-
    ment of DG is based on an index called as Power Stability
∑ ai −a j Gij − bi −b j Bij þ ai Pi −bi Qi
jϵN i Index (PSI) [12]. PSI of bus m is expressed as below (21)
   
¼ 2Pi ∑ ai −a j Bij þ bi −b j Gij −ai Qi −bi Pi ¼ −2Qi 4Rmn ðPm −Pn Þ
jϵN i PSI ¼ ð21Þ
ð18Þ ½jV m jcosðθm −θn Þ2

Thus, considering all the measurement functions – node where Rmn is the resistance of the line connecting mth and nth
voltages, line currents and load, the system measurement buses, Pm and Pn are real power at bus m and bus n respec-
function will be tively, Magnitude of voltage at mth bus is denoted as Vm. θm
2 V3 2 V3 and θn are the angles at mth and nth bus respectively. PSI is
h1 z calculated for every bus in a network. The indices are then
4 h1 I 5 x ¼ 4 z I 5 ð19Þ arranged in descending order. DG can be placed in the bus
h1 S zS which is having highest PSI index for maximum advantage.
The sizing of the DG shall be compatible with the rating of the
which is linear. system. The power delivered from the DG should be absorbed
The state estimation can now be solved using (20) within the pre-defined distribution area. There is an optimum size
 −1 of the DG in a bus, at which the losses in the system is a mini-
x ¼ H 1 T WH 1 H 1 T Wz ð20Þ
mum. If the DG size is not the optimum value, the power flow
where H1 is the Jacobian matrix and W is the inverse of the co- limit exceeds and it will create losses. Thus in order to keep the
variance matrix and the errors are having a Gaussian distribu- losses to a minimum value, the DG size is fixed to an optimum
tion with zero mean. value. Analytical approach is used for optimum DG size selec-
tion [11]. The method is easy to implement and computationally
efficient. It is based on the minimization of losses in the system.
The total loss in the system is,
LSE in the Presence of DG
n n
PLoss ¼ ∑ ∑ ½Aik ðPi Pk þ Qi Qk Þ þ Bik ðPk Qi −Pi Qk Þ  ð22Þ
Placement and Sizing of DG i¼1 k¼1

DGs are an integral part in any modern day power distribution Minimizing the power loss function with reference to the
network. Improvement of voltage profile in electric distribu- incremental power injected at node i,
tion network is an important additional motive for integration
δPLoss
of DG. Placement of DG improves the voltage profile and the ¼0
δPi
stability of the system. The SE method must consider the
interconnection of DG in the network and the algorithm needs Then the real power injected at ith bus is,
to be modified accordingly. Placement and sizing of DG in
electric distribution systems is decided considering the influ- 1 n
Pi ¼ ∑ ½Aik Pk −Bik Qk  ð23Þ
ence of the inclusion on the operating parameters of the net- Aii k¼1
work. By optimally placing DG in the electric distribution

Fig. 1 Single line diagram of

IEEE 33 bus radial distribution
System
Technol Econ Smart Grids Sustain Energy (2021) 6:3 Page 5 of 7 3

Fig. 2 Voltage magnitudes for a

IEEE 33 bus radial distribution 1.02

Voltage magnitude (p.u.)

system estimated using LSE 1
method
0.98
0.96
0.94
0.92
0.9
0.88
0.86
0.84
0.82
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33
Bus number
True value Estimated value

Rik cosðδi −δk Þ State Estimation with DG

Aik ¼
V iV k
The performance of the new SE method is to be tested with
inclusion of DG in the network. The position of DG intercon-
and
nection and the rating of the resource are very critical when the
operation of the overall network is considered. The optimum
Rik sinðδi −δk Þ
Bik ¼ position for DG interconnection is found based on criteria de-
V iV k cided by PSI values at different nodes. The sizing of the DG is
carried out analytically as in “LSE in the Presence of DG”
Pk and Qk are the real and reactive powers respectively at section and is included with the state estimation algorithm.
kth node.
The real power generated by DG(PDgi) can then be formu-
lated as,
Results
Pi ¼ PDGi −PDi ð24Þ
The SE is performed on IEEE 33 bus (Fig. 1) radial distribu-
th
where PDi is the load on the i bus. tion system. The true values are calculated using a linear load

Fig. 3 Voltage angles for the

0.015
radial distribution System (IEEE
33 bus) estimated using LSE
method 0.01
Voltage angle (p.u.)

