2013 Joint International Conference on Rural Information &

Co mmunication Technology and Electric-Vehicle Technoology (rICT & ICeV-T)

November 26-28, 2013, Bandung-Bali, Indonesia

Dynamiic Response Analysis off

Permanent Magneet Brushless DC Motor Drives
D for
City Electric C
Car Based on Matlab /Sim
mulink
Abdullah Assegaf, Aguus Purwadi, Tri Desmana Rachmildha,Yanuarsyah Harroen
email: [email protected] , [email protected]
Department of Electrrical Engineering School of Electrical Engneering&Informatics
Bandung Institute of Technology
Bandung, Indonesia

Abstract— Electric cars becomes one off interesting object Electric motor, however, has to orque-speed characteristic
for researchers since its ability to meet thhe requirements of which is close to the profile of tractive effort-speed of
environmentally friendly vehicles. The driving force of vehicles,. Since an electric vehicle is driven by electric motor,
electric car comes only from electric mottor, hence the car it is possible the system to use sing gle gear to get match the
performance is mostly determined by itss motor dynamic vehicle profile. This much simplifiies the structure of the
response. This paper presents the modelinng of PMBLDCM vehicle.
(Permanent Magnet Brushless DC Motor) with the load
characteristics of an urban city electric ccar. The dynamic
responses of torque and speed of the motor in constant torque
region is discussed Tests carried out using M
Matlab/Simulink in
an attempt to obtain data describing the dynnamic responses of
the motors driving city electric car.

Keywords—PMBLDC;motordrives;torque-sppeed
characteristics;dynamic response;Matlab/Simulin
nk;city electric car.

I. INTRODUCTION Figure 1. The tractive effort - speed proffile of vehicle with multi-
Electric vehicle developments can cerrtainly produce a gear system.
revolutionary improvement in vehicle perrformance, energy The characteristic of torque–speeed curve of the electric
conservation and reduction of pollutant em missions. Recently motor can be divided into two areas, i.e. the constant torque
the development of electric vehicle are w widely growing in area and constant power area . Thesee two operating areas of
electric vehicle, such as an electric city carr and rail vehicles electric motors is determined in two modes,
m namely basic rpm
for mass transportation system. These field of research mode (ωb) and the power limits mode. [2]
activities include electric motor drive system
m, management of • RPM based ( ωb ) Mode
energy sources and its controllers, and mechanical power This mode of operation uses base speed of the rotor as the
train system. Objective of the main researcch field of electric boundary to traverse from torque constant operation to the
cars in particular: 1) high efficiency annd low pollutant constant power operation, i.e.
emissions, 2) structure is compact and sim mple, and 3) high -Constant torque area: ω < ωb T= =TT-constant
reliability and low prices. To achieve the oobjectives, electric TT-constant= ctrl xTmax (1)
motor controllers and energy storage arre most important
components [1]. -Constant power area: ω ≥ωb T = TP-constant,
௧ ௫ ௫
The tractive effort - speed profile of aan ICE ( internal TP-constant = ఠ
(2)
combustion engine) vehicle is given in figurre (1). The profile • Power limit mode (Pmax)
has a wide range of constant power operaation area, i.e. the Motor power limit values is defined
d as:
tractive effort hyperbolically decline with increasing speed. Pmax = Tmax x ωb (3)
But an ICE car usually depends on muulti-gear system to Motor controllers in this mode works
w on constant torque
produce constant power operations, but thee torque variations region if the power absorbed by the motor is less than Pmax
with the multi-gear system cause signifiicant reduction in motors and switch to the constant po ower operation when the
efficiency. So the multi-gear configuration ffor city electric car motor output power P is greater thaan the maximum power
becomes unfavorable. Pmax, i.e.
-Constant torque area: P< Pmax. T=TT-constant

