Procedia Technology 4 ( 2012 ) 666 – 670

2212-0173 © 2012 Published by Elsevier Ltd.doi: 10.1016/j.protcy.2012.05.107

C3IT-2012

Simulation Modeling of BLDC motor drive and harmonic analysis of stator current, rotor speed and acceleration using

Discrete Wavelet Transform Technique

Madhurima Chattopadhyaya, Priyanka Royb , Sharmi Duttaa

aHeritage Institute of Technology, Kolkata- 107 bNetaji Subhash Engineering College, Kolkata- 152

Abstract: A simulation model has been developed in this paper to study the behavioral characteristic of Brushless DC motor and also analyses the harmonics present in the stator current, rotor speed and acceleration of the BLDC drive circuit. Hence in order to improve the accuracy of the motor drive control system, we have introduced a de-noising module in the feedback path. This denoised signal is compared with the original one and the deviation (error) between these two signals is further rectified by the PI controller present in the forward path. All these analyses are achieved by using discrete wavelet transform (DWT). For simulation purpose, we have used MATLAB/SIMULINK based platform and developed the models of those motors in SIMULINK from mathematical expressions derived from working principles of motor. The purpose of the PI controller is to achieve good tracking to torque (i.e. angular acceleration). The acceleration is decomposed and denoised further by choosing Coif let as mother wavelet. The study of the denoised signal is discussed in detail.

© 2011 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of C3IT

Keywords: PMBLDCM; Trapezoidal Commutation; Brushless DC Motor; DWT.

1. Introduction

Nowadays a trend to use Brushless DC motor instead of brushed DC motor in an increasing number of applications. This type of motor is categorized as of sinusoidal or trapezoidal depending on the rotational voltage (back EMF) induced. This paper describes mathematical model of a three phase Brushless permanent magnet DC motor (PMBLDCM) drive using SIMULINK/ MATLAB. Many researchers are showing interests for modeling and simulation of BLDC motor because of its novelty [1]. BLDC motors have a few advantages over conventional brushed DC motors and induction motors. These advantages are i) better speed versus torque characteristics, ii) high dynamic response, iii) high efficiency, long operating life, iv) noiseless operation and v) higher speed range[7]. In addition, BLDC motors are reliable, easy to control, and inexpensive [1].The BLDC motor is a type of synchronous motor because the magnetic field generated by the stator and the magnetic field generated by the rotor rotates at the same frequency. BLDC motors do not experience the “slip” that is normally seen in induction motors [2]. In the DC commutator motor, the current polarity is altered by the commutator and brushes. However, in the brushless DC motor, polarity reversal is performed by power transistors switching in synchronization with the rotor position. Therefore, BLDC motors often incorporate either internal or external position sensors to sense the actual

Available online at www.sciencedirect.com

667 Madhurima Chattopadhyay et al. / Procedia Technology 4 ( 2012 ) 666 – 670

rotor position, or the position can be detected [3]. The motor has trapezoidal back EMF with square wave currents to generate the constant torque. A conventional BLDC motor drive is generally implemented via a six-switch three phase inverter and three Hall-effect position sensors that provide six commutation points for each electrical cycle [4]. The BLDC motor drive with trapezoidal back-EMF and rectangular stator currents, show a very high mechanical power density [5, 6]. DWT is a useful tool to detect the disturbances of time varying signals. As it was pointed out above, induced EMF has irregular shape in commutation region and input current would carry the hidden properties of back EMF and other signatures of rotor position. The wavelet theory is applied here to detect these signatures and deduce the denoised rotor speed as well as acceleration. The number of peak appearing in wavelet is the unique feature for transient classification [8]. The amplitude of the peak indicate how rapid the transient magnitude changes. It can be used to determine the rise rate of the transient.

2. Simulation and modeling of the BLDC motor drive control

In process of design the drive control circuit of BLDCM, we have built the model of BLDC motor in SIMULINK shown in Figure 1 and BLDC motor subsystem is shown in Figure2. This will help in further modeling of BLDC motor drive

Figure 1: Model of BLDC motor in SIMULINK Figure 2: BLDC motor subsystem

In figure 3 the BLDC motor drive circuit with closed loop control system is presented with the help of SIMULINK blocks. The complete system is derived a six step inverter fed BLDC motor where two control loops are used. The inner loop synchronizes the pulses of the bridge with electromotive force and outer loop regulates the motor’s speed. The PI controller is used to get a steady output.

