PMSM Control System Based on Digital Signal ... - Semantic Scholar
924
JOURNAL OF NETWORKS, VOL. 8, NO. 4, APRIL 2013
PMSM Control System Based on Digital Signal Processor Zhu Jun, Li Wankui, Han Lili
Henan Polytechnic University, College of Electrical Engineering and Automation, Henan Jiaozuo, China Email: [email protected]
through experimental results demonstrate the superiority of the system.
Abstract—For the high power density of PMSM, designed the corresponding drive controller to improve the servo efficiency of PMSM servo system. According to the theory of math model of PMSM under the dq coordinate system, applied id = 0 vector control method as a PMSM control strategy, established PMSM controller model based on vector control. Taking DSP TMS320F2812 as controller core, built a power-driven circuit, control circuit and the main detection protection circuit. The algorithm of control program was completed on the hardware platform to present its software processes. The simulation results show that: the control system response is fast, can track the given speed and position quickly and accurately. The speed fluctuation, overshoot and steady state error are very small. The designed controller is reasonable, which has better dynamic and static characteristics, and be benefit to improve the efficiency of PMSM servo system.
A.The Mathematical Model of the PMSM By the assumption of the desired motor, can be deduced from the mathematical model of PMSM in dq coordinate system [6]: Stator flux equation: ψ d = Ld id + ψ f (1)
Index Terms—PMSM, Overall Control system, DSP,
d -q axis stator inductance; ψ f
Controller
I. INTRODUCTION With the rapid development of permanent magnet materials, power electronics and control theory, based on vector control of PMSM its excellent control performance, high power density and high efficiency, more and more used in a variety of high-performance servo systems and other fields of industrial production [1-2]. In recent years, the development of various control method [3-4] and vector control theory and methods become mature gradually, the progress of integrated circuits and computer technology, the PMSM vector control system has been a great development. The vector control of PMSM based on the mathematical model in the dq coordinate system, through conversion of vector model; realize that it has completely decoupled control of stator current. It has like a DC motor control performance, and thus more widely applied [5]. This paper introduces the mathematical model of PMSM in the dq coordinate system, modeling and analysis of PMSM vector control strategies based on the id = 0 , for DSP TMS320F2812 as controller core, built a power-driven circuit, control circuit, feedback circuit and the main auxiliary circuit. The algorithm of control program was completed based the hardware platform, and presented its software processes. Constitute a high-performance PMSM vector control system, and © 2013 ACADEMY PUBLISHER doi:10.4304/jnw.8.4.924-931
II.SYSTEM MODEL
ψ q =Lq iq
(2)
In the formula, id , iq is the d -q axis stator current;
ψ d , ψ q is the d -q axis stator flux; Ld , Lq is the is the magnetic
potential generated by the permanent magnets on the rotor. Stator voltage equation: u d = Rs id +
u q = Rs iq +
dψ d dt
dψ q dt
− ω rψ q
(3)
− ωrψ d
(4)
In the formula, ud , uq is the d -q axis stator voltage;
ωr is the rotor angular velocity; Rs is the stator
resistance. Electromagnetic torque equation:
3 3 pn (ψ d iq -ψ q id )= pn [ψ f iq -(Lq -Ld )id iq ] (5) 2 2 In the formula, Te is the electromagnetic torque; Pn is Te =
the number of pole pairs. Equations of motion: d ωr J =Te -Bωr -TL (6) dt In the formula, J is a moment of inertia( kg ⋅ m 2 );
TL is the load torque( N ⋅ m ), Te is the output torque( N ⋅ m ); B is the viscous friction coefficient. B. The Basic Principle of PMSM Vector Control In the field oriented coordinates, PMSM vector control
JOURNAL OF NETWORKS, VOL. 8, NO. 4, APRIL 2013
925
simulates the DC motor torque control law, the current vector is decomposed into the excitation current component to produce magnetic flux and the generated torque of the torque current component, the two current components are adjusted respectively, so that decoupling control is realized to control torque component and flux component of PMSM respectively [7-9]. According to the same principle of magnetic potential and power, the coordinate transformation is made to three-phase voltage, current and flux for PMSM. Stationary coordinate system of the three-phase ABC transforms into dq 0 rotating coordinate system, stator current vector is decomposed into the rotor field oriented two mutually orthogonal current component, they are excitation component id and torque component iq of stator current [6]. Set id = 0 ,the component iq is controlled,
System hardware block diagram is shown in Figure 2, the 220V AC is rectified, and filter capacitor filter to obtain a smooth DC output, and finally the alternating current supply was obtained for PMSM through the inverter circuit to convert.
PC
IO Interface Circuits
Keyboard and Display Circuit
CI S
O IP G
NT I DP P
S M T 0 23 21 F 82
MW P
no toip tO al o-s i
Detection and Protection Circuit
Rectifier Circuit
IPM DC A
PI S EP Q
Current Detection Speed and Position Detection
PMSM
it is equivalent to control the torque directly. Electromagnetic torque Te and iq are linear relationship,
Figure 2. PMSM Control System of the Block Diagram
so that the system has good mechanical properties and dynamic performance. iq Regulating reference volume
The stator phase current signal detected by hall current sensor is sent to DSP through the ADC module, it constitutes the current closed-loop control system. The rotor speed and position signal detected by optical encoder are sent to DSP through the QEP module, it constitutes the speed and position control loop. In the DSP, corresponding op conversion is made applying the soft program for the detected signal, SVPWM pulse drives signal is generated which is required in the control. It drives the IPM intelligent power module through opto-isolation circuit, controls the turn-on and turn-off time of IGBT and generates the corresponding voltage signal which controls the motor running. When the system detects fault signal, it is transmit to the DSP power protection input interrupt pins (PDPITNT) through the opto-isolation circuit, so that the power protection interrupt signal is generated to turn off the 6-channel PWM signal output pulse signal achieving the protection of system failure. In addition, the DA converter, PC connection and other external auxiliary circuits are realized through the DSP's SPI, SCI, and GIPO interface.
