See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/282291981

REVIEW OF ELECTRIC MOTOR DRIVES

Research · September 2015

DOI: 10.13140/RG.2.1.1198.6408

CITATIONS READS
4 4,945

1 author:

Cheku Dorji
College of Science and Technology
13 PUBLICATIONS 12 CITATIONS

SEE PROFILE

All content following this page was uploaded by Cheku Dorji on 29 September 2015.

The user has requested enhancement of the downloaded file.

 REVIEW OF ELECTRIC MOTOR DRIVES

Cheku Dorji
chekudorji@cst.edu.bt
College of Science & Technology
Royal University of Bhutan

ABSTRACT

This paper presents the review of electrical machine drives and the control strategies, illustrated
using Matlab-Simulink tools. Matlab Simulink facilitates the building of individual functional
blocks like motor body and control components. A complete simulation model is formed by
combining the individual blocks systematically. The review of mathematical models of DC brush
motor, brushless DC motors (BLDC) and the permanent magnet synchronous motor (PMSM) are
presented in brief to explain the theory of drives. The BLDC requires electronic commutations to
drive the system, unlike the conventional DC brush motor there are no brushes mounted on the
rotor. The commutation logic depends on rotor positions and the trapezoidal wave is necessary
to achieve the constant torque production. The PMSM drive requires the field oriented theory
(FOC) to control the space vectors of magnetic flux, current and voltage. The vector control
provides the independent control of torque and rotor field excitation which is similar to armature
current and field control of DC brush motor.

Machine Drives and control, November 2012

1.1 The Mathematical Model of DC Brush Motor

The DC motors have been most popular choice for use in high performance systems. With
advent of power electronic switches used for power conversion, DC motor enhance the ability to
control their torque and flux independently over wide range of speed. The flux in the dc motor is
controlled by varying the field winding current and torque by changing the armature winding
current [1].

The electrical circuit model of a separately excited DC brush motor consisting of two circuits.
The field excitation, voltage (𝑉𝑓 ) generates a field current (𝐼𝑓 ) that flows through a variable
resistor (𝑅𝑓 ) and through the field coil. The armature circuit, on the other hand, consists of a
back-emf(𝐸𝑏 ), an armature resistance (𝐼𝑎 ) and an armature voltage (𝑉𝑎 ) [1][2][3].

Lf R

field circuit
La
Rf
Ia Armature circuit 

if

 Va
 Vf 
Eb  Laf mi f


Te
m
Tm

Figure (1.1) Equivalent circuit of Separately Excited DC brush motor

In the motor mode (𝑉𝑎 > 𝐸𝑏 ), armature current (𝐼𝑎 ) flows into the machine. The equations
described the dynamic behavior of a separately excited DC brush motor are
𝑑𝐼𝑎
𝑉𝑎 = 𝑅𝑎 𝐼𝑎 + 𝐿𝑎 + 𝐸𝑏 (Armature circuit)
𝑑𝑡
𝑑𝐼𝑓
𝑉𝑓 = 𝑅𝑓 𝐼𝑓 + 𝐿𝑓 (Field circuit)
𝑑𝑡

The torque can be expressed as

𝑑𝜔𝑚
𝑇𝑒 = 𝑇𝑚 + 𝐵𝜔𝑚 + 𝐽 𝑑𝑡

Machine Drives and control, November 2012

 Developed torque 𝑇𝑒 = 𝐾𝑡 𝐼𝑎
Back emf 𝐸𝑏 = 𝐾𝑡 𝜔𝑚 , where 𝐾𝑡 = 𝐿𝑎𝑓 𝐼𝑓
The torque constant (𝐾𝑡 ) is related to the geometry and the magnetic properties of the structure,
Where J = rotor inertia (kgm2)
B = viscous friction coefficient (Nm/rad/sec)
Tm & Te = load torque & electrical torque (Nm)
𝐿𝑎𝑓 = mutual inductance (H)
𝜔𝑚 = rotor speed (rad/sec)
The transfer equations of dc brush motor can be written as;
𝑉𝑎 (𝑠) = (𝑅𝑎 + 𝐿𝑎 𝑆)𝐼𝑎 (𝑠) + 𝐸𝑏 (𝑠)

𝑉𝑓 (𝑠) = (𝑅𝑓 + 𝐿𝑓 𝑆)𝐼𝑓 (𝑠)

𝑇𝑒 (𝑠) = 𝑇𝑚 (𝑠) + (𝐵 + 𝐽𝑆) 𝜔𝑚 (𝑠)

Therefore the s-domain of the dc brush model can be developed using matlab Simulink as shown
in figure (1.2 & 1.3)

Figure (1.2) Simulink model of dc motor

Machine Drives and control, November 2012

 Figure (1.3) Simulink model of dc motor using Simpowersystem

1.2 PID Controllers

The characteristics and the behavior of the electric motor drives are controlled by closed loop
systems. The most popular linear controller which gives the most efficient output-input ratio of
the drive system is the PID Controllers [3][4].
Consider a unity feedback system below consisting of actual system (plant), controller and error,
which is the difference of reference and feedback signal fed into closed loop system.

