Transcript Position Sensorless Control Based on Coordinate Transformation

Position Sensorless Control Based on Coordinate
Transformation for Brushless DC Motor Drives
Yuanyuan Wu, Zhiquan Deng, Xiaolin Wang, Xing Ling, and Xin Cao;
IEEE TRANSACTIONS ON POWER ELECTRONICS,
VOL. 25, NO. 9, SEPTEMBER 2010 (2365-2371)
指導教授：王明賢
學生班級：四電資四甲
學生編號：49728017
學生姓名：林炳宏
Outline
• I.INTRODUCTION
• II.REVIEW OF THE TRADITIONAL
SENSORLESS METHOD
• III.PROPOSED ARITHMETIC FOR
SENSORLESS BLDC DRIVES
• IV.EXPERIMENTAL RESULTS
• V.CONCLUSION
• VI.REFERENCES
I.INTRODUCTION(1/3)
• Permanent magnet (PM) brushless dc
(BLDC) motor is increasingly being used in
various areas such as automotive,
computer, industrial, and household
products, etc., and its market is rapidly
growing. This is mainly due to its high
efficiency,high torque, ease of control, and
lower maintenance [1].
I.INTRODUCTION(2/3)
• The three-phase wye-connected BLDC motors are
usually supplied with a six-step 2π/3 inverter, and for
assuring the normal system operation, the rotor position
signals are necessary.
• Traditionally, a rotor position sensor such as hall element
is used for detecting the rotor position inBLDCmotor due
to its simplicity and low cost.However, the rotor position
sensors show some disadvantages from the standpoint
of total system cost, size, and reliability[2].
• Consequently, for eliminating these sensors from the
motor, many research workers have attempted to control
the speed of BLDC motors without rotor position sensors
or speed sensors [3]–[15].
I.INTRODUCTION(3/3)
• In the past, the detection of the
freewheeling diodes conduction [3], the
back electromotive force (EMF) voltage
sensing [4], [5], the third harmonic
detection [6], the flux estimation [7], and
the intelligent control [8]–[10] had been
applied to the BLDC drives.
II.REVIEW OF THE TRADITIONAL
SENSORLESS METHOD(1/7)
Fig. 1.(a) Phase-voltage method drive system.
(b) Terminal-voltage method drive system.
II.REVIEW OF THE TRADITIONAL
SENSORLESS METHOD(2/7)
• A.Traditional Sensorless-Control Methods
• The back EMF zero-crossing detection circuit is composed
of four parts, including the resistance network, the dividing
voltage circuit, the first-order low-pass filters, and the
comparators in the PVM. R2 , R3 , and R4 constitute the
resistance network which is used for simulating the motor
neutral point voltage if the neutral wire of the BLDC motor
is not available. R0 and R1 constitute the dividing voltage
circuit, which is used to reduce the terminal voltage so that
the signal sampled is between +15 and –15 V if the
comparators are supplied by the +15 V power source.
II.REVIEW OF THE TRADITIONAL
SENSORLESS METHOD(4/7)
• B. Analysis of the Rotor-Position-Detection Error
• In a PM BLDC motor, the transient value of flux linkage
in each phase winding is directly related to the rotor
position. However, the back EMF is due to the rate of
change of flux linkages, which is mainly excited by the
PM on the surface of the rotor; hence, the back EMF can
be indirectly related to the rotor position, and used for
position sensorless control.
• Nevertheless, in practical implementation, the two main
sources of error, e.g., the armature reaction field and the
low-pass filter,should be taken into account. For
implementing the proposed method,
II.REVIEW OF THE TRADITIONAL
SENSORLESS METHOD(5/7)
1) Error Caused by the Armature Reaction Field: The back EMF consists of two
components, one is the back EMF caused by the PM excitation field, which can
reflect the rotor position accurately, and the other is the reaction EMF due to the
armature reaction field, which is usually only around 10%–20% of the magnitude
of the open-circuit field [21] and may distort the motor air-gap magnetic field
excited by PM.