is given by the speed controller, the d -q axis voltage components are output after current loop adjustment, which are ud , uq . The component uα and uβ are obtained through anti-Parke transform for ud and uq in the α − β coordinate system. According to the value size of uα and uβ and SVPWM space vector method, the output of the vector control is obtained to achieve the purpose of vector Control. It is id = 0 vector control block diagram as shown in Figure 1. ∆θr
θ r* θr
Position Controller
ω r* ∆ωr ωr
Speed Controller
iqref
PI
-i idref= 0
PI
Speed
AntiUα Park SVPWM Transforgenerator mation
Ud
Uβ
q
id-
Position
Uq
Park Transformation
θr 1/s
Inverter iα
ia
Clark Transforiβ mation
ib ic
θr θr
ωr
Optical encoder
Figure 1. PMSM Vector Control Block Diagram
III. SYSTEM DESIGN A. The Hardware Structure of the Controller The controller system includes a main power circuit, control circuit and the motor. The control loop is mainly composed by the master DSP chip and sampling detection circuit. The most advanced chip TMS320F2812 DSP which frequency is 150MHz was used as core controller. The control chip is used widely in the high precision servo control, variable frequency power supply and other areas, while it is the best choice of motor and [10] digital control .
© 2013 ACADEMY PUBLISHER
B. Main Power Circuit Design The system's main power circuit consists of a rectifier circuit, filter circuit and the IPM module. The rectifier circuit part uses single-phase uncontrolled rectifier module GBJ25M. AC is rectified DC, the maximum
311V , GBJ25M rectifier modules voltage is 2 × 220 = are fully able to meet the requirements. The pressure of the filter capacitor is at least as 375V, so the paper selected two 330 µ f / 450V capacitors in parallel with a decrease of the ESR of the capacitor, the pressure value is 450V , the capacitance is 660 µ f . This paper inverter circuit part adopts IPM PS21964, it integrates high-voltage power transistor drive circuit in their own internal, and owns built-in overvoltage, over-current and overheating fault detection circuit, so that it can ensure the safe operation and reliable operation for the controller. Its rated voltage is 600V, rated current is 15A for the
926
JOURNAL OF NETWORKS, VOL. 8, NO. 4, APRIL 2013
PS21964, and the inverter output power is:
detection circuit is shown as in Figure 4.
P = 3V I cos ϕ out out out
+5
+5
(7)
+3.3 +3.3 +3.3
In the formula, Vout is the output voltage rams value; I out is the output current rms value; cos ϕ is the power factor. When inverter operates, taking into account the requirements of motor overload, inverter takes power rating of 1.5 times margin, the power factor is 0.7[11], so the maximum output power of the inverter is: (8) P = 3 × 220 × 15 × 0.7 ÷ 1.5 = 2.6 KW max It has the ability to drive 1KW motor. C. Phase Current Detection Circuit The control loop in order to achieve the current loop, speed loop and position loop three closed-loop control, DSP will detect the current signal, speed signal, and location of the signal after the program operation processing, Event Manager EVA and EVB produce 6-channel PWM signal to control the inverter turn-on and turn-off. PMSM vector control system with id = 0 control strategy must be timely detection the size of the stator phase current value. In this paper, TKC-10P takes as closed-loop hall current sensor. Motor using a Y-shaped connection can only detect two phase currents, shown in Figure 3, in order to achieve the current closed-loop control.
R2
U
TKC-10P
2 1 4 3
R3
R6 C2
+5
6N137
R5 R1 +15V
74HC14
74HC14 MM74HC14
Figure 4. Rotor Position Detection Circuit
According to U,V,W edge of the trigger signal to determine the initial position of the rotor in which the motor is started. Encoder output U,V,W signal after 74HC14, output for the three-way interaction difference of 120 electrical angle, the width of 180 electrical angle of the square wave signal, then after the isolation of high-speed optocoupler6N137,shaping the signal through the Schmitt trigger MM74HC14, to ensure the accuracy of position detection. Finally, through the QEP module is input to TMS320F2812. E. Speed Detection Circuit Optical encoder output is six-way differential signals A ±,B ±,Z ±, its output voltage range is from 0 to 5v, the signal need to be converted by using the differential receiver. Rotor speed detection circuit is shown as in Figure 5. +3.3
+5
C1 - LM358
R8 -
R3
+15
R4
+
R6 LM358
R2
+3.3
R2
+
R7
BAV99
A+
R1
+3.3 +3.3
ADCINAI A-
BAV99
+
R6
R4
C2
R7 C5
QEP
R5
C1 R3
+ LM358
QEP
R4
C1
+5
-15
BAV99
R5
R1
-15V Uo 6 OUT -V GND O/P Ui 5 +V IN
+3.3
C3
C4
6N137
74HC14
DS3486
Figure 3. Current Sensing Circuit
Figure 5. Speed Detection Circuit
The system utilizes TMS320F2812 has 12 high-resolution of the AD converter, to improve the accuracy and timeliness of the current sampling. DSP's ADC input voltage range is 0 to 3.3V, the hall current sensor output current is a bipolar signal, it needs convert the detected current signal into the voltage signal through converter circuit, so that the DSP can identify the signal. The current sensor detects the current signal of the first through the RC filter, and then through a voltage follower, then is made DC bias through the op amp circuit, finally a proportion of operations is made, the output signal range from 0 to 3.3v is obtained. Bias voltage of +5V, the op amp circuit uses a high-speed dual op amp chip LM358. Use switching diode BAV99 as the voltage clamp circuit to prevent voltage exceeds +3.3V.
