Iraq J.

Electrical and Electronic Engineering ‫اﻟﻤﺠﻠﺔ اﻟﻌﺮاﻗﻴﺔ ﻟﻠﻬﻨﺪﺳﺔ اﻟﻜﻬﺮﺑﺎﺋﻴﺔ واﻻﻟﻜﺘﺮوﻧﻴﺔ‬

Vol.13 No.1 , 2017 2017 ، 1 ‫ اﻟﻌﺪد‬، 13 ‫ﻡﺠﻠﺪ‬

Mathematical Driving Model of Three Phase, Two Level

Inverter by (Method of Interconnected Subsystem)

Mohammed .H. Ali

Electrical power and machines Engineering Department
College of engineering
Diyala University
Iraq – diyala-baquba
Mobile : +96407717218264
Email: [email protected]

Abstract In this paper describe to mathematical analysis for a three-phase, two level inverter designs. As we know the
power electronic devices (inverter) to convert the DC power to AC power (controller on output voltage and
frequency level). In Industrial applications, the inverters are used for adjustable speed (AC Drives). In this paper,
the mathematical analyses for inverter design are done by using Software packages C++ Builder and visual C++
Language. For non- linear distortions described by the load power factor in power system networks. The P.F is
reverse proportional with the harmonics distortion. Small P.F means much more of harmonic distortion, and lower
power quality for consumers. to improve the P.F, and power quality in this paper the small capacitor installed as
part of the rectified the load current has power (30 KW with P.F load 0.8), the fluctuations of the rectified voltage
must not greater than +/- 10%.The power factor proportion of the load power, with Modulation coefficient p.u
approximately unity. The calculation is achieved with different integrations steps with load power 30KW, 0.8 P.F. all
results done Based on model and experimental data..

Keywords:- Mathematical analysis, Modeling, three phase - two level inverter, Interconnected Subsystem.

I. INTRODUCTION
This Paper describes a model of PWM inverter
fed three-phase load. The model needs to be
based up by decomposition of a system into sub
circuits that are coupled by means of dependent
voltage/current sources. Such an approach
ensures high flexibility in construction of system
models along with acceptable accuracy of
computation based on model and experimental
data, this model described is built up on Figure 1 Scheme of the system with two-level
decomposition of complex system into sub inverter and load [3].
circuits interconnected via dependent
voltage/current sources [1]-[2].To highlight this This system is decomposed into SC sub circuit
method of computer model construction we shall and three-phase load block. SC is fed by current
consider the simplest system with two-level from a DC source with resistor and inductance in
converter and three phase loads (30, 100) KW a circuit of rectified voltage there is a capacitor C
show on in fig (1)..Computer models of the with current. Each leg of SC consists of a
system with load power of 30 KW and transistor along with its anti parallel diode.
semiconductor converters (SC) are widely used to Transistors and diodes are supposed to be ideal
facilitate development. gates. The static energy losses are taken into
account by resistor [4]. States of semiconductor

73
 Iraq J. Electrical and Electronic Engineering ‫اﻟﻤﺠﻠﺔ اﻟﻌﺮاﻗﻴﺔ ﻟﻠﻬﻨﺪﺳﺔ اﻟﻜﻬﺮﺑﺎﺋﻴﺔ واﻻﻟﻜﺘﺮوﻧﻴﺔ‬
Vol.13 No.1 , 2017 2017 ، 1 ‫ اﻟﻌﺪد‬، 13 ‫ﻡﺠﻠﺪ‬

elements are described by a discrete function kin,

n=1, 2, 3: if an open transistor or diode connects e0 = ( e1 + e 2 + e 3 ) / 3 (2)
n-the phase to a positive pole of the capacitor C
then kin=1, and if it connects n-the phase to a
negative pole then kin = 0. The transistors in an un = en – e0 , n = 1,2,3
arm are complementary one to the other: if one
transistor in an arm is fired, then the other is Voltage of capacitor C:
turned off. The arms of the bridge attached to the 1
c
uc  ic dt
positive pole transfer input current of the inverter
circuit, this source further.[5]-[6] (3)
Output phase voltage of ideal converter without
II. MATHEMATICAL ANALYSIS FOR THREE zero-sequence component:
PHASE TWO LEVEL INVERTER [7]. k i1  k i 2  k i 3
en  k in  u rc , u n  en (4)
3
The dividing mathematical analysis system into
sub circuit, interconnected-dependent voltage , DC voltage source current id is determined from
and current sources. As a result represented sub an equation:
circuit shown in Fig 2.
Derivative current supply:-
d in u n  rn i n
 (5)
dt lH

Currents of bridge arms:

iin  k in i n iin3  (k in  1)i n , n  1,2,3 (6)

Transistor currents itn and diodes currents idn:

Figure 2 Dividing sub scheme of the system with If
two-level inverter and load.