2) Error Caused by the Low-Pass Filters: Usually, the terminal or phase voltages
are used instead for sensorless control, as the back EMF is difficult to be
detected directly. The terminal voltage is the combination of dc and harmonic
components generated by PWM operation. The magnitude of the dc component
varies directly with the duty of PWM, and the frequencies of harmonic
components are multiples of the carrier frequency.
•
其中
；C
R0，R1 分壓電阻器
濾波電容
；ω=2πf為角頻率和F是根本頻率。
II.REVIEW OF THE TRADITIONAL
SENSORLESS METHOD(6/7)
Fig. 2. (a) Commutation instant for the normal rotation.
(b) Commutation instant for the reversing rotation.
II.REVIEW OF THE TRADITIONAL
SENSORLESS METHOD(7/7)
• C. Review of the Error Compensation Methods
• Sensorless commutation IC for BLDC motors [2] based
on the frequency-independent phase shifter [19], in
which the position error could be compensated by
decreasing the slew rate γd . In [18], the period Td was
calculated which depended strongly on the parameters
of the BLDCmotor, such as the back EMF coefficient and
the winding inductance. Another low-cost position
sensorless-control scheme [20] included a look-up-tablebased correction for the phase delay introduced by the
filter.
III.PROPOSED ARITHMETIC FOR
SENSORLESS BLDC DRIVES(1/9)
• A. Constructed Position Signals Based on Coordinate
Transformation
• Fig. 3. Coordinate transformation.
III.PROPOSED ARITHMETIC FOR
SENSORLESS BLDC DRIVES(2/9)
• To begin with, the phase relationship between Ea and
Ea and the characteristics of Ea will be analyzed. Then,
Fourier expansions for Ea , Eb , and Ec are given as (3)
where Ud is the dc bus voltage.
III.PROPOSED ARITHMETIC FOR
SENSORLESS BLDC DRIVES(3/9)
III.PROPOSED ARITHMETIC FOR
SENSORLESS BLDC DRIVES(4/9)
• The back EMF, which is trapezoidal, contains only odd
harmonics.Therefore, the phase of E_a is the same as the
fundamental phase of Ea . The advantages of the proposed
arithmetic can be summarized as follows: 1) using the angle
β, the rotorposition-detection error could be compensated; 2)
the neutral point voltage in PVM and the Ud/2 in TVM are not
needed.Because the dc component is eliminated by the
coordinate transform,the zero-crossing points of E_a , E_b ,
and E_c can be used for commutation; and 3) it can be seen
that the amplitude of E_a is 1.5 times the amplitude of the
fundamental ofEa ,which enables better estimation of the
zero-crossing points than conventional methods [11], E_a
could be regarded as fundamental wave if the higher
harmonics are neglected.
III.PROPOSED ARITHMETIC FOR
SENSORLESS BLDC DRIVES(5/9)
• Fig. 4. Proposed sensorless BLDC motor-drive system
III.PROPOSED ARITHMETIC FOR
SENSORLESS BLDC DRIVES(6/9)
• 1) Hardware Implementation: The proposed sensorless
BLDC motor drive system could be designed as shown in Fig.
4. In comparison with Fig. 1, the back EMF zero-crossing
detection circuit is substituted by the back EMF sampling
circuit,which includes the low-pass filters, the sampling circuit,
and the regulation circuit. The filters are used for eliminating
the chopped pulse caused by the PWM operation. The
sampling circuit, which is galvanically isolated from the digital
signal processor (DSP), would sample the filtered terminal
voltage to the regulation circuit, and the regulation circuit
would scale down the filtered terminal voltage between 0 and
+3 V, which could be read by the DSP.