In this paper, the encoder pulse signals are converted into a single output signal by DS3486, and then isolated by the opto-coupler 6N137, shaping the signal through the Schmitt triggers 74HC14. Finally, the signals are input to TMS320F2812 through the QEP module.
D. Position Detection Circuit Position detection is achieved by using the optical encoder, the article selected the hybrid optical encoder, its single ring pulse number is 2000.Signal output adopts differential driver, It can reduce the common mode interference of the signal transmission. The rotor position © 2013 ACADEMY PUBLISHER
F .The Design of Limiting the Current Starting Circuit The auxiliary circuit of the system by limiting the current starting circuit, DC bus over-current protection circuit, voltage detection circuit, braking circuit, power circuit, chip voltage monitor circuit, the serial communication interface circuit and its peripheral circuits. Limit the current starting circuit in order to prevent the starting current is too large, while damage to the bridge rectifier, during the motor starter. The main circuit will produce a very high current when power is connected, at this point relays no action, +5 V supply is not turned on, current through the thermostat flow into the filter capacitor, thermostat to play the role of limit the current. Delay in seconds, normal both ends of the filter capacitor
JOURNAL OF NETWORKS, VOL. 8, NO. 4, APRIL 2013
927
voltage, DC bus current is reduced, at this time+5 V power supply connected, the normally open relay contact pull, the thermostat short-circuit, end of the limit the current starting. Limit the current starting circuit shown in Figure 6.
R1 +5V C1
+3.3V
C4
C3 DCP
D1
4
R2
3
TLV2374ID +
+5V
2
K
ADC
1
-
R5
11
R3
D2 A
R4
D1 PTC
C2
Figure 8. Voltage Detection Circuit Diagrams
I. Dynamic Braking Circuit
G.DC Bus Over-current Protection Circuit When the DC bus current is too large, IPM module can easily be burned, therefore designed a DC bus over-current protection circuit, Shown in Figure 6. In the filter circuit and IPM modules in series one 0.02Ω sampling resistor, the over-current protection circuit to monitor the pressure drop across the sampling resistor. Over-current detection signal is active low, when the DC bus current is normal, the transistor Q5 is turned off, the comparator LM2903 inverting input is grounded, +15 V power input access the same phase, LM2903 output invalid high signal; When the DC bus over-current, transistor Q5 turns on, comparator LM2903 output low flow signal through opto-coupler to PDPINT pins. Immediately cut off the PWM signal output and LED alarm instructions after DSP receive over-current signal. +5V
When the servo driver brakes frequently with a large inertia load, servo motor will in the power generation state, but rectifier part of the system is uncontrollable rectifier circuit, the energy can’t be fed back into the grid, DC bus voltage of the system will rapidly increase. It is easier to be damaged for the energy storage capacitors, power module, so the dynamic braking circuit must be added for the servo drive system. The TMS320F2812 chip GPIOA pins sends out the brake signal, first through opto-isolation, and then through diode NPVIDT2227 to be amplified 15V level signal, it drives the power transistor Q4 to be turn on, the power is consumed in the resistor R10. When the DC bus voltage is less than 360v, the driven power transistor Q4 is turn off, the braking process is ended The brake circuit is shown as in Figure 9. +3.3V
C2
C3
C4 R4
R5
C3
C4
R2 2 3 NC
GPIOA
R7 +
1
D1 DCP
R6
C1
LM2903
D1
5 3
Q2
1
MMDT2227
PDPINT
2
-
4
R3
4N25
R4 R2
4
R5
2
2
R3 Q1
DCP
+5V 6
5
R1
+3.3V
Q1
+5V
6
1 C1
3 3
Figure 6. Limit the Current Starting Circuit
Figure 9. Dynamic Braking Circuit
C2
R1
J. Power Management Circuit Figure 7. DC Bus Over-current Protection Circuit
H. Voltage Detection Circuit The DC bus voltage detection circuit uses a closed-loop hall voltage sensor, the detected voltage signal is connected to voltage follower circuit through the TLV2374 of precision op amp. It is sent into DSP chip by the A/D converter. When the bus voltage is larger than 380V, the controller sends a brake signal. When the bus voltage is less than 120V, DSP takes as under voltage state, its principle is same as over voltage. The voltage detection circuit is shown as in Figure 8.
