International Journal of Engineering and Technical Research (IJETR)

ISSN: 2321-0869 (O) 2454-4698 (P), Volume-3, Issue-8, August 2015

Research on Fractional Order Adaptive Speed

Control for Brushless DC Motor Drives
Wen Jianping, Cao Xiangang

non-integral order, which can model various real materials

Abstractin order to enhance the static and dynamic more nearly than integer order ones. This study introduces
performances and robustness of the brushless DC motor fractional order calculus into ADRC. Fractional order
(BLDCM) drives, this paper introduces a new control scheme proportional integral control replaces the nonlinear state error
for speed control. Fractional order integral and active feedback control law using nonlinear function combination in
disturbance rejection controller (ADRC) technology are applied
ADRC, which can synthesize the high performance of ADRC
to establish the speed loop controller. Fractional order integral
replaces the nonlinear state error feedback control law that uses
estimating disturbances with the characteristics of fractional
nonlinear function combination in ADRC, which can utilize the order calculus spreading the stable domain of the system
characteristics of fractional order calculus describing the parameters. The bilinear transformation and continued
physical object more nearly and spreading the stable domain of fraction expansion are used to obtain discrete fractional order
the system parameters. Simulation results verify that the integral model. Simulation results verify that the proposed
proposed control method has given good dynamic response and control scheme has given good speed response
robustness. characteristics.

Index TermsActive disturbance rejection control, brushless II. ADRC OF BLDCM DRIVES
DC motor, fractional order calculus, speed Servo.

The mathematical model of BLDCM can be obtained

I. INTRODUCTION according to the following assumption.
Assumption 1: star connected windings, no neutral line,
Assumption 2: ignoring cogging effect, homogeneous
BLDCM is increasingly used in high precise servo system
distribution windings.
and small power driving system due to its simple structures,
Assumption 3: ignoring saturation of magnetic circuit and
speed conveniently, high power density and lower
eddy-current loss and magnetic hysteresis loss, no damping
electromagnetic pollution [1-2]. These applications have
winding on rotor.
raised higher requirements with fast dynamic response,
The voltage equation of stator windings is expressed as [9]:
robustness and stability precision of MLDCM speed control.
PI control can hardly meet the requirement of high ua ia ia ea
u Z i pN i e (1)
performances although it was widely used. Some new control b b b b
strategieshave been developed and applied to the speed uc ic ic ec
control of speed servo drives to obtain high driving
where, p is differential operator, ua, ub and uc are phase
performance. Predictive function control based on ARMAX
voltages of three-phase stator windings, respectively, iaib
model and fractional order PI is used to improve the
ic are phase currents of three-phase stator windings,
performance of the BLDCM [3]. To depress influence of the
disturbances to BLDCM, the variable structure controller respectively, ea, eb and ec are back-emf three-phase stator
having an extended state observer is presented [4]. In order to windings, respectively. Parameters Z and N are denoted as,
improve the PI control, fuzzy control is combined with PI respectively:
control [5-6]. Variable structure control law and repetitive R 0 0
control law are synthesized to improve the rate smoothness of Z = 0 R 0
BLDCM driven systems [7]. Adaptive control exists complex 0 0 R
adaptive control law. Chattering phenomenon can be caused
in a conventional sliding mode control system. Active L M 0 0
disturbance rejection control technique is applied in BLDCM N = 0 LM 0
speed servo system [8], which can design the controller of 0 0 L M
controlled object by using order of model, but not becoming
dependent. It improves the response by using extended state where R is resistance of the stator windings, L is
observer to estimate and compensate disturbances. self-inductance of the stator windings, M is mutual inductance
Fractional order calculus extends integer order ones to of the stator windings.
Electromagnetic torque is given by:

Wen Jianping, School of Mechanical Engineering, Xi'an University of 2 pn em im

Science and Technology, xian, China Te (2)
Cao Xiangang, School of Mechanical Engineering, Xi'an University of r
Science and Technology, xian, China
This study is supported by the Key Laboratory of Express-way where, em is amplitude of the back-emf; im is amplitude of
Construction Machinery (Changan University), Shaanxi province, no. phase currents, r is angular velocity of the rotor; pn is
310825151135.