III.PROPOSED ARITHMETIC FOR
SENSORLESS BLDC DRIVES(7/9)
• 2) Program Implementation: The program flow of the
proposed arithmetic is presented in Fig. 5(a). Since the back
EMF is not generated at standstill, a proper starting
procedure is necessary. In this paper, an open loop ramping
control scheme is adopted to accelerate the motor to a
certain speed level. Once the motor reaches the speed at
which the back EMF can be reliably detected, the control
mode will be changed from the open loop ramping control to
the proposed sensorless control. After the starting process,
the filtered terminal voltages, e.g., Ea and Eb , are sampled
in real time, and the other one Ec is constructed by Ea and
Eb . Then, the rotation speed ω can be computed out by
using traditional estimation method [2].
III.PROPOSED ARITHMETIC FOR
SENSORLESS BLDC DRIVES(8/9)
• Fig. 5. (a) Program flow diagram.
(b) Commutation logic decision diagram.
III.PROPOSED ARITHMETIC FOR
SENSORLESS BLDC DRIVES(9/9)
• The commutation logic decision is shown in Fig. 5(b). After
the commutation logic decision, the six PWM signals are
transmitted out for sensorless BLDC control by the control
unit. The polarity of the signals derived from the back EMF
would reverse when the motor rotates in the opposite
direction; hence it only needs to switch the β calculation
mode as expressed in (2).
IV.EXPERIMENTAL RESULTS(1/2)
• The proposed arithmetic for sensorless
BLDC motor drive has been successfully
implemented on the experimental BLDC
motor, and its specification is shown in
Table I. The overall system configuration
of the proposed sensorless drive is
shownin Fig. 4.
• TABLE I - MOTOR PARAMETERS
IV.EXPERIMENTAL RESULTS(2/2)
V.CONCLUSION(1/1)
• A novel arithmetic for sensorless BLDC motor drives has
been presented in this paper. The rotor-position signals,
e.g.,E_a , E_b , and E_c , are constructed by the twophase terminal voltages, which can be used for
commutation. The proposed sensorless-control method
has the following advantages: 1) it only needs two-phase
terminal voltage signals, without using the neutral
voltage signal of the motor and 2) in order to obtain the
commutation instants, the rotor-position-detection error
could be calculated in real time, and compensated by
computing the angle β, over the speed range.
Experimental results verify the analysis in Section III and
demonstrate advantages (1) and (2).
VI.REFERENCES(1/5)
• [1] T. H. Kim, H. W. Lee, and M. Ehsani, “State of the art and future
trends in position sensorless brushless DC motor/generator drives,”
in Proc. 31st Annu. Conf. IEEE Ind. Electron. Soc., 2005, p. 8.
• [2] K. Y. Cheng and Y. Y. Tzou, “Design of a sensorless
commutation IC for BLDC motors,” IEEE Trans. Power Electron., vol.
18, no. 6, pp. 1365– 1375, Nov. 2003.
• [3] S. Ogasawara and H. Akagi, “An approach to position sensorless
drive for brushlessDCmotors,” IEEE Trans. Ind. Appl., vol. 27, no. 5,
pp. 928–933, Sep. 1991.
• [4] J. X. Shen and Y. B. Liu, “Microcomputer based positionsensorless drive for brushless dc motor,” in Proc. Int. Power
Electron. Motion Control Conf., 1994, pp. 402–407.
• [5] S. J. Lee, Y. H. Yoon, and C. Y. Won, “Precise speed control of
the PM brushless DC motor for sensorless drives,” in Proc. 8th Int.
Conf. Electr. Mach. Syst., 2005, vol. 1, pp. 290–295.
VI.REFERENCES(2/5)
•
•
•
•
•
[6] J. X. Shen and S. Iwasaki, “Sensorless control of ultrahighspeed PM brushless motor using PLL and third harmonic back EMF,”
IEEE Trans. Ind. Electron., vol. 53, no. 2, pp. 421–428, Apr. 2006.
[7] Y. H. Yoon, T. W. Lee, S. H. Park, B. K. Lee, and C. Y. Won,
“New approach to rotor position detection and precision speed
control of the BLDC motor,” in Proc. 32nd Annu. Conf. IEEE Ind.