Re f error E(s) Controller U (s) Plant V (s)


 D(s) G (s)

H (s)  1

Figure (1.4) Schematic closed loop of PID controller

The each of controller (P, PD, PI, and PID) designs unique characteristics of the plant system.
The general equation for the PID control in the time domain is
𝑡
𝑑
𝑢(𝑡) = 𝑘𝑝 𝑒(𝑡) + 𝑘𝑖 ∫ 𝑒(𝑡) 𝑑𝑡 + 𝐾𝑑 𝑒(𝑡)
𝑡0 𝑑𝑡
Machine Drives and control, November 2012
Where 𝑢(𝑡) is the output signal, 𝑒(𝑡) is the error signal, 𝑘𝑝 the proportional constant, 𝑘𝑖 is an
integral constant and 𝐾𝑑 is the derivative constant.
The proportional (P) will reduces the rise time, increases the overshoot, and reduces the steady-
state error. The proportional control (P) equation be can represented in the s-domain as the
relationship between output and error, given by the open loop transfer function
𝑈(𝑠)
𝐷(𝑠) = = 𝐾𝑝 (𝑠)
𝐸(𝑠)
The proportional with an integral (PI) will decreases the rise time with effect of increase in both
the overshoot and the settling time; however it eliminates the steady-state error. Its time domain
and s-domain equations is given by
𝑡
𝑢𝑢(𝑡) = 𝑘𝑝 𝑒(𝑡) + 𝑘𝑖 ∫ 𝑒(𝑡) 𝑑𝑡
𝑡0

𝑈(𝑠) 𝐾𝑖 𝐾𝑝 . 𝑆 + 𝐾𝑖
𝐷(𝑠) = = 𝐾𝑝 + =
𝐸(𝑠) 𝑠 𝑠
The proportional plus derivative (PD) will reduce the overshoot and the settling time of the
desired response. Its time domain and s-domain equations be given by
𝑑
𝑢(𝑡) = 𝐾𝑝 𝑒(𝑡) + 𝐾𝑑 𝑒(𝑡)
𝑑𝑡
𝑈(𝑠)
𝐷(𝑠) = = 𝐾𝑑 𝑆 + 𝐾𝑝
𝐸(𝑠)
The combination of PID controllers confirmed the overall desired response of the system. In fact
each of the controllers is dependent of each other. Suppose the proportional controller (P) will
reduce the rise time but no improving of the steady-state error. An integral control (I) will
although eliminates the steady-state error it increases the overshoot and the settling time. A
derivative control (D) will improve the overall response of the system by reducing the overshoot,
and improving the transient response. Most often the PI controller is adequate to get the desired
response [4].
1.3 PI controller of DC brush motor
The current and the speed of the dc brush motor can be controlled by exploring the step response
of the PI controllers using the dc motor parameters associated with its respective transfer
function. The main objective is to track a given reference input and to reject any disturbance,
hence matching the ideal case of the output-input response of maximum efficiency [4].

Machine Drives and control, November 2012

The current controller forms the inner loop of the electrical part of the machine and it gets
feedback from the output measurement elements. The transfer equations of dc brush motor
related to output current is given armature circuit, which represents plant 𝐺(𝑠) of the open loop
system[4].
𝑉𝑎 (𝑠) = (𝑅𝑎 + 𝐿𝑎 𝑆)𝐼𝑎 (𝑠) + 𝐸𝑏 (𝑠)
𝐼𝑎 (𝑠) 1
= = 𝐺(𝑠)
𝑉𝑎 (𝑠) − 𝐸𝑏 (𝑠) 𝑅𝑎 + 𝐿𝑎 𝑆

I a ref error U (s) 1 I a (s)


 PI La S  Ra

Figure (1.5) Closed loop current controller of DC motor

Hence the closed loop transfer functions for the current control systems can be obtained as
1 𝑘𝑝 . 𝑠 + 𝑘𝑖
𝐼𝑎 (𝑠) 𝐺(𝑠)𝐷(𝑠) 𝑅𝑎 + 𝐿𝑎 𝑠 . 𝑠
= =
𝐼𝑟𝑒𝑓 (𝑠) 1 + 𝐺(𝑠)𝐷(𝑠) 1 𝑘𝑝 . 𝑠 + 𝑘𝑖
1+𝑅 +𝐿 𝑠.
𝑎 𝑎 𝑠
Simplifying it leads to
𝐼𝑎 (𝑠) 𝑘𝑝 . 𝑠 + 𝑘𝑖
=
𝐼𝑟𝑒𝑓 (𝑠) 𝐿𝑎 𝑠 2 + (𝑘𝑝 + 𝑅𝑎 )𝑠 + 𝑘𝑖
1.4 Speed Controller
Speed Controller of dc motor forms the outer loop of the mechanical part of the machine, and it
gets its feedback from which the actual motor angular speed 𝜔𝑒 is measured.

ref 
ers 1 e
 PI Js  B
e

Figure (1.6) Closed loop speed controller of DC motor

The relationship between electrical and mechanical torque is given by

𝑑𝜔𝑚
𝐽 + 𝐵𝜔𝑚 = 𝑇𝑒 − 𝑇𝑚
𝑑𝑡
Applying Laplace transform yields

Machine Drives and control, November 2012

 𝜔𝑚 (𝑠) 1
𝐺(𝑠) = =
𝑇𝑒 (𝑠) − 𝑇𝑚 (𝑠) 𝐽𝑠 + 𝐵
The closed loop transfer function of speed control is expressed as

𝜔𝑒 (𝑠) 𝐺(𝑠)𝐷(𝑠) 𝑘𝑝 . 𝑠 + 𝑘𝑖
= = 2
𝜔𝑟𝑒𝑓 (𝑠) 1 + 𝐺(𝑠)𝐷(𝑠) 𝐽𝑠 + (𝑘𝑝 + 𝐵)𝑠 + 𝑘𝑖
The PI constants (𝑘𝑝 , 𝑘𝑖 ) values can be obtained by plotting step response of the transfer
equations using trial and error values. Since this method takes time to find the true values, the
simpler method can be an automatic tuning of PID values like using simpower block.

