Hardware Implementation and analysis of a Seven Level MLI with SVPWM

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 04 Issue: 06 | June -2017 www.irjet.net p-ISSN: 2395-0072
© 2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 373
Hardware Implementation and analysis of a Seven Level MLI with
SVPWM
Aryasree G, Steffy Tresa Loui , Merin Sunny
Assistant professor ,Dept. of EEE ,Viswajyothi College of Engineering, M G University, Vazhakulam,Kerala ,India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Various modern industrial applications require
higher power rating equipment. Someofthemediumvoltage
motor drives and utility applications demand medium
voltage and megawatt power level. The Multilevel power
inverter has been developed to meet this demand more
effectively. The constraint in MLI is that of the increasing
number of switchesinvolved.ThoughSinusoidal PulseWidth
Modulation (SPWM) is most popularlyusedforinverters, the
Space Vector Pulse Width Modulation(SVPWM)isbecoming
popular in inverter systems with - better utilization of input
voltage, lesser harmonic content and better output voltage
profile, especially for a drive system. The complexity of
calculations involved though, pose a difficulty in SVPWM
especially with MLIs.
This paper proposes a simplified low computational Space
vector algorithm applied to a reduced switch MLI topology.
The simulation of the proposed model is done in
MATLAB/Simulink and the hardware prototype model at
low power was implemented with DSPACE.
Key Words: Reduced Switched Multilevel Inverter (MLI),
Low Computation Space Vector Pulse Width Modulation
(SVPWM) Algorithm
1. INTRODUCTION
High performance and easy & accurate controllability are a
demanding requirement for industrial drives in general.
Controlling of high power rated industrials drives, usually
induction machine drives deals with controlling of their
inverter topology. From various researches it has been
established that multilevel invertersaregainingprominence
in now a day due to its inbuilt capability of high voltage
handling and good harmonic rejection abilitieswithrecently
available power devices. While using a multilevel inverter
the high number of switches contributes to high switching
losses. But in recent years many efficient multilevel inverter
topologies with less number switches have been proposed
promising improvements in demanding industrial
requirements .This paper proposes one such reduced
switched multilevel topology [1-2].
Various PWM techniques have realized to control the MLI
topologies to provide a good modulation range, decreased
losses in switching, reducedTHD, also effectivenessindigital
implementation. Comparing with other pulse width
modulationmethodsSVPWMisthehighlyrecommendedone.
The typical SVPWM techniques involve tedious calculations
which increase the complexity of implementation especially
for higher level MLIs[3]. In this paper we realize a simplified
space vector modulationalgorithmavoidingthecomplexities
with determinationofsectordivision,derivationofswitching
states and its timing instants .
With higher number of levels, When employed in MLI the
hardware realization will become complex and leads to
increased memory consumption. Also the paper includes the
hardware implementation of the reduced switch MLI with
low computational SVPWM technique[4-5].
1.1 PROPOSED MLI TOPOLOGY
The proposed multilevel inverter to accomplish a sevenlevel
output with reduced switches is indicated in Fig.1.
SW1
SW2
SW3
SW4
SW5
D S
G1
D S
G2
G3
G4
G5
Vdc1
Vdc2
Vdc3
Vdc4
D
S S
D
S
S D
D
S
VaD
Fig.1. Circuit diagram of Proposed Multilevel Inverter
Topology
The maximum possible output voltage V0 is given by
V0 = (n-1)/Vdc. Where n represents the number of DC
voltage sources per phase.
TABLE I. SWITCHING STATES OF PROPOSED MLI
Voltage level Switches Turn ON
3Vdc Sw1& Sw5
2Vdc Sw2& Sw5
1Vdc Sw3& Sw5
0Vdc -
-1Vdc Sw1& Sw4
-2Vdc Sw2& Sw4
-3Vdc Sw3& Sw4
The proposed topology achieves the different levels as per
the switching sequence shown in the switching table. For
example to obtain output voltage of +Vdc, switches Sw3 and