0.005

0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33

-0.005 Bus number

-0.01

-0.015
Esmated Value True value
 3 Page 6 of 7 Technol Econ Smart Grids Sustain Energy (2021) 6:3

Fig. 4 Error in Voltage 0.05

magnitudes for a 33 bus radial
0.045

Error in voltage magnitude(pu)

distribution system using LSE
method and WLS method 0.04
0.035
0.03
0.025
0.02
0.015
0.01
0.005
0

11

13

15

17

19

21

23

25

27

29

31

33
LSE WLS Bus number

flow method. These values are then used for the calculation of value of PSI is the highest in the eighth node and therefore the
pseudo measurements. DG is placed in that node. This is the optimal place for placing
All measurements including pseudo measurements, PMU the DG. The size of the DG is based on (23) which in turn
measurements and the line current measurements form the set depends on loss minimization in the system. The size of the
of measurements for the SE. DG is found to be 238.28 kW and is placed in node 8. LSE is
The results of the newly proposed SE are shown in Figs. 2 and performed by placing the DG on node 8. The voltage profile is
3. The plots of the voltage magnitude are shown in Fig. 2. shown in Fig. 6 and is found to improve with the integration of
Estimated values are found to follow the true values of voltage DG. The maximum improvement in voltage is found to be at
magnitude. The errors are found to be comparatively less with that node 18 and the losses are also reduced.
obtained from WLS method. Load flow analysis on the network
provides the true values of various parameters of the distribution
system under study. Figure 3 shows the angles associated with the Conclusion
voltage vector at various nodes of the system. Bus 1 is taken as the
reference node for the estimation. In the case of voltage angles DSSE using LSE is proposed in this paper to overcome the
also, the estimated values closely follow the true values. drawbacks of the standard WLS method. The proposed method
The errors in magnitude and angle of the bus voltage vector is suitable where true measurements are less in number which is
with reference to the true values are depicted in Figs. 4 and 5. compensated by pseudo measurements. The reported algorithm
Figures show that the errors are high in WLS method as com- can also accommodate synchronized phasor measurement data
pared to LSE method. to obtain more precise state estimates. The errors in the esti-
A detailed study of LSE incorporating DG is also carried mates with the proposed method are found to be less when
out. DG is placed in the system based on the PSI values. The compared to WLS technique. A methodology is evolved for

Fig. 5 Error in Voltage angles for 0.008

Error in Voltage angle(degree)

a 33 bus radial distribution system

using LSE method and WLS 0.007
method
0.006

0.005

0.004

0.003

0.002

0.001

0
1

11

13

15

17

19

21

23

25

27

29

31

33

Bus number
LSE WLS
Technol Econ Smart Grids Sustain Energy (2021) 6:3 Page 7 of 7 3

Fig. 6 Improvement in voltage

1.02
profile with DG placement for a

Voltage magnitude (p.u.)

IEEE 33 bus radial distribution 1
system 0.98
0.96
0.94
0.92
0.9
0.88
0.86
0.84
0.82
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33
Bus number
Without DG With DG