668 Madhurima Chattopadhyay et al. / Procedia Technology 4 ( 2012 ) 666 – 670

Figure 3. BLDC motor control drive Figure 4. Decoder circuit

3. Harmonics Detection and Analysis

We have studied stator current harmonics using Discrete Wavelet Transform by choosing mother wavelet Coif let 5 at level 1. The decomposed signal along with details and approximation coefficients are shown in Figure 5. The Figure shows that the approximations indicate the general trend and the low frequency characteristics of the original signal while the details show the high-frequency contents of the original signal.

Figure 5: Harmonics analysis by using Coif let 5 at level 1

3.1. Scheme for improving accuracy in drive control system

Now we have introduced a velocity (speed) feedback in the control loop of the BLDC motor. The de-noising scheme of wavelet transform has been used for minimising the error between the reference and the output rotor speed. This angular velocity of the rotor is derived from numerical differentiation of angular position obtained by hall sensors. This data corresponding to angular velocity is shown in Figure 6. It has been decomposed at level 8 and the best observable peaks are available from details coefficient at level 7, which are given in figure 7 and 8 respectively.

669 Madhurima Chattopadhyay et al. / Procedia Technology 4 ( 2012 ) 666 – 670

Figure 6. Closed loop control for acceleration using velocity feedback

Figure 7. Harmonics analysis based on Coif let Figure 8. Detail co-efficient at level 1 (d1)

The angular velocity is still retaining some noise which can be easily studied from Figure 9. The main purpose of de-noising is to remove the noise from this signal before it is fed to the PI controller. In Figure 10 the actual (red) are filter signal (blue) are very close to each other at this scale.

Figure 9. Denoised (blue) and actual signal (red). Figure 10. Zoomed denoised angular velocity

The PI controller in Fig. 6, is instructed by the torque (i.e. angular acceleration) developed in rotor. For studying the torque more precisely, the angular acceleration is derived from the velocity by numerical differentiation. The acceleration is decomposed further by using Coif let at level 8 shown in Figure 11. Again we are identified the unwanted signal just after de-noising and shown in Fig. 12. Here the noises are very prominent instead of zooming action.

670 Madhurima Chattopadhyay et al. / Procedia Technology 4 ( 2012 ) 666 – 670

Fig 11. Detail coefficient at level 7 using Coif let Fig. 12 Denoised(blue) and actual signal(red).

4. Conclusions

References

1. A. Tashakori, M. Ektesabi and N. Hosseinzadeh, “Characteristic of suitable drive train for electric vehicle”, 2010 3rd International Conference on Power Electronic and Intelligent Transportation System (PEITS), 20-21 Nov 2010, Shenzhen, China.

2. Y. S. Jeon, H. S. Mok, G. H. Choe, D. K. Kim and J. S. Ryu, “A new simulation model of BLDC motor with real back EMF waveform”, IEEE CNF. On Computer and Power Electronics, 2000. COMPEL

3. Mehnet Cunkas, Omar Aydogdu,"Realization of fuzzy logic controlled BLDC motor Drives using SIMULINK".vol-15,no-2,pp-218-229,2010.

4. Padmaraja Yedamale, Microchip Technology Inc,"Brushless Dc motor Fundamental"2003. 5. Leonerd N.Elevich,"3 phase BLDC motor control with hall sensor using 56800/E digital signal controlling"2005. 6. S.Prakash and R.Dhanesekaram,"Modelling and Simulation of closed loop controller Buck converter fed PMBLDC Drive

System" 3(4): 284-289, 2011. 7. A Taskori et al,"Modelling of BLDC motor with ideal back emf for auto mative application". Vol II ,WCE 2011, July 6 - 8,

2011, London, U.K. 8. K. Kim and A. G. Parlos, “Induction motor fault diagnosis based on neuropredictors and wavelet signal processing,”

EEE/ASME Trans. Mechatronics, vol. 7, no. 2, 2002, pp. 201-219.