+5V
C1
C2
1 2 3 4 5 6 7 8 9 10 11 12 13 14
TPS767D301
28 27 26 25 24 23 22 21 20 19 18 17 16 15
DSP_RST1
R1 R2
SENSE
1.8V DSP_RST2
C3
C4
C5
C6
3.3V
Figure 10. Power Management Circuit
The circuit voltage is 3.3V for TMS320F2812 peripheral interface, its core voltage is 1.8V. In the Paper, power management chip TPS767D301is selected, output © 2013 ACADEMY PUBLISHER
928
JOURNAL OF NETWORKS, VOL. 8, NO. 4, APRIL 2013
two-way power are 1.8V and 3.3V, so that the DSP chip can normal work. DSP has ADC module, the voltage output must distinguish the digital voltage and analog voltage, so the digital ground and analog ground are connected by beads. The circuit is shown as in Figure 10. K. Serial Communication Interface Circuit TMS320F2812 chip has a serial communication interface (SCI) module, the serial port is configured as RS232 port to connect with the PC easy. Because the DSP level and RS232 level cannot match, a external circuit must be constituted by using power level conversion chip MAX232D to meet the communication functions between the control panel and PC. MAX232D uses a 3.3V supply voltage, its external four small capacitor rise the voltage from 11V to 15V. The serial communication interface circuit is shown as in Figure 11.
after the stage circuit electrical isolation, but also has a level conversion function. The IPM module fault detection circuit diagram is shown in Figure 12. M. Temperature detection protection circuit IPM module PS21964 controller uses normal working temperature does not exceed 100, the heat sink is not only to the design of the controller, but also the design temperature detection and protection circuit, to prevent overheating burned chip and other circuit. Paper, the temperature change of the temperature sensor T255 detecting controller, the temperature sensor T255 is reflected by the change in resistance to temperature change, the higher the temperature, the smaller the resistance of the temperature sensor T255.The typical o
value of the temperature sensor T255: R(25 C)=5.0KΩ .
3.3V 16 C1
C2
+3.3V
+3.3V 1
4
2
3
C5
C3 3 4
2 V+ 6 VC4 GND
fault
JP1 1 6 2 7 3 8 4 9 5
5
11 T1IN 12 R1OUT 10 T2IN 9 R2OUT
R2 1K
T255 C6
15 SCITXDA SCIRXDA SCITXDB SCIRXDB
+5V
1
VCC
14 T1OUT 13 R1IN 7 T2OUT 8 R2IN
R1 1.2K
PC181
PDPINT
R3 5.1K
Figure 13. Temperature Detection Protection Circuit
MAX232
N. The Software Structure of the Controller
Figure 11. Communication Interface Circuit
Start
L.IPM module fault detection circuit
The initialization of the system clock
+5V
+3.3V
4
1 R1 10K FO
R2 1K Q1 NPN
The timer is initialized
Initialize the system parameters
fault PDPINT
2
R3 330
PC181
3 The software module initialization
R4 5.1K
Figure 12. Temperature Detection Protection Circuit
This article uses the IPM module PS21964, internal integrated overvoltage, overcurrent, overheating protection function, IPM modules work properly, the PS21964 No.14 pin fault detection signal output terminal of the high level of +5 V,NPN type transistor C9013 and opto-coupler PC181 in the conduction state, the opto-coupler output of +3.3V high. When the IPM module fails, the fault detection signal output terminal output current of 10mA, low pulse width of 1.8ms signal, the NPN type transistor C9013 are in the OFF state and opto-coupler PC181 output is low, that is, a fault signal is active low. The opto-coupler used not only before and
© 2013 ACADEMY PUBLISHER
Timer overflow interrupt and capture interrupt enable
To detect the position of the rotor initialization
Y
Rotor initial position detection subroutine
N Interrupt latency
Y
Interrupt service routine
N
Figure 14. Main Program Flow Chart
A complete software system should include function subroutine, system initialization procedure, the timer main interrupt program, position detection interrupt program and protection interrupt program. Software
JOURNAL OF NETWORKS, VOL. 8, NO. 4, APRIL 2013
is set to 400rad, the simulation time is set to 0.5s.The rotating speed response waveform is shown in Figure 16,can see motor no-load can quickly reach the maximum speed, after slight fluctuations stabilized rated speed in 1000rad/s, when 0.1s load 4 N ⋅ m , speed fluctuate somewhat, but soon was stable in 1000rad/s. When arrived at a set position 400rad, motor speed decreased rapidly, while unload 4 N ⋅ m load, during descending process the speed is corresponding rapidly , with slight fluctuations, finally speed stabilized in 0rad/s and parking at 0.5s. 1200 1200
1000 1000
800800 n(rad/s)
mainly achieve SVPWM control algorithm, the coordinate transformation, digitized PI regulator and current / position signal detection, they are all implemented in the main interrupt program [12]. The first three do not need to control the hardware, the rest requires software and hardware properly fit in order to achieve accurate control effect. The introduction of interference in the signal detection will directly affect the control performance of the system. After control system start plus power, the first need to do is the initialization of system input and output module classes and data processing module class, and system resources, and then is the detection of the initial position angle of the PMSM. After get the initial position angle information of the rotor and then carry on backstage processing program, and wait for interrupt handling signal is used to access the system subroutine in order to control PMSM [13]. The system main program flow chart is shown in Figure 14.