Equivalent EMF Phase Inverter: when n  1,2,....., 6

= (1) Inverter input current:
Further transformation schemes shown in Fig.3. i dn  ii1  ii 2  ii 3 (7)

DC voltage source current id is determined from

an equation:
d ik u k  u rc  rd i k
 (8)
dt ld

Where current of the protection circuit:

u
i z  k z rc (9)
rz
Figure 3 Transformation schemes system with The arms of bridge currents:
two-level inverter and load. i z  k in i n , iin3  (1  k in )i n (10)
Current of capacitor:
Removal from the EMF phases of the zero
ic  i d  i z  i di (11)
sequence components.

74
 Iraq J. Electrical and Electronic Engineering ‫اﻟﻤﺠﻠﺔ اﻟﻌﺮاﻗﻴﺔ ﻟﻠﻬﻨﺪﺳﺔ اﻟﻜﻬﺮﺑﺎﺋﻴﺔ واﻻﻟﻜﺘﺮوﻧﻴﺔ‬
Vol.13 No.1 , 2017 2017 ، 1 ‫ اﻟﻌﺪد‬، 13 ‫ﻡﺠﻠﺪ‬

III. MODELING OF CONTROL SYSTEM INVERTER A and B intermediate variables, constant time
aperiodic filter.
In modeling, circuit Fig.1. saw tooth voltage is
described by the equation (12). PI control at the load current:
I  I Z  I , U ym  U yi  I .K I 0 (20)
Ton  Ton  f on .t (12)
If
U y min  U yi  U YMAX y max , then

U yi  U yi  I .K I 0
Where frequency of the reference voltage
If
Hz , intermediate variable
, The second calculation. U ym  y max , then U ym  y max

Voltage control is determined by the following If

formulas: U ym  y min , then U ym  y mix
t  t  t y (13)
Where
u y1  u y max sin( H t ) (14)
ΔI –Deviation of the actual current from
2 reference sub phase.
u y1  u y max sin( H t  ) (15)
3
Δt – Step work system control.
4
u y1  u y max sin( H t  ) (16) Uym– Voltage control amplitude.
3
Uyi– Integral component of voltage control.
Where t, time in second , angular frequency
Uymin– Minimum value control voltage.
voltage reference load by rad/sec, Uy max the
Uymax– Maximum value control voltage.
amplitude of maximum voltage control.
KIi– Coefficient integration for the deviation
IV. DEFINITION THE CURRENT LOAD CURRENT
BY INSTANTANEOUS VALUES IN PHASES current.
KIo– Coefficient of current deviation.
In one of the possible construction of a control,
system used PI controller acting load current. The IZ –The arms of bridge currents.
actual operating current of three-phase load is
determined in the process of calculating the
instantaneous variables: Simulation result saw tooth voltage described
Instantaneous value of voltage control VSI When
i12  i 22  i32 sinusoidal PWM represented in equations (21,22,
A (17) 23 and 24).
3
t y
B  B  ( A  B) (18)      * t (21)
Ti
u y1m  U ym sin(  ) (22)
I  B (19) 2
u y 2 m  U ym sin(  ) (23)
3

75
Iraq J. Electrical and Electronic Engineering ‫اﻟﻤﺠﻠﺔ اﻟﻌﺮاﻗﻴﺔ ﻟﻠﻬﻨﺪﺳﺔ اﻟﻜﻬﺮﺑﺎﺋﻴﺔ واﻻﻟﻜﺘﺮوﻧﻴﺔ‬
Vol.13 No.1 , 2017 2017 ، 1 ‫ اﻟﻌﺪد‬، 13 ‫ﻡﺠﻠﺪ‬

