Integrated Intelligent Research (IIR) International Journal of Business Intelligents

Volume: 05 Issue: 01 June 2016 Page No.45-49

ISSN: 2278-2400

Design of Sensor less Sliding-Mode BLDC motor

Speed regulator using Class of Uncertain
Takagi-Sugeno Nonlinear Systems
S. Swapna1,Joseph Henry2, K.Siddappa Naidu 3
Research scholor,Department of EEE,Veltech university, Chennai, India
1
2,3
Professor/ EEE,Veltech university, Chennai, India
Email:[email protected],[email protected], [email protected]

Abstract—Current Research Brushless dc motor (BLDC) is [1]. However, the PM BLDC motors are inherently
very popular in wide variety of applications. In comparison to electronically controlled and require rotor position information
conventional dc motor, the brushless dc motor having for proper commutation of currents in its stator windings. But
electronic commutator rather than mechanical commutator, so it is not desirable to use the Hall sensors for high temperature
it is more reliable than the dc motor. BLDC motors are applications where reliability is of utmost importance because
commonly used in high end white goods (refrigerator, washing a sensor failure may cause instability in the control system
machine, dish washer etc.), high end pump applications which [11]. Another major problem is associated with the
require higher reliability and efficiency. This paper proposed a conventional controllers that are widely used in the industry
due to its simple control structure and ease of implementation.
new method for sensor less speed control of high-Speed
But these controllers pose difficulties under the conditions of
brushless Dc motors with higher reliability of the motor. The
nonlinearity, load disturbances and parametric variations [3].
sensor less BLDC motor based on back EMF sensing and a Traditional control systems are based on mathematical models
rotor position angular detection value. The Nonlinear in which the control system described using one or more
characteristics of Sensor less BLDC motor analyzed using (T- differential equations that define the system response to its
Z) fuzzy model in MATLAB. Simulated Results shows inputs. In many cases, the mathematical model of the control
proposed control method superior than conventional process may not exist or may be too expensive in terms of
controllers. computer processing power and memory and a system based
on empirical rules may be more effective for closed loop speed
Keywords—BLDC, T-Z Fuzzy model, Backemf, nonlinear control. This paper develops to remove the drawbacks
associated with sensor control and use of traditional
I. INTRODUCTION controllers by using zero crossing point (ZCP) based on Back
Recently the increase in energy prices spurs higher demands of electromotive force (Back-EMF) sensor less control with
variable speed Permanent Magnet (PM) motors drives. Also, fuzzy logic controller. The sensor less control requires good
recent rapid proliferation of motor drives into the automobile reliability and various speed ranges with the high starting
industry, based on hybrid drives, generates a serious demand torque for BLDC motor drive system. To satisfy these
for high efficient PM motor drives. The Brushless dc motor requirements, this paper proposes an efficient sensor less
(BLDC) is a Permanent Magnet DC Synchronous motors. speed control to avoid high energy prices.
These are used in great amount of industrial sectors because
their architecture is suitable for any safety critical applications, A. principle of sensorless BLDC motor control
from a modelling perspective looks exactly like a DC motor Brushless DC motor drives have need of rotor position
and having linear relationship between current and torque, information for appropriate operation to execute phase
voltage and rpm [4,5]. The electromagnets do not move, the commutation. Position sensors are generally used to provide
permanent magnets rotate and the armature remains static, the the position information for the driver. So this type of position