929
600600
400400
IV . SIMULATION RESULTS
200200
Matlab/simulink can accurately establish PMSM vector control system simulation model which based on magnetic field orientation. Three closed-loop control system control process: through position regulator seek the velocity command value after position instruction value and feedback value compared; through velocity regulator get current (vector) size of the command value after the speed instruction value and the speed feedback value [14-15]. According to the current command value and the actual position value calculated three-phase instantaneous current command value, the stator current is close to the command value after the current closed loop control. Current regulator output value through SVPWM module, output of 6 PWM wave controlled inverter module, produce the desired phase current supply PMSM. PMSM three closed-loop vector control simulation model system diagram as shown in figure 15. This paper chooses the rated speed of the motor is 1000rad/s, pole number is 4, the position reference value
0
0
0
0
0.05
0.05
0.1
0.1
0.15
0.15
0.25
0.25 t/s
0.3
0.3
0.35
0.35
0.4
0.4
0.45
0.45
0.5
0.5
Figure 16. Response of Speed
Torque and excitation component of current waveforms are shown in Figure 17, Torque and excitation component of the stator have fluctuation when the motor start and stop, when the motor operated stably ,the fluctuation is small; At 0.1s joined the load, torque component corresponding rapidly, operated stably after a greater fluctuation; the motor speed decreases and eventually stop the rotation, during the motor brake process the torque and the excitation component of current has a bigger wave motion, finally the two components are reduced to 0 A.
Figure 15. The Block Diagram of the Control System of PMSM
© 2013 ACADEMY PUBLISHER
0.2
0.2
930
JOURNAL OF NETWORKS, VOL. 8, NO. 4, APRIL 2013
4040 3030 2020 Iq
Id Iq/A
1010 00 -10-10
Id
-20-20 -30
-30
-40-400 0
0.05
0.05
0.1
0.1
0.15
0.15
0.2
0.2
0.25
0.3
0.25 t/s
0.3
0.35
0.4
0.35
0.45
0.4
0.5
0.45
0.5
Figure 17. The Torque and the Excitation Component of Current
Electromagnetic torque and lode torque waveforms are shown in Figure 18. 4040
operation, 0.1s added load, three-phase current value rose rapidly; when the speed drop suddenly, and eventually reduced to 0 r/s, the current of each phase of the stator, first there are small fluctuations, and then stabilized and, ultimately, the phase current values down to 0. By the simulation waveforms visible, motor starts fast, and can quickly and accurately track a given speed. In the case of load increase, speed after short-term volatility can track the given speed, the speed fluctuation is very small. In the debugging system, the speed of the motor starter - stop process curve shows that motor by the rated speed quickly arrived at the designated location and is able to accurately locate parking. The control system can rapidly reach steady state, overshoot, and steady-state error is very small, and the experimental results show that the controller design is reasonable and has a good dynamic and static performance.
30
30
V. CONCLUSION
2020 TL
Te TL(N*m)
1010 00 -10-10
Te -20-20 -30-30 -40-400 0
0.05
0.05
0.1
0.1
0.15
0.15
0.2
0.2
0.25
0.25 t/s
0.3
0.3
0.35
0.35
0.4
0.4
0.45
0.45
0.5
0.5
Figure 18. Electromagnetic Torque and Lode Torque
We can see that electromagnetic torque of the motor changed always around the load torque; electromagnetic torque has greater fluctuating produced starting torque as motor start, at 0.1s joined 4 N ⋅ m , electromagnetic torque rapidly achieved 4 N ⋅ m , motor speed decreases and eventually stops rotating, electromagnetic torque drops rapidly and produce a reverse braking torque, eventually stabilizing at 0 N ⋅ m . 30
30
2020
This article used advanced digital signal processing chip and designed a PMSM digital controller. By analyzing the model and control strategy of PMSM, Proposed the main hardware structure part and the software program of the PMSM controller, and built a simulation model of the controller. Through the simulation experimental study on PMSM servo system startup, operation and parking, you can get the conclusion: the whole process of the system, speed is running smoothly, the torque ripple is small, small overshoot and precise positioning, and controller designed reasonably and has good dynamic and static performance. ACKNOWLEDGEMENTS The paper is supported by Key Projects of Henan Provincial Department of Education Science and Technology Research (12A47000), Laboratory of Fund Colleges Control Engineering Key Disciplines of Henan Province (KG2011-12), Science and Technology Research of China Coal Industry Association (MTKJ2012-376) and Dr. Foundation of Henan Polytechnic University under Grand (B2011-104).
