EVALUATION OF POWER QUALITY

IN A MICROGRID USING DATA-

DRIVEN TECHNIQUE

Mojisola Adunola JIMOH

(B.Sc. (O.A.U), M.Sc. (O.A.U), MNSE,
R.COREN)
 INTRODUCTION
• Power quality (PQ) refers to the possibility of maintaining a perfect sine
waveform of bus voltages and or bus currents of a distribution system
at their nominal magnitude and frequency at a particular point in time
(Wu et al. ,2019).
• Events or disturbances: significant variations in RMS voltage
magnitudes for varying periods. Event types are voltage sags and
swells, undervoltages, overvoltages, and interruptions (IEEE Std. 1159-
2009).
• Steady-state distortions: steady-state deviations from an ideal sine
waveform of power frequency. These are harmonics, interharmonics,
oscillatory and impulsive transients etc. (IEEE Std. 1159-2009).
 STATEMENT OF RESEARCH PROBLEM

The growing adoption of microgrids in

distribution systems driven by renewable
energy integration has enhanced reliability
and sustainability of the power industry. It
however introduced new concerns for PQ
due to the intermittent nature of their
sources, the integration of power electronic
devices, and changes in the characteristics
of loads in distribution networks.
 AIM AND OBJECTIVES

The research aim to characterise the power system’s

performance by assessing the microgrid’s PQ parameters
and verifying compliance with the IEEE standards and
specifications. The specific objectives of the study are to:
i. implement a PQ monitoring system on a hybrid DC-AC
microgrid testbed;
ii. analyse the testbed parameters for PQ events and
distortions; and
iii.estimate the PQ indices of recorded PQ problems
 CAUSES OF PQ PROBLEMS
PQ disturbances are majorly caused by systems’ operation
such as energizing and de-energizing of heavy loads, loose
connections, and faults including single line-to-ground
(SLG) and line-to-line symmetrical faults (Kachhepati,
2016; IEEE Std. 1159-2009).
Waveform distortions result from the nonlinearity of end-
users’ equipment and loads such as semiconductor-based
loads and switching power supplies (IEEE Std. 1159-2009)
 EFFECTS OF PQ PROBLEMS
RMS variations: Voltage instability, malfunctioning of
electronic loads, insulation breakdown, and
tripping of protection circuits, equipment
shutdown/failure, disruption of critical loads
(Kachhepati, 2016; Arias-Guzmán, 2017)

Waveform distortions: tripping of circuit breakers,

increased thermal stress, degraded
performance of rotating motors, higher
copper losses from cable overheating,
reduced efficiency of end-user equipment
(Kachhepati, 2016; Arias-Guzmán, 2017).
 PQ INDICES AND STANDARDS: THD
PQ indices are quantitative measures of the extent of
disruptiveness of disturbances and distortions (Ivry et al.,
2016).
Total harmonic distortion (THD): ratio of the square root of
RMS value of harmonic content up to 50th order as a
√
percentage ∑ ( 𝑉of ) the fundamental content, excluding
𝑁
h 2
𝑟𝑚𝑠
h =2
𝑇𝐻𝐷 = 𝑋 100 % (1 )
interharmonics
𝑉
1
(Adesina
𝑟𝑚𝑠 and Fakolujo, 2015).

Allowable harmonic limit of 8% VTHD is allowable for bus

voltage V ≤ 1.0 kV (IEEE Std 519, 2014).
 PQ INDICES AND STANDARDS: SAG SEVERITY

Sags are also characterized by the depth of the event, the

difference between its retained voltage and the nominal
voltage.
This involves calculating the retained voltage or depth of
individual event and1 duration
V of occurrence to a reference
Se 
curve (Arias-Guzmán 1  Vet
curveal.,
(d ) 2017).

The IEEE recommends a reference curve standard. An event on

the curve has a unit sag severity; an event above the curve is
less than one and is considered within the equipment’s
 PQ ANALYSIS

i. Characterization techniques: Fourier transforms

and time-frequency representations (TFRs) such
as short-time Fourier transform (STFT),
Stockwell transforms, wavelet transforms etc.
(Găşpăresc, 2011; Kostov et al., 2015).

ii. Classification techniques: artificial neural

networks, expert systems, support vector
 M E T H O D O LO GY

Figure 1: Block diagram of the microgrid

testbed
 M E T H O D O LO GY

Figure 2: Block diagram of research

process
 F E AT U R E E X T R A C T I O N A L G O R I T H M

Figure 3: Flow chart of STFT Algorithm

 FEATURES FOR CLASSIFICATION
 F1 is the standard deviation of the TmA vector
 F2 is the maximum value of the TmA vector
 F3 is the number of peaks in the TmA vector plus the number of local
minima in the TmA plot.
 F4 is the minimum value of the TmA vector
 F5 is the standard deviation of the FmA vector corresponding to
frequencies within the 3rd and 7th harmonics.
 F6 is the standard deviation of the FmA vector corresponding to
frequencies above the 3rd harmonics.
The dataset has 1200 set of the features under 8 PQ problems (ideal
signals, voltage sags, voltage swells, interruptions, oscillatory transients,
harmonics, sags + harmonics, and swells + harmonics (Igual et al., 2020).
 C L A S S I F I C AT I O N A L G O R I T H M

