Introduction of Doubly Fed Induction Machine in an ...

Available online at www.sciencedirect.com

ScienceDirect Energy Procedia 36 (2013) 1076 – 1084

TerraGreen 13 International Conference 2013 - Advancements in Renewable Energy and Clean Environment

Introduction of Doubly Fed Induction Machine in an Electric Vehicle R.Babouria, D.Aouzellagb, K.Ghedamsia,b* ab

Laboratory of Renewable Energy Mastery, University of Bejaia, Bejaia 06000, Algeria

Abstract

This work is aiming to study and control of drive train of an electric vehicle based on doubly fed induction machine (DFIM), the power structure of this machine and the control strategy applied allow operating over a wide range of speed variation, for both applications: engine and recovery. Therefore the power of the machine can reach twice its rated power. After modeling different parts of the drive train a numerical simulation in MATLAB / Simulink is carried out. The results show the good performance of the vector control, and the structure of power applied to the DFIM.

access under CC BY-NC-ND license. © Authors. Published by Elsevier Ltd. Open ©2013 2013The The Authors. Published by Elsevier Ltd. Selection and/or peer-review under responsibility of the TerraGreen Academy Academy. Selection and/or peer-review under responsibility of the TerraGreen

Key words: DFIM, PWM Converters, Battery, Gearbox, control of vehicles drive train.

1. Introduction Industrial applications of variable speed drives require increasingly important performance and a maximum reliability and minimum cost. Indeed, currently the use of AC machines is becoming more common as these machines are characterized by their robustness and longevity compared to commutator machines [1-2]. Literature shows the great interest shown in the double-fed induction machine (DFIM) for various applications: as a generator for wind energy and for certain industrial applications, such as rolling and traction or propulsion maritime. Indeed, most work on this machine have been the subject of the study of the structure where the stator is directly connected to the network and the rotor powered by a power electronics converter. The advantage of this solution is that the converter is sized at 30% of the rated power of the system and therefore the variation of speed limit near the speed of synchronization [3-4]. * Corresponding author. Tel.: +213-556-463-120; fax: +0-000-000-0000 . E-mail address: rabah.babouri@yahoo a .fr.

1876-6102 © 2013 The Authors. Published by Elsevier Ltd. Open access under CC BY-NC-ND license.

Selection and/or peer-review under responsibility of the TerraGreen Academy doi:10.1016/j.egypro.2013.07.123

1077

R. Babouri et al. / Energy Procedia 36 (2013) 1076 – 1084

However, the objective of this work is to operate the DFIM in a wide range of speed variation, for application in a drive train of an electric vehicle; for this the machine is connected through two power converters with pulse wide modulation control (PWM), these converters are both powered by a battery, which is a key element for development of electrical vehicles, namely the energy density is low and the charge time very long [4]. In the drive train we use only one machine (DFIM) for the motorization of the vehicle, and for recovering energy during braking. The advantage of the power structure chosen is not only to operate the machine in a wide range of speed variation, but also to give to the machine the capacity to operate up to twice its rated power. So the power density is improved. Figure (1) illustrates the schematic diagram of the drive train:

DC

Wheel

Battery AC Gearbox DFIM PWM control

Defffirential

DC AC

Wheel

P Powers Fig (1): Representative diagram of the electric vehicle drive train

The ability of the DFIM to start with high torque makes possible the elimination of clutch and gearbox. The torque is the size dimensioning; therefore the machine must be heavier and bulky, so more expensive. The use of fixed ratio gearbox overcome these problems and allows to have a simple machine that can provide the required torque. The power electronic converters used for power transfer between the battery and the DFIM are sized at 100% of rated power of the machine, those are bidirectional converters with PWM control, they absorb power from the battery when the machine operates as a motor and they provide to it when the machine operates as a generator (braking). Semiconductors used depends on power level passing converters; for low powers are used IGBT. For high power converters based on IGCT or GTO semiconductors can be used. Variable-speed drives with a rated power up to 40MW (IGCT) or 100MW (GTO) have been installed. A disadvantage of these semiconductor types is their lower switching frequency, compared with IGBT’s [5]. 2. DFIM model Two-phase equivalent model of the DFIM represented in the reference (dq) linked to the rotating field is given as follows f [6-7]:

1078


緊懸鎚鳥噺迎鎚件鎚鳥髪鯨砿鎚鳥伐降鎚砿鎚槌 """"""""""""" 衿懸鎚槌噺迎鎚件鎚槌髪鯨砿鎚槌髪降鎚砿鎚鳥 """""""""""""" 芹懸追鳥噺迎追件追鳥髪鯨砿追鳥伐岫降鎚伐降追岻砿追槌衿懸噺迎件髪鯨砿髪岫降伐降岻砿追追槌追槌鎚追追鳥菌追槌 With S: Laplace Operator In order to achieve good decoupling between the axes d and q, we define intermediate voltages as follows: 暢懸鎚鳥伐懸追鳥噺懸痛鎚鳥挑認崔暢懸追鳥伐懸鎚鳥噺懸痛追鳥崔

挑濡暢

懸挑認追槌暢懸追槌伐懸鎚槌挑

懸鎚槌伐

濡

噺懸痛鎚槌

噺懸痛追槌

(1)

(2)

(3)

Coupling terms appear to compensate; 鶏怠鳥 , 鶏怠槌 , 鶏態鳥 , 鶏態槌 , these expressions allow to obtain relations between the intermediate voltages and the stator and rotor currents in d or q axes. So: 緊懸痛鎚鳥噺迎鎚岫な髪鯨劇鎚購岻件鎚鳥髪鶏怠鳥衿懸痛鎚槌噺迎鎚岫な髪鯨劇鎚購岻件鎚槌髪鶏怠槌 (4) 芹懸痛追鳥噺迎追岫な髪鯨劇追購岻件追鳥髪鶏態鳥衿懸噺迎岫な髪鯨劇購岻件髪鶏追追追槌態槌菌痛追槌 With: Ts= Ls / Rs; stator electrical time constant; Tr = Lr / Rr; rotor electrical time constant; σ = (1-M2/ (Lr.Lr)); dispersion coefficient. The coupling terms can be expressed as follows: 暢暢緊鶏怠鳥噺伐挑認迎追件追鳥伐降鎚砿鎚槌髪降追挑認砿追槌暢暢衿衿鶏怠槌噺伐迎追件追槌髪降鎚砿鎚鳥伐降追砿追鳥挑認挑認 (5) 暢暢芹鶏態鳥噺伐迎鎚件鎚鳥髪降鎚砿鎚槌伐降追砿追槌挑濡挑濡衿衿暢暢菌鶏態鳥噺伐挑濡迎鎚件鎚槌伐降鎚挑濡砿鎚鳥髪降追砿追鳥 From system of equations (5), the transfer’s functions following are obtained: 彫

"

怠斑眺

濡匂濡緊劇鎚鳥噺噺怠貸聴脹濡蹄蝶禰濡匂貸牒迭匂衿怠斑衿彫濡忍眺濡衿劇鎚槌噺噺

怠貸聴脹濡蹄怠斑眺認芹劇噺噺追鳥蝶禰認匂貸牒鉄匂怠貸聴脹認蹄衿衿怠斑彫認忍衿眺認劇噺噺菌追槌蝶禰認忍貸牒鉄忍怠貸聴脹認蹄蝶禰濡忍貸牒迭忍彫認匂

3. Converters model The matrix giving the model of powers electronics converters used is expressed as ̨æØØæ¬œ " "

(6)

1079


懸銚津な怠煩懸長津晩噺戟待煩ど戴懸頂津伐な

伐ななど

ど鯨銚伐な晩煩鯨長晩な鯨頂

(7)