4 models of electrical drives with semiconductor

u y 3m  U ym sin(  ) (24)
3 converter and load by using C++ builder
Where t, time in second , reference value programmer. The calculations (for given load
frequency angular by rad/sec, Uy max, power 100KW and power factor load 0.5 to 0.8).
The amplitude maximum voltage Pu.
Table (1) Definitive input data

Sinusoidal PWM with zero sequence represented Emf power supply Ei 1000 V.
The inductance of power Li 0.0005 H
in equations (25,26 and 27). supply
Active resistance of the Ri 0.01 Ω
power supply
u y1m  U ym sin( )  0.13 * U ym sin(3 ) (25) Capacity of the capacitor C 0.002 F
Resistance of the capacitor Rc 0.01 Ω
2 battery
u y 2 m  U ym sin(  )  0..13 * U ym sin(3 ) (26) The resistance of protective Rz 1000 Ω
3
resistor
4 Inductive load Ln 0.022 H
u y 3m  U ym sin(  )  0.13 * u ym sin(3 ) (27) Resistance load Rn 9.24 Ω
3
The amplitude of emf load Enm 0V
The angular frequency emf omega 314.15 rad/s
V. RESULTS OF SIMULATION load
input data for control system
Frequency of the reference fop 2000 Hz
value
The Modeling system by interconnected sub Frequency rated of load f1 50 Hz
circuit to calculate transient and steady state voltage
models of VSI. For reference load power of 100 Maximum voltage control Uymx 1.8 Pu
KW and power factor 0.5 to 0.8 to hold series Maximum voltage across the Ucmx 1500 Vc
calculation. capacitor
The specified operating load Inz 32.89 А
Voltage phase calculation current
The coefficient of the integral Kii 0.25 Pu
of the current load
The coefficient of the load Kio 0.025 Pu
2
v  U m  0.38 * 1000  380v (28) current
3
Load current calculation Table (2) represented where P.F. to change (0.5 to
PL 30000 0.8) Calculation result for I, R and L by used
I    175.438A
3U * cos 380 * 0.5 * 3 (29) formula (28,29,30,31,32 and 33).

Table (2) result for I, R and L( 0.5 to 0.8)

U 380
Z   2.1660
I 175.438 (30)
0.5 0.6 0.7 0.8
175.438 146.1988 125.31 109.6491
R  Z cos  2.166 * 0.5  1.080 (31) 1.08 1.559 2.1227 3.2772
0.00597 0.0066 0.0068 0.0066
X  Z sin   0.0189 (32)
X 0.0189
L   0.061mH
 314.15 (33) VII. ALGORITHM CALCULATIONS PROGRAMS

VI. MODULATION SYSTEM CONTROL AND Figure 4 shows the flowchart programming for
CALCULATION TRANSIENT REGION calculation and solution equation by using C++
Input data for the program represented a table and visual C++ [8]- [9].
(1).The development complex of mathematical

76
 Iraq J. Electrical and Electronic Engineering ‫اﻟﻤﺠﻠﺔ اﻟﻌﺮاﻗﻴﺔ ﻟﻠﻬﻨﺪﺳﺔ اﻟﻜﻬﺮﺑﺎﺋﻴﺔ واﻻﻟﻜﺘﺮوﻧﻴﺔ‬
Vol.13 No.1 , 2017 2017 ، 1 ‫ اﻟﻌﺪد‬، 13 ‫ﻡﺠﻠﺪ‬

Figure 5 Schemes of characteristic current ( ,

, and )

Fig. 4 Algorithm calculations programs

The results are shown in Fig.(5,6 and 7) and the
table (3) when P.F.=0.5
Table (3) Harmonic analysis (

Source current: 29.918 Figure 6 Characteristic current and voltage of SC

The rectified voltage: 999.700 ( , , , and )
Current in the first switch
The current value of the curve: 37.169
The maximum value of the curve: 75.266
The minimum value of the curve: -75.273