978-1-4799-3365-5/13/$31.00 ©2013 IEEE

E
 ܶ
TT-constant= ctrl ◌ ௫ (4) mathematical modeling of the two motors are different.
PMSM uses a famous two axes dqo method, while BLDC
-Constant power area : P ≥ Pmax T = TP-constant
motor modeling uses phase variables.
◌ౌ
TP-constant= (5) The general voltage equation in cylindrical rotor of
ఠ
An electric motor which is used in electric car is required electrical machines is [9]:
to have the characteristics of high torque at low speed to ݅ ◌ௗ ݅
obtain high initial torque. On the other hand, adequate ൌ ◌ܴ ◌ ◌
acceleration and wide speed bounds for operation with (6.a) ◌ௗ௧
constant power. For city electric car, the start-go mode of or
operation (constant torque operation) is predicted to happen
∂ ∂θ ∂
[v] = [ R][i] + [ L(θ )] [i] + { [ L(θ )]}[i]
∂t
14243 144 ∂t ∂θ42444 3
most of the time . transforma tion emf rotation emf
(6.b)
Permanent magnet motor has superiorities compare to
other motors, since its torque per Ampere and efficiency is the BLDC motor are commonly of surface mounted type. This
highest [3-4]. PMBLDCM or later on called as BLDC motor, type of motor has no significant pole saliency, hence its
has the highest torque density and therefore will potentially reluctance is not a function of angle. Therefore the value of
have the lowest weight for given torque and power ratings the self inductances of each phases of the stator windings are
compare to dc motor, induction motor and switched the same, and so are the value of their mutual inductances,
reluctance motor (SRM). Unlike the PMSM which its stator i.e.:
winding is arranged for sinusoidal flux, the stator windings of Laa= Lbb = Lcc = L ( H ) and
BLDC motor are designed to produce trapezoidal back emf Lab = Lba = Lac = Lca = Lbc = Lcb =M (H))
shape. A three-phase BLDC motor only requires six discrete
The modeling of BLDC motor are relatively less presented positions in each cycle to observe its absolute position of the
in papers than paper with other motor type topics [5], so his magnetic rotor. This motor has three hall sensors placed in the
paper proposed modeling on BLDC motors in simulink. stator and are separated 600 electric angle. Thus the shape and
Experiments were carried out on BLDC motor using FPGA- position of each induced emf can be determined through its
based with IGBT inverter intended for validation of the sensor responses.
simulink model presented. Some responses from simulink Because of the shape of current and voltage emf are non-
models and experiments using FPGA component are displayed sinusoidal, the analysis is done using the variable phases abc
in the form of curves and the responses of the simulink model rather than using the popular dqo model.
has shown response with a very high accuracy. From equation (6.a) and (6.b), the motor voltage equations
Control and materials BLDC motor and PMSM are are :
constantly being developed. PID control technique in analog
and hybrid control to enhance their responses are explained in ⎡Vas ⎤ ⎡Rs 0 0⎤⎡ias ⎤ ⎡L − M 0 0⎤⎡ia ⎤ ⎡eas ⎤
many papers [6-7]. ⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥
⎢Vbs ⎥ = ⎢0 Rs 0⎥⎢ibs ⎥ + p ⎢0 L − M 0⎥⎢ib ⎥ + ⎢ebs ⎥ (7)
Operation under constant torque area of PMSM with ⎣⎢Vcs ⎢⎣0 0 Rs ⎥⎦⎣⎢ics ⎢⎣0 0 L − M ⎣⎢ecs
abrupt change of the mechanical load has been presented in ⎦⎥ ⎣⎢ic
detail. [8]. The analysis of the motor has been presented in where Rs is the resistance, L is self inductance, M is mutual
famous dqo model since the stator winding of PMSM is inductance stator winding and eas, ebs and ecs are the stator
ditributed in sinusoidal wave form. The analysis was given for back emf. The maximum emf is
motor with small value of inertia, and the results show a very Ep = λpωr (8)
good rise time value and a very small ripple response. These
are the reasons that the motor favorable for component of with λp= permanent magnet flux linkage (Weber).
drivetrain of electrical vehicle. Let L1=L-M, equation (17) can be simplified as:
This paper discusses the BLDC motor torque and speed
⎡Vas ⎤ ⎡1 0 0⎤ ⎡ias ⎤ ⎡1 0 0⎤ ⎡ias ⎤ ⎡eas ⎤
responses to sudden changes of required load torque under ⎢ ⎥ ⎢ 0 1 0⎥ ⎢i ⎥ + L⎢ 0 1 0⎥ p⎢i ⎥ +⎢ e ⎥
V
⎢ bs ⎥ = Rs⎢ ⎥ ⎢ bs ⎥ 1 ⎢ ⎥ ⎢ bs ⎥ ⎢ bs ⎥ (9)
constant torque area of operation. The speed of the motor is
only observed for the rotor mechanical speed less than the ⎢⎣Vcs ⎣⎢0 0 1⎥⎦ ⎢⎣ics ⎢⎣0 0 1 ⎢⎣ics ⎢⎣ecs
base speed. The flux weakening operation is not discussed in Electromagnetic torque can be obtained by using the
this paper. following equation :
II. MATHEMATICAL MODELING AND SIMULINK
Te=[easias + ebsibs+ecsics]/ ωr (Nm) (10)
PMSM and BLDC motor are two permanent magnet
Instantaneous induced emf can be written in the following
motors which widely used in electric cars. The main
difference is the PMSM has a sinusoidal back emf while equation :
BLDC motor generates trapezoidal back emf, so the
eas=fas(θr)λpωr
 ebs=fbs(θr)λpωr (11) ⎡ 1
0 0 0 ⎥
⎤
⎢L
ecs=fcs(θr)λpωr ⎢ 1 ⎥
⎢ 0 1
0 0 ⎥
⎢ L1 ⎥
where the function fas (θr), fbs (θr) and fcs (θr) are the hall sensor B =⎢ 1 ⎥
(20)
⎢ 0 0 0 ⎥
responses as shown in table 1. ⎢ L1 ⎥
⎢ 1⎥
⎢ 0 0 0 − ⎥
J
⎢ 0 0 0 0 ⎥⎦
⎣