4. Battery model The model of battery used for application in electric vehicle should have the specifications as follows [8]: It should simulate the variation of the battery’s terminal voltage on certain load demand or current demand; It should be simple and require limited times for mathematical calculation and iteration; The model should be involved with as few as possible or none of the parameters that are related to the battery’s chemical process. There have been many proposals battery model; one of this is the Thevenin equivalent circuit, shown in figure (2). It is a linear electrical battery model [9]. 迎継待

迎待系待

荊長

+ 撃長

-

Fig(2): (2) Thevenin h i equivalent i l circuit i i off battery b

撃長噺継待伐撃頂待伐迎荊長 5. Vehicle Dynamics Equation governing vehicle dynamics is given as following [10]: 鳥蝶蝶劇追噺盤繋繋追髪繋銚追鳥髪繋直匪堅髪 f警塚寧

(8)

(9)

鳥痛

繋銚追鳥撃塚

繋追

警塚 ̌潔剣嫌岫糠岻

警塚 ̌嫌件券岫糠岻警塚 ̌

繋追

糠

Fig (3 (3): ): Various forces applied to the vehicle

繋追噺鶏系追蝶系追噺どどな岾な髪寧峇怠待待 Which 系追 is called the rolling resistance coefficient and P is the normal load on the Wheel. 繋銚追鳥噺どの貢畦捗系鳥岫撃撃塚髪撃栂岻態 Where ρ is the air density, Af is the frontal area of the vehicle, Cd is aerodynamic coefficient, Vv vehicle speed and Vw is the wind speed.

(10) (11) (12) is the

1080


繋直噺警塚 ̌œÆº岫c岻

(13)

Where g is the earth gravity and Mv is the total weight of the vehicle. 6. Vector control of the DFIM A vector controlled doubly fed induction machine is an attractive solution for high restricted speed rang electric drive and generation application, it consists in guiding an electromagnetic flux of the DFIM along the axis d or q.[5] In our case we choose the direction of reference (d,q) according to the direct stator flux vector 剛鎚鳥 , so the model of steady DFIM will be simplified as follows:

懸鎚鳥噺迎鎚鳥件鎚鳥緊懸噺迎件髪降剛鎚槌鎚鎚槌鎚鎚鳥懸噺迎件伐降剛追槌追鳥追追鳥追芹菌懸追槌噺迎追件追槌髪降追剛追鳥

(14)

Such as:

降追噺降鎚伐降 (15) The magnetization of machine is assured by the rotor direct current, so the stator current in the d axis is taken to zero (件鎚鳥噺ど ). The current and voltage in this line are then in phase: 懸鎚槌噺懸鎚 and 件鎚槌噺件鎚 (16) In this case we obtain a unity power factor at the stator, so the stator reactive power is zero Qs = 0.These simplifications lead to the electromagnetic torque expression: 劇勅陳噺喧剛鎚件鎚槌 (17)

From the expressions of equations which have been established, we can draw a connection summary table setting the objectives of the control strategy with the references of action variables involved: Objectifs

剛鎚鳥噺剛鎚噺剛鎚津剛鎚槌噺ど

芸鎚噺ど岫潔剣嫌砿噺な岻茅劇勅陳噺劇勅陳

References 茅件追鳥噺

剛鎚津警

詣鎚件警鎚槌貸追é捗茅件鎚鳥噺ど茅劇勅陳茅件鎚槌噺計脹勅陳

茅件追槌噺伐

Table (1): control strategy applied to the DFIG model

7. Powers distribution The distribution of stator and rotor active powers is a requirement in the control strategy to be applied. Indeed, this allows increasing the range of speed variation and the power density of the machine. Such as if the stator and rotor resistance windings terms are neglected, the following relationship is imposed: 牒濡

牒認

噺

摘濡

摘認

(18)