Voltage Inverter 1 phase

The current value of the curve: 445.914
Harmonic Act. Of Phase (grad)
freq.(Hz) harmonic Value
50 379.981 -94.4640
Harmonics coefficient: 0.5233
Current 1 phase load
The current value of the curve: 52.556
Harmonic Act. Of Phase (grad)
freq.(Hz) harmonic Value
Figure 7 Characteristic of voltage PWM reference
50 52.550 -154.5692 at = 2000 Hz (carrier frequency).
Harmonics coefficient: 0.01486
Voltage control The results shows in Fig.(8, 9and 10) and the
Acting value of the curve: 0.797 table (4) when P.F.= 0.6.
Harmonic Act. Of Phase (grad)
freq.(Hz) harmonic Value
50 0.796 -94.4071
Harmonics coefficient: 0.04566

77
Iraq J. Electrical and Electronic Engineering ‫اﻟﻤﺠﻠﺔ اﻟﻌﺮاﻗﻴﺔ ﻟﻠﻬﻨﺪﺳﺔ اﻟﻜﻬﺮﺑﺎﺋﻴﺔ واﻻﻟﻜﺘﺮوﻧﻴﺔ‬
Vol.13 No.1 , 2017 2017 ، 1 ‫ اﻟﻌﺪد‬، 13 ‫ﻡﺠﻠﺪ‬

Table (4) Harmonic analysis (

Source current: 29.899

The rectified voltage: 999.702
Current in the first switch
The current value of the curve: 30.971
The maximum value of the curve: 62.698
The minimum value of the curve: -62.643
Voltage Inverter 1 phase
The current value of the curve: 444.636
Harmonic Act. Of Phase (grad)
freq.(Hz) harmonic Value Figure10 Characteristic of voltage PWM
50 378.981 -94.4701
reference at = 2000 Hz (carrier frequency)
Harmonics coefficient: 0.5248
Current 1 phase load
The current value of the curve: 43.781 The results are shown in Fig.(11,12 and 13) and
Harmonic Act. Of Phase (grad)
the table (5) when P.F.= 0.7
freq.(Hz) harmonic Value
50 43.775 -147.5263 Table (5) Harmonic analysis (
Harmonics coefficient: 0.01604
1 control voltage Source current: 29.864
Acting value of the curve: 0.787 The rectified voltage: 999.703
Harmonic Act. Of Phase (grad) Current in the first switch
freq.(Hz) harmonic Value The current value of the curve: 26.579
50 0.787 -94.4039 The maximum value of the curve: 53.810
Harmonics coefficient: 0.04563 The minimum value of the curve: -53.873

Voltage Inverter 1 phase

The current value of the curve: 445.192
Harmonic Act. Of Phase (grad)
freq.(Hz) harmonic Value
50 379.244 -94.3962
Harmonics coefficient: 0.5273
Current 1 phase load
The current value of the curve: 37.561
Harmonic Act. Of Phase (grad)
freq.(Hz) harmonic Value
50 37.561 -140.0594
Harmonics coefficient: 0.01792
Figure 8 Schemes and characteristic current ( , 1 control voltage
Acting value of the curve: 0.791
, and )
Harmonic Act. Of Phase (grad)
freq.(Hz) harmonic Value
50 0.790 -94.3779
Harmonics coefficient: 0.04567

Figure 9 Characteristic current and voltage of SC

( , , , and )
78
 Iraq J. Electrical and Electronic Engineering ‫اﻟﻤﺠﻠﺔ اﻟﻌﺮاﻗﻴﺔ ﻟﻠﻬﻨﺪﺳﺔ اﻟﻜﻬﺮﺑﺎﺋﻴﺔ واﻻﻟﻜﺘﺮوﻧﻴﺔ‬
Vol.13 No.1 , 2017 2017 ، 1 ‫ اﻟﻌﺪد‬، 13 ‫ﻡﺠﻠﺪ‬

Table (6) Harmonic analysis (

Source current: 29.880

The rectified voltage: 999.701
Current in the first switch
The current value of the curve: 23.243
The maximum value of the curve: 47.457
The minimum value of the curve: -47.056
Voltage Inverter 1 phase
The current value of the curve: 444.757
Harmonic Act. Of Phase (grad)
freq.(Hz) harmonic Value
50 378.632 -94.4475
Harmonics coefficient: 0.5246
Current 1 phase load
Figure 11 Schemes and characteristic current ( , The current value of the curve: 32.843
, and ) Harmonic Act. Of Phase (grad)
freq.(Hz) harmonic Value
50 32.836 -131.2152
Harmonics coefficient: 0.02118
1 control voltage
Acting value of the curve: 0.789
Harmonic Act. Of Phase (grad)
freq.(Hz) harmonic Value
50 0.788 -94.3920
Harmonics coefficient: 0.04575

Figure 12 Characteristic current and voltage of

SC ( , , , and ).