1.5 Simulation Result

The closed loop system with PI controller of DC motor and the simulation results are shown in
figure (1.7) and (1.8) respectively.

Figure (1.7) Simulink model of DC motor drive with PI controller

Machine Drives and control, November 2012

 Figure (1.8) simulation results of characteristics of DC brush motor

DC Brush Motor parameters

La=0.012;
Ra=0.6;
J=1;
B=0;
Rf=240;
Lf=120;
Laf=1.8;
P=6;
% Ref. Speed= 1220 rpm or 1220x pi/30= 127.8 rad/sec
% Rated output Power= 5HP = 5x746= 3.730 KW
% Tm=3730/127.8= 29.2 N-m (mechanical load at rated power)
% Power input= Vdc x Ia= 240x16.2= 3.888 KW
% Te=3888/127.8= 30 N-m (electrical load)
2.1 The Mathematical model of Brushless DC Motor

The BLDC motor is an AC synchronous motor with permanent magnet mounted on the rotor and
stator windings. The equivalent circuit of BLDC motor can be modeled as 3-phase, star-
connected stator circuits as shown in the figure (2.1). The following assumptions are made in the
derivation [1][7][8][9].

(i) The 3-phase windings are symmetrical

(ii) Magnetic flux is constant and its saturation effect is neglected
(iii) Inherent resistance, inductance and mutual inductance of each winding as R, L, M
respectively. .

Machine Drives and control, November 2012

 R L
ia
Va

M Ea
M
R L
ib
Vb Eb Ec

M
R
L
ic
Vc

Fig. (2.1) The equivalent circuit of BLDCM

The 3-phase dynamic equations can be expressed in the matrix and derivative forms as

𝑑
𝑖
𝑉𝑎 𝑑𝑡 𝑎
𝑅 0 0 𝑖𝑎 𝐿−𝑀 0 0 𝑑
𝐸𝑎
[𝑉𝑏 ] = [ 0 𝑅 0 ] [ 𝑖𝑏 ] + [ 0 𝐿−𝑀 0 ] 𝑖𝑏 + [ 𝐸𝑏 ]
𝑉𝑐 0 0 𝑅 𝑖𝑐 0 0 𝐿−𝑀 𝑑𝑡 𝐸𝑐
𝑑
[ 𝑑𝑡 𝑖𝑐 ]

𝑑𝐼𝑎
𝑉𝑎 = 𝑅𝐼𝑎 + (𝐿 − 𝑀) + 𝐸𝑎
𝑑𝑡
dIb
Vb = RIb + (L − M) + Eb
dt
dIc
Vc = RIc + (L − M) + Ec
dt
where ia, ib, ic are rectangular shaped phase currents and Ea, Eb, Ec are the trapezoidal back emf
produced by each of phase windings.
. The back emf with each phase of 120° depends on the rotor angular positions as

𝐸𝑎 = 𝐾𝑒 𝜔𝑒 𝑓𝑎 (𝜃𝑒 )

𝐸𝑏 = 𝐾𝑒 𝜔𝑒 𝑓𝑏 (𝜃𝑒 )

𝐸𝑐 = 𝐾𝑒 𝜔𝑒 𝑓𝑐 (𝜃𝑒 )

And the electrical torque can be expressed as

1
𝑇𝑒 = 𝜔 (𝐸𝑎 𝐼𝑎 + 𝐸𝑏 𝐼𝑏 + 𝐸𝑐 𝐼𝑐 ) or

𝑇𝑒 = 𝐾𝑒 [𝐼𝑎 𝑓𝑎 (𝜃𝑒 ) + 𝐼𝑏 𝑓𝑏 (𝜃𝑒 )+𝐼𝑐 𝑓𝑐 (𝜃𝑒 )]

Machine Drives and control, November 2012

The relationship between electrical and mechanical system is expressed as
𝑑𝜔𝑚 𝑃
𝑇𝑒 − 𝑇𝑚 = 𝐵𝜔𝑚 + 𝐽 , 𝜃𝑒 = 2 𝜃𝑚
𝑑𝑡

Where 𝜃𝑒 𝜔𝑒 are an electrical angle and angular speed

𝜃𝑚 , 𝜔𝑚 are mechanical angle and angular speed
𝑇𝑒 , 𝑇𝑚 are the electrical and mechanical torque respectively.
𝐵 is friction constant, J- moment of inertia of rotor and coupled shaft.
The trapezoidal back emf and the torque equations can be modeled as function of rotor angle
with 60° or 120° as shown in table (2.1) [8];

Theta-electrical 𝒇𝒂 (𝜽𝒆 ) 𝒇𝒃 (𝜽𝒆 ) 𝒇𝒄 (𝜽𝒆 )