position sensor and intelligent electronic controller are sensors is not used in sensor less drives. The advantage of
required to transfer current to a moving armature, which sensor less drives comprises of less hardware cost, increased
performs the same power distribution as a brushed DC motor. system reliability, decreased system size and reduced feedback
Regular uses of cheap induction motor drives, which have units. And also they are free from mechanical and
around 10% lower efficiency than adjustable PM motor drives environmental constraints [2]. Various control methods arises
for energy saving applications [10]. BLDC motor have many for sensor less drive, in which a back-EMF is the most cost
advantages over brushed DC motors and induction motors, effective method to obtain the commutation sequence in the
such as a better speed versus torque characteristics, high star wound motors and current sensing provides enough
dynamic response, high efficiency and reliability, long information to estimate with sufficient rotor position to drive
operating life (no brush erosion), noiseless operation, higher the motor with synchronous phase currents. BLDC motor
speed ranges and reduction of electromagnetic interference drives that do not require position sensors but it contains
(EMI). In addition, the ratio of delivered torque to the size of electrical dimensions are called a sensor less drive. The BLDC
the motor is higher, making it useful in application where motor provides sensor less operation based on the nature of its
space and weight are critical factors, PM motors have been excitation intrinsically suggest a low-cost way to take out rotor
widely used in automotive (hybrid vehicles), computers, position information from motor-terminal voltages. In the
household products and especially in aerospace application excitation of a 3 phase BLDC motor, apart from the phase-
commutation periods, two of the three phase windings are
45
Integrated Intelligent Research (IIR) International Journal of Business Intelligents
Volume: 05 Issue: 01 June 2016 Page No.45-49
ISSN: 2278-2400
functioning at a time and no conducting phase carries in the the integrated area of the back-EMFs shown in Figure 11 is
back-EMF as shown in Fig. 1. Since back-EMF is zero at approximately the same at all speeds. The integration starts
standstill and proportional to speed, the measured terminal when the silent phase’s back-EMF crosses zero. When the
voltage that has large signal-to-noise ratio cannot detect zero integrated value reaches a pre-defined threshold value, which
crossing at low speeds. That is the reason why in all back- corresponds to a commutation point, the phase current is
EMF-based sensor less methods the low-speed performance is commutated. If flux weakening operation is required, current
limited, and an open-loop starting strategy is required [7, 8]. advance can be achieved by changing the threshold voltage.
The integration approach is less sensitive to switching noise
B. Related Work and automatically adjusts for speed changes, but low speed
In this paper, a brief review of the conventional back EMF operation is poor due to the error accumulation and offset
detection will be given first. Then, the proposed novel back voltage problems from the integration [9]. As the back-EMF is
EMF detection will be described. Experiment results assumed to vary linearly from positive to negative (trapezoidal
demonstrate the advantages of the novel back EMF sensing back-EMF assumed), and this linear slope is assumed speed-
scheme and the sensor less system. Specially, a low cost insensitive, the threshold voltage is kept constant throughout
mixed-signal microcontroller that is the first commercial one the speed range.
dedicated for sensor less BLDC drives is developed,
integrating the detection circuit and motor control peripherals
with the standard 8-bit microcontroller core.