10
10 I/A
REFERENCES 00
[1] -10
-10
[2]
-20-20 -30-30 0 0
0.05
0.05
0.1
0.1
0.15
0.15
0.2
0.2
0.25
0.25 t/s
0.3
0.3
0.35
0.35
0.4
0.4
0.45
0.45
0.5
0.5
[3]
Figure 19. Stator Current Waveform
The three-phase stator current waveform as shown in Figure 19, as can be seen from the graph, when starting the motor current has fluctuation; the current value is very small close to 0 A when the motor unloaded stable
© 2013 ACADEMY PUBLISHER
[4]
Miao H, Zhenhua L,“Design of Sensorless PMSM Controllers Based on DSP”. Small & Special Electrical Machines,vol. 38, no. 10, pp.63-65, 2010. Wen D, Lin G, Deliang L, Ping S,“Modeling and Simulation of Vector Control System for Permanent Magnet Synchronous Motors”. Micromotors,vol. 43, no. 12, pp.66-71, 2009. Bin Chen, Hong-zhen Yang, Li-wen Wang, “Single Neuron PID Control of Aircraft Deicing Fluids Rapid Heating System”. Journal of Networks. vol. 8, no. 2, pp.405-412, 2013. Tengyue Mao, Zhengquan Xu, Rui Hou, Min Peng, “Efficient Satellite Scheduling Based on Improved Vector Evaluated Genetic Algorithm”. Journal of Networks. Vol.7, no.3, 2012
JOURNAL OF NETWORKS, VOL. 8, NO. 4, APRIL 2013
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
Jianhong L, Yangzhong Z,“The Design for a Permanent Magnet Synchronous Motor Servo Control System Based on DSP”. Power Electronics, vol. 46, no. 1, pp.79-81, 2012. Xiaoli C, Linna H,“The Design of Drives System for PMSM Based on DSP”. Power Electronics, vol. 45, no.11, pp.115-117, 2011. Bingqiang L, Hui L,“Direct Control of Current Vector for Surface-mounted Permanent Magnet Synchronous Motor”. Proceedings of the CSEE, vol. 31, no.13, pp.288-294, 2011. Tiebing L, Chunjiang L, Yuejin Z, Huichen M,“The Study of PMSM Vector Control System Based on DSP”. Micromotors, vol. 43, no. 12, pp.34-38, 2010. Liu Yanping, Liu Shuhong, Wang Huajun. “DSP control implement of permanent magnet synchronous AC servo motor based on vector control”. 2009 IEEE International Conference on Mechatronics and Automation, pp. 866-870, 2009. Hui Z, Chao L,“The Research of Control Strategy for PMSM Based on SVPWM”. Electrical Measurement & Instrumentation, vol. 46, no. 7, pp.34-38, 2009. Naiqing X, Jishen P, Jian L,“Servo Drive Control of PMSM Research Based on IPM”. Control & Automation, vol. 27, no. 5, pp.51-54, 2011. Junfeng J, Xuejuan T,“Digital Vector Control System of Permanent Magnet Synchronous Motor Based on DSP”. Electric Machines & Control Application, vol. 37, no. 2, pp.43-46, 2010. Yannan X, “Research on Servo Vector Control System Based on TMS320F2812”. Micromotors, vol. 44, no. 1, pp.74-77, 2011. Junmei Guo,Yiwei Zhao, “DSP Practical Solutions for Motor Control Using DSP-Controller”. 2010 IEEE International Conference on Networking and Digital Society, pp. 629-634,2010. Murat Karabacak,Halil Ibrahim,“Design, modelling and simulation of a new nonlinear and full adaptive backstepping speed tracking controller for uncertain PMSM”.Applied Mathenatical Modelling, vol. 27, no. 21, pp.1-15, 2011.
Jun Zhu, male, born in Wulanchabu City, Inner Mongolia Autonomous Region, China, in October, 1984. He obtained PH.D degree in college of Mechanical and Electrical Engineering, Inner Mongolia Agricultural University, China, in 2010. His research fields are special motor drive, nonlinear motor control and motion control. He has been working at Henan Polytechnic University since 2001, and is interesting in teaching and scientific research. He is a researcher of Special Electric Machines & Drives (SEMD) in Henan Polytechnic University. His research works mainly engaged in novel control theory research and the new control system development of special motor. Dr. Zhu is a member of Henan Province institute of special electric machine and drives. In recent years, he presided 3 provincial and one country research projects, and published more than 20 academic papers. He obtained 3 national invention patents, and 2 provincial academic rewards.
© 2013 ACADEMY PUBLISHER
931
Li Wankui, male, born in Xinyang City, Henan Province, China, in August, 1988. He is studying at the college of Electrical Henan Engineering and Automation, Polytechnic University, China, since 2011. His research fields are special motor drive, PMSM control and motion control. He obtained bachelor's degree in college of Electrical Engineering and Automation, Henan Polytechnic University, China, in 2011.His research works mainly engaged in novel control theory research and the new control system development of PMSM.In recent years, he published one academic paper. Han Lili, female, born in Puyang City, Henan Province, China, in August, 1987. She is a postgraduate student at Henan Polytechnic University. Her research fields are Permanent magnet synchronous motor control Kalman filter algorithm.
JOURNAL OF NETWORKS, VOL. 8, NO. 4, APRIL 2013
PMSM Control System Based on Digital Signal Processor Zhu Jun, Li Wankui, Han Lili
Henan Polytechnic University, College of Electrical Engineering and Automation, Henan Jiaozuo, China Email: [email protected]
through experimental results demonstrate the superiority of the system.
Abstract—For the high power density of PMSM, designed the corresponding drive controller to improve the servo efficiency of PMSM servo system. According to the theory of math model of PMSM under the dq coordinate system, applied id = 0 vector control method as a PMSM control strategy, established PMSM controller model based on vector control. Taking DSP TMS320F2812 as controller core, built a power-driven circuit, control circuit and the main detection protection circuit. The algorithm of control program was completed on the hardware platform to present its software processes. The simulation results show that: the control system response is fast, can track the given speed and position quickly and accurately. The speed fluctuation, overshoot and steady state error are very small. The designed controller is reasonable, which has better dynamic and static characteristics, and be benefit to improve the efficiency of PMSM servo system.