Therefore, the stator and rotor active powers distribution, involve the stator and rotor pulses distribution and vice versa. Working with a slip g = -1 we obtain the following relationship: 摘濡貸摘摘噺認噺伐な (19) 摘濡

So:

摘濡

1081

R. Babouri et al. / Energy Procedia 36 (2013) 1076 – 1084 摘

降鎚噺伐降追噺賑 (20) 態 The diagram representing the complete system with the control strategy applied is given by figure (7). is_abc abc bcc is

AC Ω

*

Whee e l

+-

Geearbox x

Deffirentiall

+ Battery -

DC

Ω DF D FIM M

P M PW generator

PI Tem*

ib

ir_ab _a a c Wheel

ir AC C

ab bc

φ*

d dq

DC vs*_dq

Reffereence c s currents calculati tion ti on n

Cu C urrents ttss control Repartition off R pullsation oon n

降鎚

降追

vs*_abc bcc

d dq vr*_dq

1/ss

abc bcc

PWM M generator

vr*_abc bcc

肯鎚

肯追

Fig (4): Control diagram of the electric vehicle drive train

8. Simulation results A speed control with pulses repartition 岫嫌健件喧噺伐な岻 is applied to the DFIM for a path of a road with variable slopes. The overall system simulation is performed on the MATLAB / Simulink, the following figures show the simulation results: DFIM speed and reference speed (rpm)

3500 3000 2500 2000 1500 1000 500 0 -500

0

1

2

3

4 t (s)

5

6

7

DFIM, reference and resistance torque (N.m)

Fig (5): DFIM speed and reference speed (rpm) 800 :Cem (N.m) :Cem*(N.m) :Cr (N.m)

600 400 200 0 -200 -400 -600 -800

0

1

2

3

4 t ((s))

5

6

Fig (6): DFIM, reference, and resistance torque (N.m)

7

8

1082

R. Babouri et al. / Energy Procedia 36 (2013) 1076 – 1084 5

x 10

Stator, rotor and battery powers (W)

3

2

1

0

-1

-2

0

1

2

3

4 t(s)

5

6

7

Fig (7): Stator, rotor and battery powers (W)

Stator and rotor pulsations (rd/s)

400 300 200 100 0 -100 -200 -300 -400

0

1

2

3

4 t (s)

5

6

7

Fig (8): Stator and rotor pulsations (rd/s) 620

Battery voltage (V)

600 580 560 540 520 500

0

1

2

3

4 t (s)

5

6

7

Fig (9): Battery voltage (V)

9.

Results interpretation According to the simulation results it is noted that the DFIM operates over a wide range of speed variation (twice the nominal speed), while following the reference imposed. The distribution of pulses applied for control of DFIM has allowed to have the distribution of stator and rotor actives powers, however, there is a slight difference and this is due to the stator resistance that is greater than the rotor resistance, and we note that the total power exchanged between the battery and DFIM equal to the sum of the stator and rotor actives powers. Indeed, when the driver requests the machine to beat a rib with a given speed the, DFIM can provide power equal to twice its rated power. So for positive slopes the driver requests the DFIM to provide the torque required to overcome the resistive torque imposed by the vehicle, so the machine is operating in


motor mode, and absorbs power from the battery, for negative slopes (braking) the DFIM operates in generator mode and recharges the battery. 10. Conclusion The aim of this work is to integrate DFIM in a drive train of an electric vehicle and show its performances. Indeed, the simulation results obtained show that the DFIM can operate over wide range of speed variation, which allows to reduce the size of gearbox which is complicated and very expensive systems. And the power of the machine rises up twice of its rated power, which allows to increase its power density. Seen these benefit the DFIM is a very good alternative for use in a drive train of electric vehicles.