256 www.erpublication.org
 Research on Fractional Order Adaptive Speed Control for Brushless DC Motor Drives

number of pole pairs. iod K p ( zd1 r* ) Ki Dt ( zd1 r* ) (10)

d r 1
(Te Fr TL ) (3) To obtain discrete mathematical model of numerical
dt J control to be used, it is required that fractional order integral
operator is discretized. Using bilinear transformation method,
where, J is combined inertia of rotor and load; F is combined the fractional order system can be discretized into irrational
viscous friction of rotor and load; TL is shaft load torque. equation of z, which is then transformed in the continued
III. DESIGN OF FRACTIONAL ORDER ADRC fraction expansion way to obtain discrete model of s .
When equals to 0.4, sampling period equals 0.001 sec and
A. ADRC model of BLDCM the order of the polynomial function equals 6, the
Lets plug (2) into (3) yields approximate discretized model of fractional order integral is
obtained as follows
d r 1 2p e
dt
( n m im F r TL )
J r
Z s 0.4 nu
de
(11)
(4)
1 where nu and de are expressed as follows
=Fk wd im
J r nu = - 2.69 - 1.08 z-1 + 3.48z-2 + 1.09 z-3 - 1.09 z-4
0.21 z-5 + 0.05 z-6
TL 1 2 pn em 1 de = - 56.36 + 22.54 z-1 + 72.75 z-2 - 22.79 z-3
where, Fk = - Fr; wd im . 22.73 z-4 + 4.33 z-5 + z-6
J J r The block diagram of BLDCM drives is shown in Fig. 2.
The second-order ESO for r is constructed as follows:
ed zd 1 r
IV. SIMULATION ANALYSIS
im
zd 1 Fk zd 2 d 1 fal (ed , 0.5, ) (5) In this section, the proposed control scheme is tested by
J r means of simulations. The data of BLDCM are as follows: the
z fal (e , 0.25, ) stator inductances are 8.5 mH; the stator resistance is 0.2 ;
d2 d2 d
the flux by magnet is 0.175 Wb; the number of poles is 4; the
where, r is actual measured value; zd1 is estimate of r; zd2 is moment of inertia is 0.012 kgm2; sampling period of current
estimate of wd; d 1 and d 2 are adjustable parameters which loop is set to 2 s.
are determined by sampling time; = 0.01. To verify dynamic performance and robustness, a reference
step input of 100 rad/s is applied at t = 0.01 s. A load
PI control law can be described as: disturbance of 30Nm is used at t = 0.6 s.
iod K p ( zd1 r* ) Ki ( zd1 r* )dt (6) Fig. 2 shows speed response of BLDCM when no load is
applied. The fractional order ADRC (FO ADRC) controller
where, Kp and Ki are proportion coefficient and integral gives fast speed response and smaller overshoot.
coefficient respectively. Fig. 3 shows the speed response of BLDCM when load
The control input id for BLDCM drives is expressed as: disturbance is applied. The designed algorithm has better
id J r [iod ( Fk zd 2 )] (7) performance to resist disturbance. The amplitude of
undershoot is much lower and the time of speed adjustability
Substituting (7) into (4) yields is much lower.
r iod (8) Fig. 4 shows the phase current of motor stator when FO
ADRC controller is applied to control the BLDCM drives.
B. Fractional order PI control law Fig. 4 shows the phase current of motor stator when PI
controller is applied to control the BLDCM drives.
Fractional order calculus can be expressed in calculus
operator of fractional order and integral order [10]

d dt , R 0


Dt 1, R 0 (9)
t
a d , R 0

where a is lower bound of calculus operator; t is upper bound
of calculus operator; is order number of calculus operator;
order number of calculus equals to integral order when is
integer; R( ) represents real number; Dt indicates
fractional differentiation when R( ) is greater than zero; Dt
indicates fractional differentiation when R( ) is less than Fig. 1 The block diagram of BLDCM drives
zero.
The fractional order PI control law in ADRC can be
expressed as