C=[1 0 0 0 0;0 1 0 0 0;0 0 1 0 0; 0 0 0 1 0;0 0 0 0 1] (21)

Table 1. Hall sensor modelled as a function of rotor angle
D= [0 0 0 0 ; 0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0] (22)
Theta_elec (θ r) fa (θr) fb (θr) fc (θr)
00- 600 BLDC motor in state space with Simulink modeling is
1 -1 1 − 6θr /π
600- 1200 1 6θr/π −3 -1 shown in figure 3.
1200- 1800 5 − 6θr/π 1 -1 Te
1
1800- 2400 -1 1 6θ r/π −7 Va 5
6

2400- 3000 -1 9 − 6θr/π 1 2

x' = Ax+Bu
EMF a angle
Vb BLDC_cons mEMF b
3000- 3600 6θ r/π−11 -1 1 3 y = Cx+Du EMF c
SS Func Motor
Substitute of equation (11) to (10),we will get a new Vc
constants
electromagnetic torque equation as follows: Friction B
Te=λp[fas(θr) ias + fbs(θr) ibs + fcs(θr) ics ]
1
(12) 4
T_load
ia
2
Dynamic equation of a system with a moment of inertia J, m ib
3
friction coefficient B with the load torque TL is: K- ic
Gain 4
Te = Bωr +Jdωr/dt + TL (13) wn

And the relationship between the position rotor (θr) with

rotor angular velocity is:
dθr/dt = (P/2) ωr (14) Figure 2. Simulink model of BLDC Motor
Equation Eq.(12) through Eq(16) can be written in the III. THE SIMULATION RESULTS
form of state space equations. The general equation in the state
Simulations performed on BLDC motor with power rating
space is:
10 hp, 72 V and 1500 RPM speed and rotor flux 0.3 Weber,
•
X = Ax + Bu pole number 4 and the stator resistance R = 0.0575 Ohm and
(15)
small value of Inertia J=2e-4 kgm2. Torque and speed
y = Cx + Du
(16) variations of the motor are shown in table 2.
where Tabel 2. Mecanical load variations with time
x = state variable vector, Time Mechanical Mecanical
= first-order derivative of the state variables, ( s) Torque (Nm) speed (RPM)
y = output vector,
u = forcing function, 0 0 500
A = state matrix, 0.1 0 500
B = input matrix, 0.1 37.5 500
C = output matrix
D = transmission matrix. 0.15 37.5 500
The state space form of the BLDC motor variables are:: 0.15 37.5 1000
x= [ias ibs ics ωr θr]t (17) 0.2 37.5 1000
u = [vas vbs vcs TL]t (18) 0.2 12.5 1000
⎡ RS λp ⎤
⎢− 0 0 −
L1
f as (θ r ) 0 ⎥ 0.25 12.5 1000
⎢ L1 ⎥
⎢
⎢0 −
RS
0 −
λp
f bs (θ r ) 0 ⎥
⎥ 0.25 12.5 500
⎢ L1 L1 ⎥
⎢ RS λp ⎥ tstop 12.5 500
⎢
A= 0
⎢
0 −
L1
−
L1
f cs (θ r ) 0 ⎥
⎥ (19) Dynamic responses of torque an speed of BLDC motor are
⎢ ⎥
⎢ λ p f (θ ) λp
f bs (θr )
λp
f cs (θr )
B
− 0 ⎥ shown in the following figures.
⎢ J as r J J J ⎥
⎢ P ⎥
⎢0 0 0 0⎥
⎢ 2 ⎥
⎣ ⎦
 Torque ( E l & Load) ( Nm)
50