Sw5 are turned ON. Similarly all the other switching levels
are realized. The switching losses and complexity in
multilevel inverters can be reduced by reducing the number
of switches.
1.2 SPACE VECTOR PULSE WIDTH MODULATION
Implementation of SVPWM strategy for controlling offers
less harmonic distortion in the output voltagesandcurrents.
It gives flexibility in optimizing switching pattern design. A
Low computational algorithm SVPWM for a multilevel
multiphase system is proposed here [6]. The following flow
chart summarizes the steps of the algorithm.
Fig. 2. Flow chart for the SVPWM implementation
1) Initially the normalized reference, Vr is obtained from
the reference voltage vector Vdc using the expression.
2) Then decompose the normalized reference vector into
the sum of its integer part, vi, and its fractional part, vf,
by means of the equations
3) The next step is to calculate the permutation matrix P
that sorts the vector vf in descending order in
accordance with
4) Now rearrange the rows of the triangular matrix D in
order to obtain the matrix D by means of
5) Further extract the displaced switchingvectors,vdj,from
the matrix D by taking into account the expression -
6) Now obtain the final switching vectors, vsj, byaddingthe
integer part of the reference, vi, to the displaced
switching vectors vdj according to equation.
7) Lastly the time corresponding to each switching vector
is calculated from the components of the vector vf by
means of the expression
As is evident, this is a generalized algorithm handling all the
switching states, providing a sorted switching vector
sequence which reduces the number of switching of the
switches. This is hence capable of supportinganymulti-level
multiphase inverter system, enhancing the capability of the
algorithm further. In this regard the algorithmisusedforthe
proposed MLI topology here.
1.3 SIMULATION RESULTS
The software modeling of the system is carried out using
MATLAB/SIMULINK. The Simulation parameters are shown
in TABLE II –
TABLE II .Simulation Parameters
Parameter Value
Cell Voltage 4.5V
Control signal frequency 50Hz
Switching frequency 2KHz
Load
R load = 100 Ω
L load = 100µH
C load = 220µF
The proposed model was simulatedforfeeding passiveloads
- a resistive load and a resistive-inductive-capacitive load
and the results are given below. Fig.3 shows the SVPWM
pulses generated by the low computationSVPWMalgorithm.
Normalization of
the reference
voltage
Decomposition into
integer and
fractional parts
Calculation of the
permutation matrix
Formation of
coefficient matrix
Extraction and
calculation of
switching vectors
Calculation of
switching time
Gate signal
generation

Fig.4 and Fig.5 shows the MLI output waveform with R load
and R-L-C load respectively.
0 0.02 0.04 0.06 0.08 0.1
-0.5
0
0.5
1
1.5
Time in Sec
Amplitude
Gate Pulse
Fig.3. SVPWM pulses
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1
-20
-10
0
10
20
Time in Sec
Vaph(V)
MLI Output- Phase Voltage
Fig.4. Output waveform with R load
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
-20
-10
0
10
20
Time in Sec
Vaph(V)
MLI Output- Phase Voltage
Fig.5. Output waveform with R-L-C load
The five level output is thus observed from the proposed
model simulation. The pulses of the various switches can be
observed to be different from eachotherduetothetracingof
the levels as per the proposed switching states. Applying for
an R-L-C load shows the approximately sinusoidal output
waveform that is obtained. Filtering the output can provide
near sinusoidal waveforms, promisingly better thanVoltage
Source Inverters with the voltage levels involved. Also this
proves the effectiveness of the MLI topology as well,
compared to others for the same output voltage levels,
attained with reduced number of switches though.
1.4 EXPERIMENTAL SETUP
In addition to the simulation, a low power prototype of the
proposed model was implemented. The proposed algorithm
has been implementedusingMATLABinterfacewithdSPACE
DS1104. The obvious advantage of using hardware-in-loop
simulations is that performance of systems for various
operating conditions can be compared. Another feature of
the dSPACE is the control desk which allows the graphical
user interface, through the controldesk theusercanobserve
the response of the system, give command to the system
through the interface [7-8]. The schematic of the
experimental is shown in Fig.6
Fig.6. Hardware implementation schematic diagram
The hardware implementation of the proposed system can
be divided into three parts:- Pulse generation circuit,
Isolation circuit, and Power circuit.
The Pulse generation circuit consist mainly the
MATLAB/Simulink model interfaced with dSPACE control
card. The system implemented in MATLAB/Simulink, the
model is converted to real time codes using a Real-Time
Interface (RTI) of the control by the dSPACE DS1104control
card. The gate pulses generated by the model can be
obtained at the Digital I/O pins of the dSPACE DS1104 I/O
interface.
The pulses obtained from the interface are given to an
Isolation circuit, employed to protect the pulse generation
circuitry from the power circuit. A 12 pulse optocoupler
isolator board is used as the isolation circuit. The isolated
pulse outputs are given to the proposed MLI power Circuit.
The Power circuit consists of a 3 phase reduced MLI
topology. Each phase has 5 switches with 2 of them bi-
directional. Each phase of the power circuit needs 4 DC
sources, which is provided by the Regulated DC Power
Supply. The load used in this hardware prototype model is a
resistive load and resistive-inductive-capacitive load.
Fig.7 and Fig.8 shows the experimental setup of the
proposed system.
MATLA
B/
simulin
k
Model
dSPACE
(DS1104)
Opto-
Coupler
Board
MLI
Circu
it