Electron., 2006, pp. 1305–1310.
[8] C. Wei and X. Changliang, “Sensorless control of brushless DC
motor based on fuzzy logic,” in Proc. 6th World Congr. Intell. Control
Autom., 2006, pp. 6298–6302.
[9] B. G. Park, T. S. Kim, J. S. Ryu, and D. S. Hyun, “Fuzzy back
EMF observer for improving performance of sensorless brushless
DC motor drive,” in Proc. 21st Annu. IEEE Appl. Power Electron.
Conf. Expo., 2006, p. 5.
[10] T. S. Kim, J. S. Ryu, and D. S. Hyun, “Unknown input observer
for a novel sensorless drive of brushless DC motors,” in Proc. 21st
Annu. IEEE Appl. Power Electron. Conf. Expo., 2006, p. 6.
VI.REFERENCES(3/5)
•
•
•
•
•
[11] J. Shao, D. Nolan,M. Teissier, and D. Swanson, “A
novelmicrocontrollerbased sensorless brushless DC (BLDC) motor
drive for automotive fuel pumps,” IEEE Trans. Ind. Appl., vol. 39, no.
6, pp. 1734–1740, Nov./Dec. 2003.
[12] J. Shao, “An improved microcontroller-based sensorless
brushless DC (BLDC) motor drive for automotive applications,” IEEE
Trans. Ind. Appl., vol. 42, no. 5, pp. 1216–1221, Sep./Oct. 2006.
[13] G. H. Jang and M. G. Kim, “Optimal commutation of a BLDC
motor by utilizing the symmetric terminal voltage,” IEEE Trans.
Magn., vol. 42, no. 10, pp. 3473–3475, Oct. 2006.
[14] T. H. Kim and M. Ehsani, “Sensorless control of the BLDC
motors from near-zero to high speeds,” IEEE Trans. Power Electron.,
vol. 19, no. 6, pp. 1635–1645, Nov. 2004.
[15] L. Zicheng, C. Shanmei, Q. Yi, and C. Kai, “A novel line-to-line
back EMF calculation for sensorless brushless DC motor drives,” in
Proc. Int. Conf. Electr. Mach. Syst., 2008, pp. 1406–1411.
VI.REFERENCES(4/5)
•
•
•
•
•
[16] T. H. Kim and M. Ehsani, “An error analysis of the sensorless
position estimation for BLDC motors,” in Proc. 38th IAS Annu.
Meeting Conf. Record Ind. Appl. Conf., 2003, vol. 1, pp. 611–617.
[17] D. M. Lee and W. C. Lee, “Analysis of relationship between
abnormal current and position detection error in sensorless
controller for interior permanent-magnet brushless DC motors,”
IEEE Trans. Magn., vol. 44, no. 8, pp. 2074–2081, Aug. 2008.
[18] J. X. Shen and K. J. Tseng, “Analyses and compensation of
rotor position detection error in sensorless PMbrushless DCmotor
drives,” IEEE Trans. Energy Convers., vol. 18, no. 1, pp. 87–93, Mar.
2003.
[19] D. H. Jung and I. J. Ha, “Low-cost sensorless control of
brushless DC motors using a frequency-independent phase shifter,”
IEEE Trans. Power Electron., vol. 15, no. 4, pp. 744–752, Jul. 2000.
[20] G. J. Su and J. W. Mckeever, “Low-cost sensorless control of
brushless DC motors with improved speed range,” IEEE Trans.
Power Electron., vol. 19, no. 2, pp. 296–302, Mar. 2004.
VI.REFERENCES(5/5)
•
[21] Z. Q. Zhu and D. Howe, “Instantaneous magnetic field
distribution in brushless permanent magnet DC motors. II. Armaturereaction field,” IEEE Trans. Magn., vol. 29, no. 1, pp. 136–142, Jan.
1993.