A.The Mathematical Model of the PMSM By the assumption of the desired motor, can be deduced from the mathematical model of PMSM in dq coordinate system [6]: Stator flux equation: ψ d = Ld id + ψ f (1)
Index Terms—PMSM, Overall Control system, DSP,
d -q axis stator inductance; ψ f
Controller
I. INTRODUCTION With the rapid development of permanent magnet materials, power electronics and control theory, based on vector control of PMSM its excellent control performance, high power density and high efficiency, more and more used in a variety of high-performance servo systems and other fields of industrial production [1-2]. In recent years, the development of various control method [3-4] and vector control theory and methods become mature gradually, the progress of integrated circuits and computer technology, the PMSM vector control system has been a great development. The vector control of PMSM based on the mathematical model in the dq coordinate system, through conversion of vector model; realize that it has completely decoupled control of stator current. It has like a DC motor control performance, and thus more widely applied [5]. This paper introduces the mathematical model of PMSM in the dq coordinate system, modeling and analysis of PMSM vector control strategies based on the id = 0 , for DSP TMS320F2812 as controller core, built a power-driven circuit, control circuit, feedback circuit and the main auxiliary circuit. The algorithm of control program was completed based the hardware platform, and presented its software processes. Constitute a high-performance PMSM vector control system, and © 2013 ACADEMY PUBLISHER doi:10.4304/jnw.8.4.924-931
II.SYSTEM MODEL
ψ q =Lq iq
(2)
In the formula, id , iq is the d -q axis stator current;
ψ d , ψ q is the d -q axis stator flux; Ld , Lq is the is the magnetic
potential generated by the permanent magnets on the rotor. Stator voltage equation: u d = Rs id +
u q = Rs iq +
dψ d dt
dψ q dt
− ω rψ q
(3)
− ωrψ d
(4)
In the formula, ud , uq is the d -q axis stator voltage;
ωr is the rotor angular velocity; Rs is the stator
resistance. Electromagnetic torque equation:
3 3 pn (ψ d iq -ψ q id )= pn [ψ f iq -(Lq -Ld )id iq ] (5) 2 2 In the formula, Te is the electromagnetic torque; Pn is Te =
the number of pole pairs. Equations of motion: d ωr J =Te -Bωr -TL (6) dt In the formula, J is a moment of inertia( kg ⋅ m 2 );
TL is the load torque( N ⋅ m ), Te is the output torque( N ⋅ m ); B is the viscous friction coefficient. B. The Basic Principle of PMSM Vector Control In the field oriented coordinates, PMSM vector control
JOURNAL OF NETWORKS, VOL. 8, NO. 4, APRIL 2013
925
simulates the DC motor torque control law, the current vector is decomposed into the excitation current component to produce magnetic flux and the generated torque of the torque current component, the two current components are adjusted respectively, so that decoupling control is realized to control torque component and flux component of PMSM respectively [7-9]. According to the same principle of magnetic potential and power, the coordinate transformation is made to three-phase voltage, current and flux for PMSM. Stationary coordinate system of the three-phase ABC transforms into dq 0 rotating coordinate system, stator current vector is decomposed into the rotor field oriented two mutually orthogonal current component, they are excitation component id and torque component iq of stator current [6]. Set id = 0 ,the component iq is controlled,
System hardware block diagram is shown in Figure 2, the 220V AC is rectified, and filter capacitor filter to obtain a smooth DC output, and finally the alternating current supply was obtained for PMSM through the inverter circuit to convert.
PC
IO Interface Circuits
Keyboard and Display Circuit
CI S
O IP G
NT I DP P
S M T 0 23 21 F 82
MW P
no toip tO al o-s i
Detection and Protection Circuit
Rectifier Circuit
IPM DC A
PI S EP Q
Current Detection Speed and Position Detection
PMSM
it is equivalent to control the torque directly. Electromagnetic torque Te and iq are linear relationship,
Figure 2. PMSM Control System of the Block Diagram
so that the system has good mechanical properties and dynamic performance. iq Regulating reference volume
The stator phase current signal detected by hall current sensor is sent to DSP through the ADC module, it constitutes the current closed-loop control system. The rotor speed and position signal detected by optical encoder are sent to DSP through the QEP module, it constitutes the speed and position control loop. In the DSP, corresponding op conversion is made applying the soft program for the detected signal, SVPWM pulse drives signal is generated which is required in the control. It drives the IPM intelligent power module through opto-isolation circuit, controls the turn-on and turn-off time of IGBT and generates the corresponding voltage signal which controls the motor running. When the system detects fault signal, it is transmit to the DSP power protection input interrupt pins (PDPITNT) through the opto-isolation circuit, so that the power protection interrupt signal is generated to turn off the 6-channel PWM signal output pulse signal achieving the protection of system failure. In addition, the DA converter, PC connection and other external auxiliary circuits are realized through the DSP's SPI, SCI, and GIPO interface.
is given by the speed controller, the d -q axis voltage components are output after current loop adjustment, which are ud , uq . The component uα and uβ are obtained through anti-Parke transform for ud and uq in the α − β coordinate system. According to the value size of uα and uβ and SVPWM space vector method, the output of the vector control is obtained to achieve the purpose of vector Control. It is id = 0 vector control block diagram as shown in Figure 1. ∆θr
θ r* θr
Position Controller
ω r* ∆ωr ωr
Speed Controller
iqref
PI
-i idref= 0
PI
Speed
AntiUα Park SVPWM Transforgenerator mation
Ud
Uβ
q
id-
Position
Uq
Park Transformation
θr 1/s
Inverter iα
ia
Clark Transforiβ mation
ib ic
θr θr
ωr
Optical encoder
Figure 1. PMSM Vector Control Block Diagram
III. SYSTEM DESIGN A. The Hardware Structure of the Controller The controller system includes a main power circuit, control circuit and the motor. The control loop is mainly composed by the master DSP chip and sampling detection circuit. The most advanced chip TMS320F2812 DSP which frequency is 150MHz was used as core controller. The control chip is used widely in the high precision servo control, variable frequency power supply and other areas, while it is the best choice of motor and [10] digital control .
