Comparison Between Two Inverter Topologies for ...

4 downloads 650 Views 430KB Size Report
characteristics of slip energy recovery drives (SERD) employing a three phase bridge recovery inverter. (topology “A”), are compared.with the characteristics of ...
Comparison Between Two Inverter Topologies for Application In Industrial Drives M.N.Eskander 0.Arafa Electronics Research Institute Cairo, Egypt

M.E.Adelhakiem S.A.Elhakiem Faculty of Engineering, Cairo University Cairo, Egypt

SERD employing a three phase bridge inverter (topology “A”), and the SERD employing three single phase bridge inverters working in the flywheeling mode (topology “B”). The importance of such study lies in deciding whether the advantages of type “B” inverter ovemdes the complexity of its control circuits. This decision depends on the amount of increase in the system power factor, as well as the decrease in the harmonics injected into the grid due to employing such inverter I the SERD.. In this paper the characteristics of the SERD employing a three phase line commutated inverter are deduced and compared with the characteristics of SERD employing three single phase line-commutated bridge inverters with controlled flywheeling. The comparison comprises the waveforms of voltages and currents of the stator, rotor, d.c. link, inverter output, and recovery transformer output. Also the turns ratio of the recovery transformer used with each inverter is calculated on the basis of equal speed ranges for SERD employing them. The harmonics injected into the grid, and the power factor of the system employing inverter “A” are compared with those of the system employing inverter “B”. Experimental implementation of SERD employng three single phase bridge inverters with controlled flywheeling is done . The waveforms deduced theoretically compared with experimental ones verified the theoretical results. The deduced results are in favor of employing inverter “B”, in spite of the complexity of control circuit.

ABSTRACT In this paper the steady state characteristics of slip energy recovery drives (SERD) employing a three phase bridge recovery inverter (topology “A”), are compared.with the characteristics of SERD employing three, single phase bridge inverters operating in the flywheeling mode (topology ’73”). The comparison comprises the waveforms of voltages and currents of the stator, rotor, d.c. link, inverter output, and recovery transformer output. For accurate and fair comparison, the turns ratio of the recovery transformer used with each inverter topology is calculated on the basis of equal speed ranges for the drives employing them. The harmonics injected into the grid, and the power factor of the system employing inverter “A” are compared with those of the system employing inverter “B”. Experimental results are given for the SERD employing the three single

-

phase inverters topology. These results verified the theoretical analysis. 1.INTRODUCTION The slip-energy recovery drives (SERD) are efficient variable speed drives applicable in high power industrial applications. The drawback of SERD is its high reactive power requirements, particularly at the upper end of the speed range. Although previous work [ 1-31 showed that the application of the controlled fly-wheeling to the recovery inverter improves the supply power factor and reduces the reactive power requirements, no comparison was done between the characteristics of the 0-7803-7090-21011$10.0002001 IEEE.

1100

TSIE 2001, Pusan, KOREA

-Vh-

V, V., =

Vb,

-Vm-

-R,+L,P L p MP -MP/2 -MP/2

-L,w

-MP

-(MP+&Mm)/2

(-MP+&Mw)/2

R,+L,P Ma, (&MP+&Mw)22 0 R , + L,P MY 3MP/2 M,P R, + L,P 3MP/2 M,P M,P

-(&MP+&Mm)/2 M*P M P R, + L,P

-

-IhI, I,, 1, -_I" -

(1)

inverters operating in controlled flywheeling mode (type B). For both topologies the following assumptions are made: 1.The ac supply and the step up transformer have negligible intemal impedance. 2.The semiconductor switches are ideal.

Where,L, and L,.are the self inductance for stator and rotor respectively, R, and Rr are the stator and rotor phase resistance respectively, M is the mutual inductance between a stator phase and a rotor phase, Mr is the rotor mutual inductance, P=d/dt, vds and V,, are the transformed stator voltages, v,r,vbr,v,r are the 3-phase rotor voltages, and oe is the synchronous speed. The electrical torque is given by: Te = (M.p/2) [ 3.4, .(Lr - Ibr) + 4 s .(2&tr- Ibr - Lr)] (2) The electro-mechanicaltorque equation is : T, = TI + J.P.0, + B. .Om (3) Where J is the moment of inertia, B is the damping coefficient, .Om is the rotor speed, and p is the number of pole pairs

IV.SERD PERFORMANCE CHARACTERISTICS The characteristics of the SERD employing inverter topology A are simulated and compared with characteristics of SERD employing inverter topology B .However, to have a fair and accurate comparison, a basis of comparison should be set first. Comparing the performances of inverters A and B lead to the following relation: m=2*ml Where m and ml are the turns ratio of transformers for type A and type B inverters respectively. According to the basis of comparison determined above, the recovery transformer turns ratio used with inverter B is simulated with an input/output ratio equals 110/380, while this ratio is doubled for the recovery transformer used with inverter type A.