𝜽𝒆 °
𝜋 1 -1 6
0 𝑡𝑜 1 − 𝜃
3 𝜋 𝑒
𝜋 2𝜋 1 6
𝑡𝑜 −3 + 𝜃𝑒
3 3 𝜋 -1
2𝜋 6 1 -1
𝑡𝑜 𝜋 5 − 𝜃𝑒
3 𝜋
4𝜋 -1 1 6
𝜋 𝑡𝑜 −7 + 𝜃𝑒
3 𝜋
4𝜋 5𝜋 -1 6
𝑡𝑜 9 − 𝜃𝑒
3 3 𝜋 1
5𝜋 6 -1 1
𝑡𝑜 2𝜋 −11 + 𝜃𝑒
3 𝜋
Table (2.1) function of rotor angle positions

The dynamic mathematical model of BLDC motor can be developed using the basic Simulink
tools found in Matlab package as shown in figure (2.2).

Machine Drives and control, November 2012

 Figure (2.2) Simulink model of BLDC

2.2 Control System of BLDC Motor

The BLDC drive system consists of PWM control 3-phase inverter with dc source system, rotor
position sensor, current controller and speed controller. The error signal between reference speed
and actual speeds is fed to PI controller to generated reference stator current (Is). Then the stator
current (Is) as a function of rotor positions generated 3-phase reference currents. The PWM
current controller forced the actual motor currents equal to the reference current and compared
with triangular wave to provide pulses for the MOSFET power switches[6][10].
Motor currents

iabc
Vabc
iref abc
Ref. Speed Current PWM Control 3-phase BLDCM
 PI Ref. Current controller Inverter


e Position angle position

sensor
e Speed

Figure (2.3) Schematic current controlled BLDC drive system

Machine Drives and control, November 2012

2.3 Voltage Source PWM Inverter with DC Bus

The 3-phase inverter with dc source can be build using MOSFET/Diode or IGBT/Diode power
switches. Most often the MOSFET/Diode Bridge is selected as it is considered a universal power
device for low power and low voltage applications [11].

Figure (2.4) Simulink model of PWM inverter

The classical sine-triangle PWM technique is implemented to obtain the switching pulses for the
inverter. The inverter switching frequency is as same the frequency of the triangular carrier
wave. The error signals of the actual and reference 3-phase currents are compared with carrier
triangular wave. The output of the comparators produce the switching scheme used to turn
particular power switches on or off. When the error signal (Vcontrol) has magnitude higher than
the triangular wave the comparator output voltage is equal to positive half of Vdc, Otherwise the
output voltage is equal to negative half of Vdc [11].

S1 S3 S5
Vdc
2
Va Vb Vc
Vdc
2 S4 S6 S2

Figure (2.5) 3-phase inverter switching logic

Machine Drives and control, November 2012

 𝑉𝑑𝑐⁄
𝑉𝑐𝑜𝑛 > 𝑉𝑡𝑟𝑖 , 𝑆1 𝑖𝑠 𝑜𝑛 𝑉𝑎0 = 2
−𝑉𝑑𝑐⁄
𝑉𝑐𝑜𝑛 < 𝑉𝑡𝑟𝑖 , 𝑆4 𝑖𝑠 𝑜𝑛 𝑉𝑎0 = 2
𝑉
𝑉𝑐𝑜𝑛 > 𝑉𝑡𝑟𝑖 , 𝑆3 𝑖𝑠 𝑜𝑛 𝑉𝑏0 = 𝑑𝑐⁄2
−𝑉𝑑𝑐⁄
𝑉𝑐𝑜𝑛 < 𝑉𝑡𝑟𝑖 , 𝑆6 𝑖𝑠 𝑜𝑛 𝑉𝑏0 = 2
𝑉
𝑉𝑐𝑜𝑛 > 𝑉𝑡𝑟𝑖 , 𝑆5 𝑖𝑠 𝑜𝑛 𝑉𝑐0 = 𝑑𝑐⁄2
−𝑉𝑑𝑐⁄
𝑉𝑐𝑜𝑛 < 𝑉𝑡𝑟𝑖 , 𝑆2 𝑖𝑠 𝑜𝑛 𝑉𝑐0 = 2
𝑉𝑎𝑏 = 𝑉𝑎0 − 𝑉𝑏0
𝑉𝑏𝑐 = 𝑉𝑏0 − 𝑉𝑐0
𝑉𝑐𝑎 = 𝑉𝑐0 − 𝑉𝑎0

Figure (2.6) Wave forms of 3-phase sine PWM inverter

(Courtesy: lect25.ppt, Ohio state University)
𝑉𝑐𝑜𝑛
The ratio of 𝑉𝑡𝑟𝑖
= 𝑚𝑎, is called modulation index. The value of 𝑚𝑎 should be odd integer multiple of

3 in order to suppress the dc components and even harmonics presents in the output waveforms [11].
2.4 PWM Current Controller

There are two types of current controller namely linear PI- PWM current controller and non-
linear hysteresis current controller. The hysteresis control is easy to implement in Matlab
Simulink. However, it is not so easy to implement it in the DSP based hardware system due to its

Machine Drives and control, November 2012

variable switching frequency. In this drive the Sine PWM controller is implemented to obtain the
switching pulses, as it is mostly preferred in the real time simulation of the ac motor drives [13].