Figure 4: k-NN Algorithm

 TFR OF DETECTED PQ PROBLEMS

(a)

(b) (c)

Figure 5: TFRs on SMPS (a) Spectrogram (b) TmA Plot (c)

FmA plot
 TFR OF DETECTED PQ PROBLEMS

(a)

(b) (c)

Figure 6: TFRs on energizing 160 W resistive load (a) Spectrogram

(b) TmA Plot (c) FmA plot
 C L A SS I F I C AT I O N O F D E T E C T E D P Q P R O B L E M S

Table 1: Classifier output of detected PQ problem

Featur Unloaded On SMPS load Energizing of 160 W load
es
Phase Phase Phase C Phase Phase B Phase C Phase A Phase B Phase C
A B A

F1 0.0051 0.0116 0.00417 0.0174 0.04006 0.03050 0.05822 0.04471 0.06631

3 8 6

F2 1.0034 0.9980 1.02518 1.0068 0.95139 0.97298 0.99220 0.97599 1.04128

3 2 9

F3 6.0000 3.0000 7.00000 5.0000 6.00000 5.00000 1.00000 2.00000 1.00000

0 0 0

F4 0.9857 0.9576 1.00543 0.9441 0.76928 0.84529 0.80282 0.85557 0.83706

8 8 6
C L A S S I F I C AT I O N M O D E L E VA LU AT I O N

Figure 7 : Confusion Matrix of the k-NN

classifier
 PQ INDICES

Table 2: PQ indices of recorded disturbances

THD % Sag Severity
Unloaded Phase A 1.473 N/A
Phase B 1.481 N/A
Phase C 1.425 N/A
Energizing 160 Phase A 1.489 0.99
W resistive Phase B 1.556 0.72
loads
Phase C 1.674 0.95

On SMPSs Phase A 1.667 N/A

Phase B 1.782 N/A
Phase C 2.043 N/A
 C O N C LU S I O N

Performance of the overall system, tolerance to

various load conditions, safety of equipment load of
the system were established.

The findings of these study further established

microgrids are susceptible to harmonics, and voltage
sags.

The findings further showed that combination of

 R E C O M M E N D AT I O N S

PQ monitoring of microgrids should be in real-time as

the desire to have a safe supply is as important as
having the supply.
The use of a hybrid harmonic mitigation and
compensation strategy for PQ using a combination of
active and passive filters in the DC-AC converter to
further limit harmonic distortions in the supply.
Automated load shedding can help reduce demand,
mitigating against voltage sags and sustained
interruption.
THANK YOU FOR
LISTENING !!!
 REFERENCES

Adesina, L. M., and Fakolujo, O. A. (2015). Determination of Total Harmonic Distortion

(THD) on a 33kV Distribution Network: A Case Study of Island Business
District. Nigerian Journal of Technology, 34(4), 773-780.
https://doi.org/10.4314/njt.v34i4.15
Arias-Guzmán, S., Ruiz-Guzmán, O. A., Garcia-Arías, L. F., Jaramillo-Gonzáles, M.,
Cardona-Orozco, P. D., Ustariz-Farfán, A. J., Cano-Plata, E. A., and Salazar-Jiménez, A. F.
(2017). Analysis of Voltage Sag Severity Case Study in an Industrial Circuit. IEEE
Transactions on Industry Applications, 53(1), 15–21.
https://doi.org/10.1109/TIA.2016.2603470
Găşpăresc, G. (2011). Methodes of Power Quality Analysis. In Power Quality
Monitoring, Analysis and Enhancement (pp. 101-118). IntechOpen.
https://doi.org/10.5772/18370
Igual, R., Miraftabzadeh, S. M., Foiadelli, F., and Medrano, C. (2020, July).
Quantification of Feature Importance in Automatic Classification of Power Quality
Distortions. In 2020 19th International Conference on Harmonics and Quality of Power
(ICHQP) (pp. 1-6). IEEE.
Institute of Electrical and Electronics Engineers. (2009). IEEE Std 1159-IEEE
Recommended Practice for Monitoring Electric Power Quality. In IEEE Std 1159-2009
 REFERENCES

Ivry, P. M., Thomas, D. W., and Sumner, M. (2016, May). Assessment of Power
Quality in a Microgrid with Power Electronic Converters. In 2016 Asia-Pacific
International Symposium on Electromagnetic Compatibility (APEMC) (pp. 825-
827). IEEE.
Kostov, M., Gegov, B., Atanasovski, M., and Petkovski, M. (2015, June). Short
Time Fourier Transform for Power Disturbances Analysis. In Proceedings of the
International Scientific Conference on Information, Communication and
Energy Systems and Technologies, Sofia, Bulgaria (pp. 24-26).
Kachhepati B. (2016). Application of Short Time Fourier Transform (STFT) in
Power Quality Monitoring and Event Classification, Thesis (M.Sc.)-New Mexico
State University.
Singh, U. (2020). A Research Review on Detection and Classification of Power
Quality Disturbances caused by Integration of Renewable Energy Sources.
http://arxiv.org/abs/2009.11426
Wu, M., Niu, X., Gao, S., and Wu, J. (2019). Power quality assessment for
AC/DC hybrid micro grid based on on-site measurements. Energy Procedia,