Ihm phu. upply

,,~,

mmrter

Fig.(l) Schematic Of Slip Energy Recovery Drive 1II.INVERTER TOPOLOGIES Figure (2) illustrates the flywheeling inverter topology studied in this paper, namely the three single- phase bridge

1101

ISIE 2001, Pusan, KOREA

The algorithm cited in [3] was used to calculate the instantaneous value of the 3-phase active power, reactive power, and power factor supply side. These are shown in figures (7). From these figures, it is concluded that supply reactive power demand for inverter “A” is nearly twice that of inverter “B”. This leads to an improvement in the power factor of the system involving inverter “B” by 17.2%.

Figures (3) to (13) illustrate the two systems performance characteristics as described below: The rotor line voltages are shown in Fig.(3). The rotor voltage is observed to be composed of two main components; a high frequency component induced by the rotor current ripples, superimposed on a low frequency component equal to the rotor slip frequency. The main observed difference between the rotor voltage of the two cases is the larger amplitude of the high frequency harmonics for the system involving converter “ A”. The recovery transformer currents for both the low voltage side (motor side) , and the high voltage side (grid side) are shown in figures (4)and ( 5 ) respectively. The figures show that the pulse width of the inverter “B” current is narrower than that of “A”, leading to higher harmonic contents in the currents of “B”. However, due to the delta connection of the recovery transformer of inverter ‘B”, the transformer currents are composed of two adjacent narrow pulses, and its waveform depends on the duration of the freewheeling period. During the freewheeling period, the dc current does not flow in the recovery transformer, leading to an improvement in the power factor of the recovery circuit of inverter “B”. The supply current of both systems shown in fig.(6) consist of the instantaneous sum of the high voltage side current from the recovery transformer, and the stator current. It shows a higher harmonic content than stator current, for both inverters A and B, due to adding the inverter harmonics. It is obvious that the supply current of inverter “A” is larger than of inverter “B” due to the contribution of the higher reactive component of inverter “A”.