Figure (2.7) Simulink model of square-triangular PWM controller

2.5 Reference Current Controller

The Simulink block of reference current controller of BLDC motor is built as the function of
rotor position making rectangular wave forms as shown in table (2.2) [6].

Theta-electrical 𝒊𝒂𝒓 𝒊𝒃𝒓 𝒊𝒄𝒓

𝜽𝒆 °
𝜋 Is -Is 0
0 𝑡𝑜
3
𝜋 2𝜋 Is 0 -Is
𝑡𝑜
3 3
2𝜋 0 Is -Is
𝑡𝑜 𝜋
3
4𝜋 -Is Is 0
𝜋 𝑡𝑜
3
4𝜋 5𝜋 -Is 0 Is
𝑡𝑜
3 3
5𝜋 0 -Is Is
𝑡𝑜 2𝜋
3
Table (2.2) reference current as a function of rotor angle

Machine Drives and control, November 2012

 Figure (2.8) Simulink model of 3-phase reference currents

2.6 SPEED CONTROLLER

The Speed control loop is provided to track the actual motor speed to a desired value and then to
generate the reference currents as shown in figure (2.9) [6].

Figure (2.9) Simulink model of PI-speed controller

2.7 SIMULINK DRIVE AND RESULTS

The systematic combination of above blocks in Simulink provides the fundamental basic current
control system of brushless dc motor.

Machine Drives and control, November 2012

 Figure (2.10) Simulink model of complete BLDC drive

The model is simulated with the following motor parameters; L=0.00505; R=4.98; M=0;
Ke=0.05623; J=0.00001517; B=0; P=4, Vdc= 30V, Tm=0.1 N-m, Ref.speed= 1500 Rpm,
Switching Frequency Fs=2 KHz
The simulated characteristics of this model were validated by comparing with few of the
previous proposed models. It’s found that torque of the BLDC motor mainly depends on the
trapezoidal back emf and the rectangular currents fed by closed loop system. The torque ripples
occurs due to inaccuracy in back emf wave form and the phase commutation. The spurious
current wave forms due to PWM control which needs low pass filter. The inaccuracy in the
output curves also depends on the design of motor such slot shape, skew and the magnet material
used. However, the proposed Simulink model provides basic insight of the motor characteristics
in the real time applications.

Machine Drives and control, November 2012

Figure (2.11) Simulation results of characteristics of Brushless DC motor Drive

Machine Drives and control, November 2012

3.1 Mathematical Model of PMSM
Permanent Magnet Synchronous Machines (PMSM) are three phase AC machines that involve
the interaction of the stator flux and rotor flux are controlled by the vector controlled motor
drive. Basically there are two types of motors based on placement of rotor configuration; the
surface-mounted and inset magnets as shown figure (3.1) [16] [20] [22].

Figure (3.1) Types of PMSM

The dynamic model of the PMSM can be derived by using two-phase machine in direct and
quadrature axes with rotor frame of reference in which the d-axis is along the N S poles or along
the flux vector of the rotor, and the q-axis is at 90 degree angle with respect to the d-axis [22].
The following assumptions are made for the model development;
(i) Rotor flux is concentrated along d-axis while there is zero flux along q-axis
(ii) The induced emf is sinusoidal.
(iii)Machine core losses are negligible.
(iv) Saturation is neglected
The simplest method of transforming three phase model into a two-dimensional model is based
on the Park transformation. The transformation changes a complex non-linear ac model into a
much simpler dc signal model [14] [15].
The machine modeling in the 3-phase reference frame equations can be derived as

𝑉𝑎 𝑅 0 0 𝑖𝑎 λ
𝑑 𝑎
𝑉
[ 𝑏] = [0 𝑅 𝑖
0 ] [ 𝑏 ] + [ λ𝑏 ]
𝑑𝑡
𝑉𝑐 0 0 𝑅 𝑖𝑐 λ𝑐

Machine Drives and control, November 2012

Where 𝑉𝑎𝑏𝑐 are the machine voltages referenced to the ground, 𝑖𝑎𝑏𝑐 are the machine phase
currents, R is the stator phase resistance, and λ𝑎𝑏𝑐 are the magnetic fluxes associated with each
phase[22].
Using the park transformation, the three-phase abc system can be converted to the d-q reference
frame and equivalent circuits as shown figure (3.2)[13] [16].
R R
 id  iq

Vd Ld Vq Lq

    

r q r d

Figure (3.2) PMSM equivalent circuit from dynamic equations

The stator voltage equations are expressed as;

𝑑λ𝑑
𝑉𝑑 = 𝑅𝑖𝑑 + − 𝜔𝑟 λ𝑞
𝑑𝑡
𝑑λ𝑞
𝑉𝑞 = 𝑅𝑖𝑞 + + 𝜔𝑟 λ𝑑
𝑑𝑡
Where
𝜔𝑟 is electrical rotor speed
The stator flux linkages are given by
λ𝑑 = L𝑑 𝑖𝑑 + λ𝑎𝑓
λ𝑞 = L 𝑞 𝑖 𝑞
Substituting equations in first two equations gives;
𝑑𝑖𝑑
𝑉𝑑 = 𝑅𝑖𝑑 + L𝑑 − 𝜔𝑟 L𝑞 𝑖𝑞
𝑑𝑡
𝑑i𝑞
𝑉𝑞 = 𝑅𝑖𝑞 + L𝑞 + 𝜔𝑟 L𝑑 𝑖𝑑 + λ𝑎𝑓 𝜔𝑟
𝑑𝑡