C. Conventional Back Emf Detection Schemes

For three-phase BLDC motor, typically, it is driven with six-

step 120 degree conducting mode. At one time instant, only Fig-1: Back EMF integration in single ZCP
two out of three phases are conducting current. For example,
when phase A and phase B conduct current, phase C is Once the integrated value reaches the threshold voltage, a reset
floating. This conducting interval lasts 60 electrical degrees, signal is asserted to zero the integrator output. To prevent the
which is called one step. A transition from one step to another integrator from starting to integrate again, the reset signal is
different step is called commutation. So totally, there are 6 kept on long enough to insure that the integrator does not start
steps in one cycle. In previous chapter, the first step is AB, until the residual current in the open phase has passed a
then to AC, to BC, to BA, to CA, to CB and then just repeats zerocrossing.
this pattern. Usually, the current is commutated in such way
that the current is in phase with the phase Back EMF to get the
optimal control and maximum torque/ampere. The
commutation time is determined by the rotor position. Since
the shape of back EMF indicates the rotor position, it is
possible to determine the commutation timing if the back EMF
is known. In Fig.3.1, the phase current is in phase with the
phase Back EMF. If the zero crossing of the phase back EMF
can be measured, we will know when to commutate the
current. As mentioned before, at one time instant, since only
two phases are conducting current, the third winding is open. Fig-2: Proposed T_S Fuzzy BLDC regulator model
This opens a window to detect the back EMF in the floating
winding The use of discrete current sensors for each motor phase will
provide complete current feedback, but the cost associated
with individual current sensors (e.g., current transformers or
D.Problem Statement Hall-effect sensors) is often prohibitive. An appealing
The phase back electromotive force (emf) zero-crossing points alternative is the use of current sensors which are integrated
(zcps) detection method [7] – [10]] is widely used in the into the power switches, such as power MOSFET’S and
position sensor less control. The zcps detection of phase IGBT’s, which are available from several device
voltages is accomplished using the motor neutral voltage in manufacturers with ratings up to several hundreds of volts and
[7].the virtual neutral point is adopted in [8] for the motor in several tens of amps. However, embedded current sensors
which the neutral point voltage cannot be easily achieved. impose their own constraints; for example, the current sensing
Nonlinear behavior should not possible conventional BACK terminal is not electrically isolated from the associated power
EMF technique. The proposed speed regulator address the device. Also, the availability of new power integrated circuits
above mentioned problems makes it possible to take more complete advantage of these
II.PROPOSED SENSORLESS T-S FUZZY BLDC SPEED sensors for the combined purposes of current regulation and
REGULATOR overcurrent protection. Finally, the back-EMF integration
approach provides significantly improved performance
A. Back-Emf Integration Method compared to the zero-crossing algorithm explained before.
In this technique, the commutation instant is determined by Instead of using the zero-crossing point of the back-EMF
integration of the silent phase’s back-EMF (that is the waveform to trigger a timer, the rectified back-EMF waveform
unexcited phase’s back-EMF). The main characteristic is that is fed to an integrator, whose output is compared to pre-set
threshold. The adoption of an integrator provides dual
46
Integrated Intelligent Research (IIR) International Journal of Business Intelligents
Volume: 05 Issue: 01 June 2016 Page No.45-49
ISSN: 2278-2400
advantages of reduced switching noise sensitivity and input variable. Thus, in our implementation, p=q and
automatic adjustment of the inverter switching instants z1 (t )  y1 (t ), , z q (t )  y q (t ) . In other words, if only one
according to changes in rotor speed
output is considered, there is only one variable in the premise
B. Sliding Mode Observer part. It can be expected that the number of fuzzy rules is
For controlling BLDC motor, it is necessary to know an significantly reduced.
absolute position of the rotor, so an absolute encoder or
resolver can be used for sensing the rotor position. But, these III. SIMULATION RESULT
position sensors are expensive and require a special The Proposed sensor less BLDC motor system with a buck
arrangement for mounting. Also, the state equation of BLDC converter is established to run the simulation in
motor is nonlinear, so it is difficult for the linear control theory MATLAB/Simulink.
to be applied and the stability of position and velocity
estimation have not been clarified. To improve the mechanical
robustness and to reduce the cost of the drive system, several
estimation techniques eliminating the encoder or resolver can
be applied [7]. Some relevant methods have been developed
using the sliding-mode observer. The SMO is widely studied
in the field of a motion control, and it can be applied to
nonlinear systems, such as BLDC motors [5]. This technique
applied to control systems encounters restrictions in practice,
due to the high voltage values of the power supply needed and
severe stress given to the static power converters. On the other Fig-3: Proposed Sensor less T-S Fuzzy BLDC Speed
hand, the sliding mode has been shown very efficient in the Regulator
state estimation due to its salient features, i.e., robustness to
parameter variations and disturbances including the Error (e) and change in error (ce) are the inputs for the fuzzy
measurement noise. The use of sliding mode in state observer controller whereas the output of the controller is change in
does not present physical restrictions relative to the duty cycle (Δdc). The error is defined as the difference
convergence condition (the estimation error moves toward between the ref speed and actual speed, the change in error is
zero) and does not subject the system to undesirable chattering defined as the difference between the present error and
.These problems can be alleviated using a binary observer with previous error and the output, the Change in duty cycle which
continuous inertial Coordinate-Operator Feedback could be either positive or negative and added with the
existing duty-cycle to determine the new duty-cycle. Fuzzy
C.T-S Fuzzy ModelTakagi and Sugeno proposed an elegant logic uses linguistic variables instead of numerical variables.
modeling method, which is often referred to as the T-S fuzzy The process of converting a numerical variable in to a
linguistic variable is called fuzzification. Fuzzy logic linguistic
model, to represent or approximate a nonlinear dynamical terms are most often expressed in the form of logical
system. In each fuzzy rule, a linear model is given to describe implications, such as If-Then rules. These rules define a range
of values known as fuzzy membership functions [7]. Fuzzy
the nonlinear system locally. The overall system is described membership functions may be in the form of a triangle, a
by fuzzily “blending” those local models. A typical IF-THEN trapezoid or a bell.
rule of T-S fuzzy models is written as