Nomenclature 畦捗 : Frontal area of the vehicle 系待 : Battery capacity

系鳥 : Aerodynamics coefficient 繋銚追鳥 : Aerodynamic force

繋直 : Gravitational strength 繋追 : rolling force

継待 : Open circuit voltage 訣: Earth gravity

荊長 : Battery current

件鎚鳥件鎚槌件追鳥件追槌 : Direct and quadrature of stator and rotor currents 詣鎚 ,詣追 : Stator and rotor inductances 警: Mutual inductance

警塚 : Vehicle total weight

迎: Internal battery resistance 堅: Wheels radius

迎鎚 ,迎追 : Stator and rotor resistances 劇勅陳 : Electromagnetic torque 茅劇勅陳 : Reference torque

撃頂待 : The double layer capacity voltage

懸鎚鳥懸鎚槌懸追鳥懸追槌 : Direct and quadrature of stator and rotor voltages 撃塚 : Vehicle speed 撃栂 : Wind speed

撃長 : Battery voltage

1083

1084


砿鎚鳥砿鎚槌砿追鳥砿追槌 : Direct and quadrature of stator and rotor flux 降鎚降追 : Stator and rotor pulsations 硬: DFIM speed

硬茅 : Reference speed 貢: Air density References [1]: R.Campagne, L-A.Dessaint, H.Fortin-Blanchette. Real Time Simulation of Electrical Drive. Mathematique and computrs in simulation 63 (2003), 173-181. [2]: A. CHAIBA, R. ABDESSEMED, M.L. BENDAAS et A. DENDOUGA. Controle of torque and unity stator side power factor of the doubly-fed induction generator. Conférence sur le génie électrique CGE’04, EMP Alger, 2005. [3]:Ghania Mouna, Chaffaa, Khirddine, Benmahammed Khier. Adaptive Type-2 Fuzzy Control For Induction motor. second international conference on electrical systems, ICES’06, 2006, Oum El Bouaghi, Algeria. [4]: Lota Gidwania*, Harpal Tiwanib, R.C.Bansalc. Improving power quality of wind energy conversion system with unconventional power electronic interface. Electrical Power and Energy Systems 44 (2013) 445-453. [5]: Joris Soens, Karel de Brabondere, Johan Driesen, Ronnie Belmans, .Doubly Fed induction Machine: Operating Regions and Dynamique Simulation. EPE 2003-Toulouse, ISBN:90-75815-07-7. [6] : C. Belfedal, S. Gherbi, M. Sedraoui, S. Moreau, G. Champenois, T. Allaoui, M.A. Denai. Robust control of doubly feed induction generator for stand-alons applications."Electric Power Systems Research 80 (2010) 230–239. [7]: S.Persada, A.Tilli, A.Tonielli. Robust active-reactive Powers Control of a Doubly Fed Induction Machine. proc. of IEEEIECON’98, Aachen, Germany, sept.1998.pp.1621-1625. [8]: Nang Janping, ChenQuanshi, Cao Binggong. Support vector machine based battery model for electric vehicles. Xi’an Jiaotong University, Xi’an 710049, PR, China. [9]: Ziyad M. Salameb, Margaret A. Casacca, William A. Lynch. A Mathematical Model For Lead-Acid Battery. IEEE Transaction on Energy Conversion, Vol.7, No.1, 1992. [10]: Siavash Zoroofi. Modeling and Simulation of Vehicular Power Systemes. Master’s thesis in Electric Power Engineering, Chalmers University of Technology, 2008.

Appendix Drive train parameters """""畦捗噺なぱ"兼態 ,系鳥噺どぬに訣噺ひぱ

陳

鎚鉄

詣追噺どどなのねの茎詣鎚噺どどなのねの茎警噺どどなのな"茎 """"""""""""""""""""""""""

""""警塚噺なのどど"倦訣迎噺どどね"硬堅鳥噺どぬ"兼迎追噺どどにどひに硬迎鎚噺どどぬののに硬喧噺に """""""""""""""""""""""""""""""""""""" """"鶏津噺ばの"倦激戟津噺ねどど"撃硬津噺なのどど"堅喧兼 t 噺なにぱ"倦訣兼戴 ".