Machine Drives and control, November 2012

3.2 The Park or dq0 Transformation
The set of non-linear dynamic equations describing the 3-phase PM motor can be transformed to
steady state dqo model using the matrix called Park transformation [T]. This transformation
results in elimination of the time varying quantities (inductances, rotor positions) and
momentarily becomes time invariant steady state dq0 quantities [22]. In other words the ac
variables were perceived as dc signal for the easy of dynamic analysis.
The q- and d-axes variables are related to abc variables as [21] [22];
𝑉𝑑𝑞0 = [T][𝑉𝑎𝑏𝑐 ]
𝑖𝑑𝑞0 = [T][𝑖𝑎𝑏𝑐 ]
Where [T] is called as the Park transformation given by
cos(𝜃) 𝑐𝑜𝑠(𝜃 − 2𝜋⁄3) 𝑐𝑜𝑠(𝜃 + 2𝜋⁄3)
2 𝑠𝑖𝑛(𝜃) 𝑠𝑖𝑛(𝜃 − 2𝜋⁄3) 𝑠𝑖𝑛(𝜃 + 2𝜋⁄3)
[T] =
3 1 1 1
[ 2 2 2 ]
Hence the park transformation of 𝑉𝑑 ,𝑉𝑞 & 𝑉0 can be expressed;

𝑐𝑜𝑠(𝜃 − 2𝜋⁄3)
cos(𝜃) 𝑐𝑜𝑠(𝜃 + 2𝜋⁄3)
𝑣𝑑 𝑣
2 −𝑠𝑖𝑛(𝜃) −𝑠𝑖𝑛(𝜃 − 2𝜋⁄ ) −𝑠𝑖𝑛(𝜃 + 2𝜋⁄ ) 𝑎
𝑣
[ 𝑞] = 3 𝑣
3 [ 𝑏]
𝑣0 3 1 1 1 𝑣𝑐
[ 2 2 2 ]
Similarly the inverse park transformation can be written as
𝑉𝑎𝑏𝑐 = [T]−1 [𝑉𝑑𝑞0 ]
𝐼𝑎𝑏𝑐 = [T]−1 [𝑖𝑑𝑞0 ]
cos(𝜃) −𝑠𝑖𝑛(𝜃) 1 𝑖
𝑖𝑎 𝑑
𝑐𝑜𝑠(𝜃 − 2𝜋⁄ ) −𝑠𝑖𝑛(𝜃 − 2𝜋⁄ ) 1 𝑖
[ 𝑖𝑏 ] = [ 3 3 ] [ 𝑞]
𝑖𝑐 2𝜋 2𝜋
𝑐𝑜𝑠(𝜃 + ⁄3) −𝑠𝑖𝑛(𝜃 + ⁄3) 1 𝑖0

The instantaneous 3-phase power output of stator is given by [22];

𝑃𝑡 = [𝑉𝑎𝑏𝑐 ]𝑡 [𝐼𝑎𝑏𝑐 ] = 𝑉𝑎 𝐼𝑎 + 𝑉𝑏 𝐼𝑏 + 𝑉𝑐 𝐼𝑐
Using the park transformation or dq0 components we can eliminate phase voltages and currents
𝑉𝑎𝑏𝑐 = [T]−1 [𝑉𝑑𝑞0 ]

Machine Drives and control, November 2012

 𝐼𝑎𝑏𝑐 = [T]−1 [𝑖𝑑𝑞0 ]
Therefore the power output may be written as
3
𝑃𝑡 = (V 𝑖 + V𝑞 𝑖𝑞 + 2𝑉0 𝑖0 )
2 𝑑𝑑
At steady state or balanced condition the zero sequence element V0=i0=0
3
𝑃𝑡 = (V 𝑖 + V𝑞 𝑖𝑞 )
2 𝑑𝑑
Substituting for V𝑑, V𝑞 the steady state power equation can be derived as
3
𝑃𝑡 = (λ 𝑖 − λ𝑞 𝑖𝑑 )𝜔𝑟
2 𝑑𝑞
And the steady state torque equation as
𝑃𝑡
𝑇𝑒 = ωm
3
𝑇𝑒 = (λ i − λq id)ω 𝜔𝑟
2 d q mech

𝑃
Since 𝜔𝑟 = 2 𝜔𝑚

Where
ωm , rotor mechanical speed
3
𝑇𝑒 =
2
(λd iq − λq id)𝑃2
Or
3 𝑃
𝑇𝑒 = (λ i +(L𝑑 −L𝑞) iq id) 2
2 af q