Rule i: IF
THEN

, for
Fig-4: Basic structure of FLC with BLDC motor.
(1)
where z1 (t )  z p (t ) are the premise variables, and M ij is
the corresponding fuzzy set for j= 1, …, p,
x(t )  x1 (t ), x 2 (t ), , xn (t )
T
is the state vector,
u(t )  is the input vector, A i   , B i   nm ,
m n n

q n
and Ci   are the system matrixes for Rule i,
y (t ) q is the output and r is the rule number. In Fig-5 Membership functions of fuzzy controller.
traditional fuzzy control approaches, z1 (t )  z p (t ) are the
Fig. 5 illustrates the membership function of fuzzy logic
state variables. In our study, the static output is used as the controller that used the fuzzification of two input values and
47
Integrated Intelligent Research (IIR) International Journal of Business Intelligents
Volume: 05 Issue: 01 June 2016 Page No.45-49
ISSN: 2278-2400
defuzzification output of the controller. There are seven Fig-10: Speed response of sensor less drive technique using T–
clusters in the membership functions, with seven linguistic S Fuzzy controller.
variables defined as Negative Big (NB), Negative Medium
(NM), Negative Small (NS), Zero (Z), Positive Small (PS),
Positive Medium (PM), and Positive Big (PB). Fig. 7 shows IV. CONCLUSIONS
the MATLAB simulation diagram of fuzzy logic controller. A
sliding mode rule-base, used in the fuzzy logic controller is A fuzzy logic controller (FLC) has been employed for the
given in Table I. The fuzzy inference operation is speed control of PMBLDC motor drive and analysis of results
implemented by using the 49 rules. Fig. 8 Simulation diagram of the performance of a fuzzy controller is presented. The
of proposed sensor less speed control of BLDC.Fig. 6 modeling and simulation of the complete drive system is
illustrates the membership function of fuzzy logic controller described in this paper. Effectiveness of the model is
that used the fuzzification of two input values and established by performance prediction over a wide range of
defuzzification output of the controller. There are seven operating conditions. The simulation of Fuzzy Logic
clusters in the membership functions, with seven linguistic controller, using MATLAB to control the speed of flexible
variables defined as Negative Big (NB), Negative Medium BLDC Motor, proves that the desired speed is attained with a
(NM), Negative Small (NS), Zero (Z), Positive Small (PS), shorter response time, when compared with conventional
Positive Medium (PM), and Positive Big (PB). Fig. 7 shows controllers. Sensor less Sliding-Mode BLDC MOTOR Speed
the MATLAB simulation diagram of fuzzy logic controller. regulator using Class of Uncertain Takagi-Sugeno Nonlinear
Systems controller is a superior controller and is capable of
controlling the motor drive over wide speed range in that time
of uncertainty
REFERENCES
[1] C. Sheeba Joice, S. R. Paranjothi, and V. Jawahar Senthil Kumar
“Digital Control Strategy for Four Quadrant Operation of Three Phase
BLDC Motor With Load Variations, ” IEEE Transactions On Industrial
Informatics, Vol. 9, No. 2, pp.974 – 982, May 2013.
Fig -6: Fuzzy Rule Viewer of BLDC motor [2] B. Mahesh Kumar, G. Ravi, and R. Chakrabarti “Sensorless Speed
Control of Brushless DC Motor with Fuzzy Based Estimation,” Iranian
Journal Of Electrical and Computer Engineering, Vol. 8, N0. 2, pp.119-
125, Summer-Fall, 2011.