1000
40 Speed (Re erence & A ctual Value) ( RPM
R )

500
30

0
20

-500
10

-1000
0

-1500
-1 0
0 0.0 5 0.1 0.1 5 0.2
2 0.25 0.3
Time(s)

a) -2000
0.0995 0.1 0.1005 0.101 0.1015 0.102 0.1025
Time(s)
0.103 0 .1035 0.104 0 .10 45

T orque ( E l & Load) ( N m)

50
c)
40
Figure 4. Speed Respoonse for:
30 a) full period of test 0 -0.3 second b) Zooming view at mechanical
20
speed change from 0 RPM to 500 RPM M c) zooming view at 0.1
second on disturbance of torrque variation.
10

IV. CONCLUSION:
-10
0.0995 0.1 0.1005 0.101 0.1015 0.102 0.1025 0.103 0.1 035 0.104 0 .1045
Time(s)
The torque of BLDC motor has a very
v fast response, i.e. the
b) change of load torque from zero to 75% of its rated torque( 50
Nm), its rise time is less than 0.00255 second and its ripple is
36
about 1 %. At the instant of referencee speed change from 500
35
RPM to 1,000 RPM, the torque sufffers maximum overshoot
34 almost 25% of its steady state value, but
b it dies out in less than
33 0.0025 second. The speed response due to speed reference
32
value variations is very good with settling time less than
31
0.00025 second, but the speed respon nse due to the occurrence
of disturbance of torque variation haas a much longer settling
30
time(0.003 s). It is clear that BLDC motor
m has a good dynamic
29
response to propel city electric car, but
b of course the electric
28
0.1495 0.15 0.1505 0 .151 0.15 15 0.152 0.15525 0 .15 3 0.1535 0.154
motor control must be carefully designned.
Time(s)

c)
Figure 3 Torque Response :
a)full period of test 0 -0.3 second b) Zooming vieew at 0.1 second on
insertion of load c)Zooming view at 0.15 secondd on speed change
Speed (Reference & Actual) (RPM)

1500

1000

500

-500

-1000

-1500
0 0.0 5 0.1 0.15 0.2 0 .25 0 .3
Time(s)

a)
S pe e d ( Re fe re nce & Ac tua l value ) (RPM
55 0

50 0

45 0

40 0

35 0

30 0

25 0

20 0

15 0

10 0

50

0
0 0.5 1 1.5 2 2.5 3 3.5
Time(s) x 10
-3

b)
V. REFERENCE:
1) M. Ehsani, Yimin Gao and Sebastian Gay,” 6) Fernando Rodriguez and Ali Emadi “A Novel
Characterization of Electric Motor Drives for Digital Control Technique for Brushless DC
Traction Applications” Advanced Vehicle Motor Drives” IEEE Transactions On
systems Research Program Department of Industrial Electronics, Vol. 54, No. 5, October
electrical engineering Texas A&M University, 2007
IEEE, 2003
7) “Brushless DC Motor Control Made Easy
2) C.L. Chu,. M. C. Tsai and H. Y. Chen, " “Microchip Technology Inc,2002.
Torque Control of Brushless DC Motors
8) Abdullah Assegaf et.all” Dynamic Response
Applied to Electric Vehicles “ IEEE,2001
Analysis of Permanent Magnet Synchronous
3) Pragasen Pillay and Ramu Krishnan Motor Drives for City Electric Car” , The 5th
”modeling, Simulation, and Analysis of International Conference on Information
Permanent-Magnet Motor Drives, Part II: The Technology and Electrical Engineering, 2013
Brushless DC Motor
9) Haroen, Yanuarsyah ”Electric Machines
Drive”, IEEE Transactions On Industry
Dynamics: Modelling, Analysis and
Applications, Vol. 25, No. 2, March -April
Applications”, 1st edition, Bandung Institute
1989.
of Tecnology publisher, 2010 (originally in
4) Lin Bai,” Electric Drive System with BLDC Indonesian language)
Motor” Department of Electrical and
Computer Engineering Purdue University
Calumet Hammond, IEEE, 2011
5) Ming-Fa Tsai, et. all, “Model Construction and
Verification of a BLDC Motor Using
MATLAB/SIMULINK and FPGA Control”
6th IEEE Conference on Industrial Electronics
and Applications,2011