For surface PM motor L𝑑 = L𝑞 and for IPM L𝑑 ≠ L𝑞

The relationship between torque and speed is given by
2 𝑑𝜔𝑟
𝑇𝑒 − 𝑇𝑚 = (𝐽 + 𝐵𝜔𝑟 )
𝑃 𝑑𝑡
For dynamic simulation the state equations can be expressed;
𝑑𝑖𝑑 1
= (𝑉𝑑 − 𝑅𝑖𝑑 + 𝜔𝑟 L𝑞 iq )
𝑑𝑡 𝐿𝑑
𝑑𝑖𝑞 1
= (𝑉𝑞 − 𝑅𝑖𝑞 − 𝜔𝑟 L𝑑 id − 𝜔𝑟 λ𝑎𝑓 )
𝑑𝑡 𝐿𝑞
𝑑𝜔𝑟 𝑃 2
= (𝑇𝑒 − 𝑇𝑚 − 𝐵𝜔𝑟 )
𝑑𝑡 2𝐽 𝑃

Machine Drives and control, November 2012

 Figure (3.3) the Simulink model of PMSM
3.3 Vector Control of PMSM Drive
The control system of PMSM is similar to that of BLDC motor drive except the reference current
controller. The sinusoidal 3-phase reference current of PMSM can be developed with help of the
field oriented control (FOC) or the Vector Controlled Method of AC drives [15].
The vector controlled provides the independent control of flux and the torque of the AC drives
similar to the DC motor drive system [17] [18].
The simplest method of vector controlled of PMSM drive is the constant torque control, in which
the d-axis current is made zero. The electromagnetic torque given by the expression
3 𝑃
𝑇𝑒 = (λ i +(L𝑑 −L𝑞) iq id) 2
2 af q

When 𝑖𝑑 = 0, the electromagnetic torque is proportional to 𝑖𝑞 and it becomes equal to stator

phase current( 𝑖𝑠 = 𝑖𝑞 ) as shown in figure (3.4) [13] [14]

Machine Drives and control, November 2012

 Vq
iq  is

 r

Vd o id  0 af

 iq

Figure (3.4) the phasor diagram when 𝑖𝑑 = 0

For torque angle δ = 90° or 𝑖𝑑 = 0, 𝑖𝑞 = 𝑖𝑠
The torque becomes
𝑇
𝑇𝑒 =
3𝑃
22
(λaf is ) = kt is or 𝑖𝑠 = 𝐾𝑒 = 𝑖𝑞 ∗
𝑡

3.𝑃
Where 𝑘𝑡 = λ
2.2 af

Which is similar to the dc motor, the torque is dependent of the armature current.
Since speed is directly related to torque, and the torque is directly related to the q-axis current,
motor speed can be controlled by varying the q-axis current [13].
The 3-phase reference current can be developed using the inverse park transformation [15]
When
𝑖𝑑 ∗ = 0, 𝑖𝑞 ∗ = 𝑖𝑠 ∗

𝐼𝑟_𝑎𝑏𝑐 = [T]−1 [𝑖𝑑𝑞0 ∗ ]

𝑖𝑎 ∗ cos(𝜃) −𝑠𝑖𝑛(𝜃) 1 𝑖 ∗
𝑑
2𝜋 2𝜋
[𝑖𝑏 ∗ ] = [𝑐𝑜𝑠(𝜃 − ⁄3) −𝑠𝑖𝑛(𝜃 − ⁄3) 1] [𝑖𝑞 ∗ ]
𝑖𝑐 ∗ 𝑐𝑜𝑠(𝜃 + 2𝜋⁄3) −𝑠𝑖𝑛(𝜃 + 2𝜋⁄3) 1 𝑖0

Machine Drives and control, November 2012

 Figure (3.5) the simulink model of PMSM drive
3.4 Simulation Results

The developed model is simulated with the following motor specifications; R=1.4; Ld=0.0066;
Lq=0.0058; Yaf=0.1546; P=6; J=0.00176; B=0.00038818; Vdc= 30V, Tm=3 N-m,
Ref.speed= 1200 Rpm, Switching Frequency Fs=2 KHz
The results were compared with few of the previous proposed models and it somehow shown the
similar characteristics. At 0.025 sec a load torque of 3 Nm is applied to the motor, which causes
a slight decrease in the speed. Initially the commanded torque is equal to maximum capability of
the motor and when loaded the electrical torque increases to satisfy the desired load torque. The
motor currents and inverter output voltages were similar to that of brushless dc motor and except
back emf which is sinusoidal in shape.

Machine Drives and control, November 2012

 Figure (3.6) Simulation results of characteristics of PMSM Drive

4.1 Comparison of BLDC and PMSM motors

The simulation of BLDC and PMSM drive system does not provide clear difference between the
two motors. Some of the literature survey also reveals that there is confusion of BLDC as the DC
motor [24].
In contrast to DC brush motor, BLDC has permanent magnet poles in place of wound rotor,
electronic commutation in place of the mechanical brushes and fed with 3-phase ac supply
through inverter system. The motor drive control looks like DC motor as the phase

Machine Drives and control, November 2012

commutations logic is similar to the operation of mechanical brushes. The 2-phase commutation
at a time generates trapezoidal back emf based on the rotor position angle (60° and 120°). The
main disadvantage of BLDC is the torque ripples occurs during commutations and limits its
applications to low power ends such as refrigerators, mixers and many other home appliances
[24] [25].
The limitation of BLDC is overcome by the advent of power switches and the vector control
method, of which PMSM offers the precise control of servo system. The PMSM has permanent
magnet rotor poles and stator armatures fed with sinusoidal 3-phase ac supply through inverter
system. The field oriented control method (FOC) of PMSM provides the independent control of
torque and rotor field similar to DC brush motor. The applications of PM motors are widely
found in automobiles, aerospace and robotics. However the control drive of PMSM is
sophisticated and requires indebt knowledge of electric drive systems [24] [25].