[3] Radu Duma, Petru Dobra, Mirela Dobra and Ioan Valentin Sita “Low
Cost Embedded Solution for BLDC Motor Control,” International
conference on System Theory, Control and Computing,pp.1-6,Aug 2011.
[4] Anand Sathyan, Nikola Milivojevic, Young-Joo Lee, Mahesh
Krishnamurthy and Ali Emadi “An FPGA-Based Novel Digital PWM
Control Scheme for BLDC Motor Drives,” IEEE Transactions On
Fig-7 Surface Viewer of T-S fuzzy logic Industrial Electronics, Vol. 56, No. 8,pp.3040-3049 Aug 2009.
[5] Pooya Alaeinovin, Juri Jatskevich, “Filtering of Hall-Sensor Signals for
Improved Operation of Brushless DC Motors” IEEE Transactions On
Energy Conservation, Vol. 27, No. 2, pp.547-549,Jun 2012.
[6] Chwan-Lu Tseng, Shun-Yuan Wang, Shao-Chuan Chien, and Chaur-
Yang Chang “Development of a Self-Tuning TSK-Fuzzy Speed Control
Strategy for Switched Reluctance Motor,” IEEE Transactions on Power
Electronics, vol. 27, No. 4,pp 2141- 2151,April 2012.
[7] Shun-Chung Wang and Yi-Hwa Liu “A Modified PI-Like Fuzzy Logic
Controller for Switched Reluctance Motor Drives, IEEE Transactions on
Fig-8: Back EMF of sensor less BLDC motor Industrial Electronics, Vol. 58, No. 5, pp.1812- 1825, May 2011.
[8] V. U, S. Pola, and K. P. Vittal, “Simulation of four quadrant operation &
speed control of BLDC motor on MATLAB/SIMULINK,” in Proc.IEEE
Region 10 Conference,pp.1-6, Nov 2008.
[9] Amit Vilas Sant and K. R. Rajagopal “PM Synchronous Motor Speed
Control Using Hybrid Fuzzy-PI With Novel Switching Functions” IEEE
Transactions On Magnetics, Vol.45, N0,10,pp 4672- 4675 October
2009.
[10] Vicente Milanés, Jorge Villagrá, Jorge Godoy, and Carlos González,
Fig-9: Torque characteristics of sensor less BLDC motor “Comparing Fuzzy and Intelligent PI Controllers in Stop-and-Go
Manoeuvres” IEEE Transactions On Control Systems Technology, Vol.
20, No. 3, pp.770-778, May 2012.
[11] Yee-Pien Yang and Yi-Yuan TingKumar “Improved Angular
Displacement Estimation Based on Hall-Effect Sensors for Driving a
Brushless Permanent-Magnet Motor,” IEEE Transactions On Industrial
Electronics, Vol. 61, No. 1, pp.504-511 Jan2014.
[12] M. Surya Kalavathi, and C. Subba Rami Reddy “Performance
Evaluation of Classical and Fuzzy Logic Control Techniques for
48
Integrated Intelligent Research (IIR) International Journal of Business Intelligents
Volume: 05 Issue: 01 June 2016 Page No.45-49
ISSN: 2278-2400
Brushless DC Motor Drive” IEEE Transactions On Industrial
Electronics, Vol. 61, No. 1 , pp .488-491, Jul 2012.
[13] Xiang Wang, Mei Li “Rotor Position Simulation of Switched Reluctance
Motor Based on Fuzzy Inference Rules,” International Conference on
Innovation Management, pp.75-78, Sep 2009.
[14] Chang-Han Jou, Jian-Shiun Lu, and Mei-Yung Chen “Adaptive Fuzzy
Controller for a Precision Positioner Using Electro-Magnetic Actuator,”
International Journal of Fuzzy Systems, Vol. 14, No. 1,pp.110-117,
March 2012.
[15] Han Ho Choi and Jin-Woo Jung, “Discrete-Time Fuzzy Speed Regulator
Design for PM Synchronous Motor” IEEE Transactions On Industrial
Electronics, Vol. 60, No. 2,pp.600-607, Feb 2013.
[16] N.T.-T. Vu, H.H. Choi, R.-Y. Kim, J.-W. Jung “Robust speed control
method for permanent Magnet synchronous motor,” IET Electric Power
Applications, vol.6, No.7, pp.399 -411, Feb 2012.
[17] Timothy J. Ross, Fuzzy Logic with Engineering Applications, 2nd ed,
England: John Wiley & Sons Ltd, 2004.

49