257 www.erpublication.org
 International Journal of Engineering and Technical Research (IJETR)
ISSN: 2321-0869 (O) 2454-4698 (P), Volume-3, Issue-8, August 2015
V. CONCLUSIONS
This paper proposes a new speed control algorithm for
BLDCM drives. Fractional order integral is introduced to
active rejection disturbance control to replace the nonlinear
state error feedback control. The fractional order integral is
discretized into irrational equation of z by using the bilinear
transformation. Then, the irrational model is transformed in
continued fraction expansion way to get discrete model.
Simulation results show the effectiveness of the proposed
algorithm.

REFERENCES
[1] A. S. O. Al-Mashakbeh, Proportional Integral Derivative Control of
Brushless DC Motor, European Journal of Scientific Research, vol.
35, no. 2, 2009, pp. 198 - 203.
Fig. 2 speed step response in no-load
[2] F. Rodriguez, and A. Emadi, A Novel Digital Control Technique for
Brushless DC Motor Drives, IEEE Trans. on Industrial Electronics,
vol. 54, no. 5, Oct. 2007, pp. 2365 2373.
[3] W. Guo, Y. Chen, L. Zhou, and J. Ni, Fractinal-order PI Predictive
Function Control Research of the Brushless DC Motor, Journal of
Wuhan University of Technology, 2013, vol. 35, no. 4, pp. 130 - 135.
[4] C. Xia, J. Liu, W. Yu, and Z. Li, Variable structure control of
BLDCM based on extended state observer, The IEEE International
Conf. on Mechatronics and Automation, vol. 2, Aug. 2005, pp. 568
571.
[5] Z. Ji, Y. Chen, and J. Jiang, A Novel Fuzzy PI Intelligent Control
Method of BLDCM Speed Servo System, Electric Machines and
Control, vol. 7, no. 3, Sept. 2003, pp. 248 254.
[6] R. Arulmzhiyal, and R. Handiban, Design of Fuzzy PID Controller for
Brushless DC Motor, the International Conf. on Computer
Communication and Informatics, Jan. 2012, pp. 1-7.
[7] Y. Liu, B. Su, and M. Zeng, A Robust Repetitive Control Approach
for Improvement of Rate Smoothness of BLDCM Driven Systems,
Proceeding of the CSEE, vol. 25, no. 10, May 2005, pp. 144 148.
Fig. 3 speed response to load disturbances [8] Q. Dong, and Q. Li, Current Control of BLDCM Based on Fuzzy
Adaptive ADRC, The Ninth International Conf. on Hybrid Intelligent
Systems, vol. 3, Aug. 2009, pp. 12 14.
[9] B. Abdi, H. Baharami, and S. Mirtalaei, Simplified Design and
Optimization of Slotless Brushless DC Machine for Micro-satellites
Eelectro-mechanical Batteries, Journal of Electrical Engineering and
Technology, vol. 8, no. 1, Jul. 2013, pp. 124 - 129.
[10] M. Kleinz, and T. J. Osler, A Childs Garden of Fractional
Derivatives, The college mathematics journal, vol. 31, no. 2, 2000,
pp. 82-88.

Wen Jianping He received the Ph. D. degree in mechanical engineering

from Xian Jiaorong University. His research interests in the areas of the
technology on the control of drive for EV.
Cao Xiangang His research interests in the areas of equipment state
monitoring and fault diagnosis technology.

Fig. 4 the phase current curve (FO ADRC)

Fig. 5 the phase current (PI controller)

258 www.erpublication.org