,-

~~~

~

............................................................................

,,M)

..............“.........

............................................

0

i.“ 33

02

21

nm

i

0.5

04

sec

Inverter “B”

I

0

I

01

?I 6ma

03

sec

c4

os

Inverter “A” Fig.(3) Rotor Line Voltages

Fig.(2) Flywheeling Inverter Type “B”

1102

ISIE 2001, Pusan, KOREA

5 4

.;I

_

1 ............. ......-......-......._..^...........-..-..............

?

E . ............ ;............;........... ;... ........ 2 I..... ......A... .. .....A... ....-..i.. ....-.9 1 ..........-.. 5 0----- -: .: . . . . . . . . .............. ........ 3 ....

;

~

5 ......-.....i..” . . ...... ....-..-. .’........._. ......-... .1 : .

I.,”............ ;......

.......-....f.-.........

I .

......

__”

-3 ............. ;............ ;............ ;......................... p

............L;........... ;........... i ....................... . 1

4,

J

. t)

I

I

002

0

bm

Inverter Type”A”-

Tim

I

om

004

0.08

sa

0.1

Inverter Type “B”

sec

T m Slc

Fig.(4) Low Voltage Side Current Inverter “A”

Inverter Type “A”

---

Fig.(6) Supply Current

I ............. ..............;..............;..............................

“Ob

62

011

Inverter “B”

01

ds

04

a

>

-1

rJ

002

004

006

Tlmo

sec

1

008

01

Inverter “A”

................... t 01

?I

i 04

013

J5

Fig.(’l)Sup PlY Powers and P.F.

s-

* 2 .....-...... .._.._.._.._.._.._..i.-......_.. ~.........._

l

4

. :.............. 4 ............. :.............. :.............. :...............

.#

,

o

o

m

5

!

j&

ON

oO~

308

01

iine Sec

1103

ISIE 2001, Pusan, KOREA

thyristors, and the interfaces to the PC controlling the drive. The firing and control circuits and the associated developed software were fully described in a paper under publication by the authors of this paper [4]. The machine employed as SERD is a 1.SKw,3-phase, 4-pole, 380/220volt, staddelta, slip ring induction machine. The predicted and measured waveforms of the rotor voltage, rotor current, inverter current, and supply current are shown in Figures (19) to (22).. The theoretical and experimental results are identical except for a slight deviation between those of the supply current waveforms,.due to approximations made in modeling the recovery transformer. However this deviation did not lead to a difference between the experimental and theoretical results of the family of speed-a characteristics shown in Fig.(23). These curves are plotted as function of different values of load torque. In short, the experimental results are in good agreement with theoretical results, hence confirming the accuracy of the analysis.

IV.HARMONIC ANALYSIS An algorithm based on Fast Fourier Transform (FFT) is used to obtain the spectrum of the different current waveforms for both systems (with inverters “A” and “B) The fiequencies of the rotor current harmonics are given by : f, = n.S. f1 where, S is the slip, and fi is the supply frequency, and n is an integer which could be positive or negative according to the sequence of the harmonic.The stator current harmonics are induced by the rotor current harmonics, and are given by : f, = fi [ 1 & 6.K.S] Where K =1,2,3.. .etc.

It is clear in the supply current spectra shown in Fig.(9) that the 5thand 7th harmonics are more pronounced in the supply current of inverter “A”. This is due to the attenuation effect of the lower turns ratio of the recovery transformer used with inverter “B”. These results give inverter topology “B” the upper hand over inverter topology “A”. Fig.(9) Supply Current Spectra for “B” %fwa”suoara

.-..-..--...-- --.-i__..____. :.._.____._ .-..; .--.---i i _.--....;_____L..-

4

I . _ _ . . _ _ _

d

...........i____.-__ .... ....... i....------.-I ._ 08 -.-.. ......i. i............_ .............. : .._.......... ! 0.5 -.-.. ......i .......-.....i............- I.............. i ..-......-.. 0.7 -*-..

@ 0 4 -.-..

..............................................................

-----. ..-.-i-----.-..-.i-.----.-~----..--..-...i..-.--..-.. . . f . j o t -.-....--.,... i.... __^_._ i_...._..-.. .j.._.___._._ 0 1 -... .. .....i.......-.....1............i.-............i...-.......... 03

1‘

f

i

I . . .

V.EXPERIMENTAL, RESULTS A slip power recovery scheme employing 3-single phase bridge inverters operating in the flywheeling mode was built together with the required firing circuits of the bridges’

1104

ISIE 2001, Pusan, KOREA

bridge recovery inverters operating in the flywheeling mode . The comparison comprises the waveforms of voltages and currents of the stator, rotor, d.c. link, and inverter. The harmonics injected into the grid from system employing inverter “A” are compared with those of the system employing inverter “B”. Results showed that the power factor of the SERD employing inverter “B” topology was higher than the power factor of the SERD employing inverter “A” topology by 17.2%. Results showed also that the harmonics content of the recovery current injected into the grid when using inverter “B” is less than those injected when using inverter “A”. These results ,encourage employing the inverter topology “B” with SERD, in spite of the relatively more complex requirements of its control circuit. Experimental set up was built for the SERD employing the three single - phase inverters topology, together with the necessary firing circuits and interfaces with the PC. The measured values are in good accordance with theoretical results .

fig.(1 1) Theoretical and Experimental invertercurrent

fig.(10) Theoretical and Experimental Rotor Current

REFERENCES 1.F.Xiaogang ,Chen Boshi, “Constant Slip Control of Induction Motor At Light Load”, Proc. Of IECEC, Washington D.C., August 1996,. 2. W.S.Zakaria, S.R. Alwash, A.A. Shaltout, “A Novel Double CircuitRotor Balanced Induction Motor For Improved Slip-Energy Recovery Drive Performance”,IEEE Trans. on E.C.,Vol.11,No.3, Sept.1996 3. W.S.Zakaria, S.R. Alwash, A.A. Shaltout, “A Novel Double CircuitRotor Balanced Induction Motor For Improved Slip-Energy Recovery Drive Performance”,IEEE Trans. on E.C.,Vol.11,No& Sept.1996.

4

1 ,

._.......... L..-.. 01

2”

......i............. .__.._.._ i

ol Elm

1105

:9

se:

4;

os

ISIE 2001, Pusan, KOREA