Figure 14 Schemes and characteristic current ( ,

, and

Figure 13 Characteristic of voltage PWM .

Reference at = 2000 Hz (carrier frequency
The results shows in Fig.(14, 15 and 16) and the
table (6) when P.F.= 0.8
Figure 15 Characteristic current and voltage of
SC ( , , , and )

79
 Iraq J. Electrical and Electronic Engineering ‫اﻟﻤﺠﻠﺔ اﻟﻌﺮاﻗﻴﺔ ﻟﻠﻬﻨﺪﺳﺔ اﻟﻜﻬﺮﺑﺎﺋﻴﺔ واﻻﻟﻜﺘﺮوﻧﻴﺔ‬
Vol.13 No.1 , 2017 2017 ، 1 ‫ اﻟﻌﺪد‬، 13 ‫ﻡﺠﻠﺪ‬

When input data for the program represented

in Table (8). The development of complex
mathematical models of electrical drives with
semiconductor converter and load by using C++
builder programmer. The series of calculations
(for given load power 30KW and power factor
load 0.8, exchange capacitor.

Table (8) Definitive input data.

Figure16 Characteristic of voltage PWM Emf power supply Ei 1000 V.
reference at = 2000 Hz (carrier frequency). The inductance of power Li 0.0005 H
supply
Active resistance of the Ri 0.01 Ω
VIII. MODULATION AND CALCULATION power supply
3Ø – 2 LEVEL INVERTOR . Capacity of the capacitor C 0.002 F
Resistance of the capacitor Rc 0.01 Ω
battery
Installation the small value of capacitance in The resistance of protective Rz 1000 Ω
part of the rectified current at a given load power resistor
of 30 KW with P.F. 0.8. Table (7) represented Inductive load Ln 0.022 H
Resistance load Rn 9.24 Ω
parameter of the load: The amplitude of emf load Enm 0V
The angular frequency emf omega 314.15 rad/s
Table (7) given load power of 30 KW with P.F. 0.8. load
input data for control system
Po, KW Frequency of the reference fop 2000 Hz
,Volt
value
30 0.8 Frequency rated of load f1 50 Hz
voltage, Hz
We calculated parameters by the following equas. Maximum voltage control Uymx 1.8 Pu
Maximum voltage across the Ucmx 1500 Vc
2 capacitor
V  U m  0.38 * 1000  380V The specified operating load Inz 32.89 А
3 current
The coefficient of the integral Kii 0.25 Pu
of the current load
The coefficient of the load Kio 0.025 Pu
current

The results are shown in Fig.(17, 18 and 19) and

the table (9) when C= .

80
Iraq J. Electrical and Electronic Engineering ‫اﻟﻤﺠﻠﺔ اﻟﻌﺮاﻗﻴﺔ ﻟﻠﻬﻨﺪﺳﺔ اﻟﻜﻬﺮﺑﺎﺋﻴﺔ واﻻﻟﻜﺘﺮوﻧﻴﺔ‬
Vol.13 No.1 , 2017 2017 ، 1 ‫ اﻟﻌﺪد‬، 13 ‫ﻡﺠﻠﺪ‬

Table (9) Harmonic analysis.