using LSE in DG interconnected distribution network and it is 13. Gelagaev R, Vermeyen P, Driesen J (Oct.2008) SE in distribution
grids. Int Conference on Harmonics and Quality of Power
tested and verified. Voltage profile improvement is observed by
14. Lin WM, Teng JH, Chen SJ (2001) A highly efficient algorithm in
placing the DG in the system. A comprehensive LSE algorithm treating current measurements for the branch-current-based DSSE.
is evolved, tested, verified and reported considering different IEEE Tr. on Power Delivery 16(3):433–439
aspects of a modern day electric distribution network. 15. Zhou M, Centeno VA, Thorp JS, Phadke AG (2006) An alternative
for including phasor measurements in SE. IEEE Tr. on Power Syst
21(4):1930–1937
16. Hu Y, Kuh A, Yang T, Kavcic A (2011) A belief propagation based
power distribution system state estimator. IEEE Computational
References Intelligence Magazine 6(3):36–46
17. Radhoush S, Shabaninia F, Lin J (Feb. 2018) DSSE with measure-
1. Primadianto A, Lu C-N (2017) A review On DSSE. IEEE Tr Power ment data using different compression methods. IEEE Power and
Syst 32(5):3875–3883 Energy Conference
2. Dehghanpour K, Wang Z, Wang J, Yuan Y, Bu F (2019) A survey 18. Huang YF, Werner S, Huang J, Kashyap N, Gupta V (2012) SE in
on SE techniques and challenges in smart distribution systems. electric power grids: Meeting new challenges presented by the re-
IEEE Tr.on Smart Grid 10(2):2312–2322 quirements of the future grid. IEEE Signal Proc Mag 29(5):33–43
3. Baran ME, Kelley AW (1994) SE for real time monitoring of dis- 19. Korres GN, Hatziargyriou ND, Katsikas PJ (2011) SE in multi-
tribution system. IEEE Tr Power Syst 9(3):1601–1609 microgrids. Eur Tr Electr Power 21(2):1178–1199
4. Lu CN, Teng JH, Liu WHE (1995) Distribution system state esti- 20. Huang J, Gupta V, Huang YF (2012) Electric grid SE for distribu-
mation. IEEE Tr Power Syst 10(1):229–240 tion systems with microgrids. In Proc Inf Sci Syst (CISS), 46th Ann
5. Baran ME, Kelley AW (1995) A BCSE method for distribution Conf:1–6
systems. IEEE Tr. Power Syst 10(1):483–491 21. Korres G, Manousakis N (2012) A SE algorithm for monitoring
6. Ghosh AK, Lubkeman DL, Downey MJ, Jones RH (1997) DSSE topology changes in distribution systems. In Proc IEEE Power
using probabilistic approach. IEEE Tr. In Power Syst 12(1):45–51 Energy Soc Gen Meeting:1–8
7. Roytelman I, Shahidehpour SM (1993) SE for electric power dis- 22. Wang H, Schulz N (Feb. 2004) A revised branch current-based
tribution systems in quasi real time conditions. IEEE Tr. On Power DSSE algorithm and meter placement impact. IEEE Tr. on Power
Delivery 8(4):2009–2015 Syst 19(1):207–213
8. Kong X, Chen Y, Xu T, Wang C, Yong C, Lil P, Yu L (2018) A
23. Gómez-Quiles C, Romero-Ramos E, de la Villa-Jaén A, Gómez-
hybrid SE based on SCADA and PMU measurements for medium
Expósito A (June 2020) Compensated load flow solutions for dis-
voltage distribution system. Journal of Applied Sciences
tribution system state estimation. Energies:1–13. https://doi.org/10.
9. Sandhya K, Chatterjee K, Ghose T (Dec. 14–16, 2018) SE for DG
3390/en13123100
penetrated adaptive distribution system during disaster.
24. Yao Y, Liu X, Zhao D, Li ZI (March 2019) DSSE-A semidefinite
Proceedings of the National Power Systems Conference (NPSC)
programming approach. IEEE Tr. on Smart Grid 10(4):4369–4378
10. Fathi M, Ghiasi M (June 2019) Optimal DG placement to find
optimal voltage profile considering minimum DG investment cost 25. Carcangiu S, Fanni A, Pegoraro PA, Sias G, Sulis S (February
in smart neighborhood. Smart Cities, MDPI Journal 2020) Forecasting-aided monitoring for the distribution system
11. Viswanatharao J, Praveena E, Srinivasarao R (2017) Placement and state estimation” ,WILEY:1–15. https://doi.org/10.1155/2020/
sizing of DG for distribution system by using a novel algorithm. 4281219
IEEE Conference ICICCT
12. Chatterjee S, Roy BK (2015) An analytic method for allocation of Publisher’s Note Springer Nature remains neutral with regard to jurisdic-
DG in radial distribution system. IEEE Con INDICON tional claims in published maps and institutional affiliations.