Machine Drives and control, November 2012

REFERENCE

1. “Mathematical Modeling of DC Machines”,

http://faculty.ksu.edu.sa/fayez/Fayez%20Documents/Project%201.pdf
2. “DC_Machine.ppt”,
http://faculty.ksu.edu.sa/Rizk_Hamouda/430EE%20Doc/DC_MACHINE.ppt
3. Robert Babuska and Stefano Stramigioli, “Matlab and Simulink for Modeling and
Control”, www.dcsc.tudelft.nl/~sc4070/transp/refresher.pdf
4. “DC Motor Speed: PID Controller Design”,
http://ctms.engin.umich.edu/CTMS/index.php?example=MotorSpeed&section=ControlPID
5. Tibor, B.; Fedak, V.; Durovsky, F. “Modeling and Simulation of BLDC Motor in
MATLAB GUI”, Industrial Electronics (ISIE), 2011 IEEE International
6. Lin Bai, “Electric drive system with BLDC motor”, Electric Information and Control
Engineering (ICEICE), 2011, Page(s): 359 - 363, IEEE
7. Ji Hua, Li Zhiyong, “Simulation of sensorless permanent magnetic
brushless DC motor control system”, Automation and Logistics, 2008. ICAL 2008.
IEEE. Page(s): 2847 - 2851
8. Preetha Philip, Dr. Meenakshy, “Modeling of Brushless DC Motor Drive Using
Sensored and Sensorless Control (Back EMF zero crossing detection). www.ijetae.com
(ISSN 2250-2459, Volume-2, Issue-8, August 2012)
9. “Brushless DC Motor Control Made Easy”,
ftp://ftp.dei.polimi.it/.../Motor%20control/brushless
10. Vidhya M.P, Jaya B, A.S. shajilal, “Simulation of Brushless DC Motor”,
http://117.211.100.42:8180/jspui/bitstream/123456789/1201/1/EEEM12.pdf
11. Prof. Ali Keyhani, “Pulse-Width Modulation (PWM) Techniques”,
http://academic.udayton.edu/markpatterson/ECT459/lect25.ppt
12. Dr. Nik Rumzi Nik Idris, “Converters in Electric Drive Systems Module 2”,
http://encon.fke.utm.my/courses/mep1523/converter3.pps
13. Pragasen Pillay, R.Krishnan, “Modeling, simulation, and analysis of permanent-magnet
motor drives. I. The permanent-magnet synchronous motor drive”, Industry
Applications, 1989 IEEE Transactions on Volume:2, Issue: 2, pages: 265- 273

Machine Drives and control, November 2012

 14. ZhangHaigang, Qian Weiguo, Wu Yanxiang, Gan Shihong,and Yu Yuan, “Modeling
and Simulation of the Permanent-Magnet Synchronous Motor Drive”, IEEE, 2011,
Pages: 256-260
15. S.D.Sdhoff, “A FLUX-WEAKENING STRATEGY FOR CURRENT-REGULATED
SURFACE MOUNTED PERMANENT-MAGNET MACHEW DRIVES”, IEEE
Transactions on Energy Conversion, Vol. 10, No. 3, September 1995
16. M. E. Haque, M. F. Rahman, “Dynamic Model and Parameter Measurement of Interior
Permanent Magnet Synchronous Motor”,
http://itee.uq.edu.au/~aupec/aupec06/htdocs/content/pdf/151.pdf
17. “Implementation of a Speed Field Oriented Control of 3-phase PMSM Motor using
TMS320F240”, http://www.ti.com/lit/an/spra588/spra588.pdf
18. “Field Oriented Control of 3-Phase AC motors”,
http://www.ti.com/lit/an/bpra073/bpra073.pdf
19. “Field Oriented Control in Permanent Magnet Synchronous Motors”,
http://ebookbrowse.com/field-oriented-control-in-permanent-magnet-synchronous-
motors-arias-jcee-06-pdf-d41357667
20. “ECE 504 – EXPERIMENT 7 Permanent Magnet Synchronous Machine Drives”,
www.coursehero.com
21. “SIMULATION OF ELECTRIC MACHINE AND DRIVE SYSTEMS USING
MATLAB AND SIMULINK”, http://www.ece.umn.edu/users/riaz/macsim/info.pdf
22. SALIH BARIS OZTURK, “MODELING, SIMULATION AND ANALYSIS OF LOW-
COST DIRECT TORQUE CONTROL OF PMSM USING HALL-EFFECT
SENSORS”, master Thesis http://txspace.di.tamu.edu/bitstream/handle/1969.1/4905/etd-
tamu-2005C-ELEN-Ozturk.pdf?sequence=1
23. SIMULINK, Dynamic System Simulation for MATLAB,
www.emse.fr/~pbreuil/auto/sl_using.pdf
24. “The difference between BLDC and PMSM”,
http://www.control.com/thread/1026172925
25. Daniel Torres, “Motor Control and Power Supply Design Using BLDC, PMSM and
SMPS”,http://www.freescale.com/files/microcontrollers/doc/support_info/BeyondBits2a
rticle08.pdf

Machine Drives and control, November 2012

View publication stats