Fig.7. Complete hardware setup
Fig.8. Hardware setup – Power circuit
Fig.9 shows the generated SVPWM pulses obtained after the
isolation circuit, which are given to the switches. Fig.10 and
Fig.11 presents the hardware output waveform with R load
and R-L-C load respectively.
Fig.9. Hardware SVPWM pulses
Fig.10.Hardware output waveform with Resistive load
Fig.11. Hardware output waveform with Resistive-
Capacitive load
The Hardware results can be observed thus to verify and
validate the Simulation results. The output waveforms of
both R and R-L-C loads are observed to be almost same in
the simulation and hardware setup. The proposed system
was thus successfully implemented as a prototype and its
effectiveness was clearly observed.
2. CONCLUSIONS
The paper implemented satisfactorily a simplified low
computational SVPWM algorithm for a reduced switch MLI
topology. The observed results from both the simulation and
the hardware implementation confirmed the successful
Regulated
DC power
supply Power circuit
dSPACE
DC power
supply
Isolation
circuit
Power circuit

workingofthe proposedsystemofboththeSVPWMalgorithm
as well as the reduced switch MLI topology.
This proves the effectiveness of the simplified SVPWM
algorithm generallya dauntingissuefordigitalprocessorsand
also validates the improvement of the MLI topology. The
results of the implemented prototype system hence promise
improvedand efficient performanceinthehighpowersystem
as well. This is especially useful for a drive system for which
the quality of source voltage and performance is of major
concern.
REFERENCES
[1] Drazen Dujic, Martin Jones, Emil Levi, “A space
vector PWM scheme for multi frequency output voltage
generation with multiphase voltage-source inverter”,
IEEE Trans. Electronics, Vol.55 no.5. may2008
[2] C.Bharatiraja, Dr S. Jeevananthan, R. Latha, Dr S. S.
Dash, “A Space vector pulse width modulationapproach
for DC link voltage balancing in diode clamped
multilevel inverter”, Science direct AASRI Procedia 3-
2012,133-140
[3] Jesmin. F. Khan, Sharif M. A Bhuiyan, Kazi. M.
Rahman, Gregory. V. Murphy, “Space vector PWM for a
two phase VSI”, Science direct, Electrical power and
energy system, 51 (2013) 265-277.
[4] Shieheng Zheng, Taotao Cui, Mu-Tian Cheng, Long
Chen, “Research on variable frequency with three level
topology and SVPWM control”, IEEE conference paper
June 2012,978-1-4577-2088-8/11
[5] Thomas Nathenas, Heorgios Adamidis, “A new
approach for SVPWM of a three- level inverterinduction
motor fed- neutral point balancing algorithm”, Science
direct , 2012
[6] Zixin Li, Ping Wang, Haibin Zhu, Yaohua Li, “A New
Simplified and space vector PWM Method for N-level
Inverters”, IEEE Trans. Electronics, 2008, 978-1-4244-
1718-6/08.
[7] Renukadevi. G; Rajambal. K., "Field programmable
gate array implementation of space-vector pulse-width
modulation technique for five-phase voltage source
inverter," Power Electronics, IET , vol.7, no.2,
pp.376,389, February 2014
[8] P. M. Meshram, Dipesh Hanote, M. M. Renge,” A
Simplified space vector PWM for three level inverter
applied to passive and motor load”, ICIEA 2009

Hardware Implementation and analysis of a Seven Level MLI with SVPWM

More Related Content

What's hot (18)

Similar to Hardware Implementation and analysis of a Seven Level MLI with SVPWM (20)

More from IRJET Journal (20)

Recently uploaded (20)

Hardware Implementation and analysis of a Seven Level MLI with SVPWM