current supply 30.327

The voltage of the capacitor battery
The average value of the curve: 999.817
Maximum value of the curve: 1006.159
Maximum value of the curve: 992.737
current capacitor bank 17.56
Un1 load voltage Figure 19 The rectified voltage when С=0.025
Harmonic Act. Of Phase (grad) .
freq.(Hz) harmonic Value Further calculation: we used another values for
50 444.785 -94.4570
In1 load current capacitor changing from 2 to 0.035 .
The current value of the curve: 32.813
Table 10 represented results of calculation
Harmonic Act. Of Phase (grad)
freq.(Hz) harmonic Value Current of
50 32.813 -131.2492 The Voltage of the
Ripple of the capacitor
capacity of capacitor bank
rectified bank (rms
Power Supply: = 30.327*999.824 = the (maximum value of
capacitor voltage %
value) V the curve)
31268,49W bank C,
,A
Load Power: = 3*378.356*32.804*0.8 = 2 1001.210 0,2 15.717
1 1002.110 0,4 15.763
29787.816 W 0.1 1022.744 4,4 16.825
0.05 1057.155 11,4 18.528
Load Power Ref.: Рref = 30000 W 0.025 1092.040 18,4 20.567
0.035 1157.938 23,14 25.257
= -94.4950 – (-131.2492) = 36,7542
Cos ( ) = 0.801

Fig. 20. Represented the relation between ripple

Figure 17 Capacitor current and load voltage of the rectified voltage and capacity of the
capacitor bank

IX. CONCLUSION

In this paper, the Mathematical Driving Model

of Three Phase, Two Level Inverter designs is
done. The degree of linear load is described.
Power factor is the proportion of power at the
first harmonic of the current of the total power
Figure 18 The rectified voltage when С=2

81
Iraq J. Electrical and Electronic Engineering ‫اﻟﻤﺠﻠﺔ اﻟﻌﺮاﻗﻴﺔ ﻟﻠﻬﻨﺪﺳﺔ اﻟﻜﻬﺮﺑﺎﺋﻴﺔ واﻻﻟﻜﺘﺮوﻧﻴﺔ‬
Vol.13 No.1 , 2017 2017 ، 1 ‫ اﻟﻌﺪد‬، 13 ‫ﻡﺠﻠﺪ‬

consumed by the load. For each nonlinear (моделирование, расчет, применение)." ЕА

distortion have P.F and are introduced. In this Крутякова СПб, Силовыемашины
paper The fluctuations of the rectified voltage Электросила, Russia , p 252 ,2004.
does not greater than +/- 10% with installation of
[8] Г. Корн, Т. Корн," Справочник по
capacitor bank C > 0.030 Micro F. It is shown математике для научных работников и
that an approach of taking into account influence
инженеров". М., «Наука», p 832 с, 1978.
of current distribution in the rotor on starting
characteristics used in building up the model [9] Пронин, Михаил Васильевич, and А. Г.
proved applicable for evaluation of HF energy Воронцов." Силовые полностью управляемые
losses caused by PWM SC. This Mathematical полупроводниковые преобразователи
model of electric drives done with synchronous (моделирование и расчет). СПб, ОАО
machines has been developed. It can use in the Электросила", Russia ,2003.
real time mode with the help of personal
computers. The system is intended for debugging
transistor drive microprocessor-based control
units and based on use of mathematical models.
REFERENCES
[1] M. Pronin, "Simulation and analysis of the
system with multiphase induction generator and
multi-stage active rectifier", Electrotechnika,
Russia, p. 55-61, no. 5, 2006.
[2] M .Pronin. A. Vorontsov,"Dependence of
current pulsations of multi-phase electrical
machine on reduction of winding pitch and
scheme of semiconductor converter". EPE-
PEMC, Portoroz, Slovenia, 2006.
[3] C. Cester, A.Kedous -Lebouc, and B. Cornut,
"Iron loss under practical working conditions of a
PWM powered induction motor". Magnetics,
IEEE Transactions, Vol. 33, Issue 5, Part 2, pp.
3766-3768, 1997.
[4] A. J.Moses, F. Anayi ,"Effect of PWM
Voltage Excitation in Iron Loss of Inverter Fed
Motors", IEEE Transactions on Magnetics,vol. 40
(2), no 2, pp. 762-765, 2004.
[5] A.Ruderman, R. Welch," Electrical Machine
PWM Loss Evaluation Basics", EEMODS 2005,
5 – 8 Heidelberg / Germany, September 2005..
[6] N.Serov, P.Kalachikov, M. Pronin, and
A.Vorontsov," Electro-transmission of dump
trucks BelA Z with capacity up to 136 t // An.
Mining equipment and electro mechanics",
Russia, no5, p. 22-25., 2005.
[7] М. В .Пронин, "Электроприводы и
системы с электрическими машинами и
полупроводниковыми преобразователями

82