© 2013 ACADEMY PUBLISHER
B. Main Power Circuit Design The system's main power circuit consists of a rectifier circuit, filter circuit and the IPM module. The rectifier circuit part uses single-phase uncontrolled rectifier module GBJ25M. AC is rectified DC, the maximum
311V , GBJ25M rectifier modules voltage is 2 × 220 = are fully able to meet the requirements. The pressure of the filter capacitor is at least as 375V, so the paper selected two 330 µ f / 450V capacitors in parallel with a decrease of the ESR of the capacitor, the pressure value is 450V , the capacitance is 660 µ f . This paper inverter circuit part adopts IPM PS21964, it integrates high-voltage power transistor drive circuit in their own internal, and owns built-in overvoltage, over-current and overheating fault detection circuit, so that it can ensure the safe operation and reliable operation for the controller. Its rated voltage is 600V, rated current is 15A for the
926
JOURNAL OF NETWORKS, VOL. 8, NO. 4, APRIL 2013
PS21964, and the inverter output power is:
detection circuit is shown as in Figure 4.
P = 3V I cos ϕ out out out
+5
+5
(7)
+3.3 +3.3 +3.3
In the formula, Vout is the output voltage rams value; I out is the output current rms value; cos ϕ is the power factor. When inverter operates, taking into account the requirements of motor overload, inverter takes power rating of 1.5 times margin, the power factor is 0.7[11], so the maximum output power of the inverter is: (8) P = 3 × 220 × 15 × 0.7 ÷ 1.5 = 2.6 KW max It has the ability to drive 1KW motor. C. Phase Current Detection Circuit The control loop in order to achieve the current loop, speed loop and position loop three closed-loop control, DSP will detect the current signal, speed signal, and location of the signal after the program operation processing, Event Manager EVA and EVB produce 6-channel PWM signal to control the inverter turn-on and turn-off. PMSM vector control system with id = 0 control strategy must be timely detection the size of the stator phase current value. In this paper, TKC-10P takes as closed-loop hall current sensor. Motor using a Y-shaped connection can only detect two phase currents, shown in Figure 3, in order to achieve the current closed-loop control.
R2
U
TKC-10P
2 1 4 3
R3
R6 C2
+5
6N137
R5 R1 +15V
74HC14
74HC14 MM74HC14
Figure 4. Rotor Position Detection Circuit
According to U,V,W edge of the trigger signal to determine the initial position of the rotor in which the motor is started. Encoder output U,V,W signal after 74HC14, output for the three-way interaction difference of 120 electrical angle, the width of 180 electrical angle of the square wave signal, then after the isolation of high-speed optocoupler6N137,shaping the signal through the Schmitt trigger MM74HC14, to ensure the accuracy of position detection. Finally, through the QEP module is input to TMS320F2812. E. Speed Detection Circuit Optical encoder output is six-way differential signals A ±,B ±,Z ±, its output voltage range is from 0 to 5v, the signal need to be converted by using the differential receiver. Rotor speed detection circuit is shown as in Figure 5. +3.3
+5
C1 - LM358
R8 -
R3
+15
R4
+
R6 LM358
R2
+3.3
R2
+
R7
BAV99
A+
R1
+3.3 +3.3
ADCINAI A-
BAV99
+
R6
R4
C2
R7 C5
QEP
R5
C1 R3
+ LM358
QEP
R4
C1
+5
-15
BAV99
R5
R1
-15V Uo 6 OUT -V GND O/P Ui 5 +V IN
+3.3
C3
C4
6N137
74HC14
DS3486
Figure 3. Current Sensing Circuit
Figure 5. Speed Detection Circuit
The system utilizes TMS320F2812 has 12 high-resolution of the AD converter, to improve the accuracy and timeliness of the current sampling. DSP's ADC input voltage range is 0 to 3.3V, the hall current sensor output current is a bipolar signal, it needs convert the detected current signal into the voltage signal through converter circuit, so that the DSP can identify the signal. The current sensor detects the current signal of the first through the RC filter, and then through a voltage follower, then is made DC bias through the op amp circuit, finally a proportion of operations is made, the output signal range from 0 to 3.3v is obtained. Bias voltage of +5V, the op amp circuit uses a high-speed dual op amp chip LM358. Use switching diode BAV99 as the voltage clamp circuit to prevent voltage exceeds +3.3V.
In this paper, the encoder pulse signals are converted into a single output signal by DS3486, and then isolated by the opto-coupler 6N137, shaping the signal through the Schmitt triggers 74HC14. Finally, the signals are input to TMS320F2812 through the QEP module.
D. Position Detection Circuit Position detection is achieved by using the optical encoder, the article selected the hybrid optical encoder, its single ring pulse number is 2000.Signal output adopts differential driver, It can reduce the common mode interference of the signal transmission. The rotor position © 2013 ACADEMY PUBLISHER
F .The Design of Limiting the Current Starting Circuit The auxiliary circuit of the system by limiting the current starting circuit, DC bus over-current protection circuit, voltage detection circuit, braking circuit, power circuit, chip voltage monitor circuit, the serial communication interface circuit and its peripheral circuits. Limit the current starting circuit in order to prevent the starting current is too large, while damage to the bridge rectifier, during the motor starter. The main circuit will produce a very high current when power is connected, at this point relays no action, +5 V supply is not turned on, current through the thermostat flow into the filter capacitor, thermostat to play the role of limit the current. Delay in seconds, normal both ends of the filter capacitor
JOURNAL OF NETWORKS, VOL. 8, NO. 4, APRIL 2013
927
voltage, DC bus current is reduced, at this time+5 V power supply connected, the normally open relay contact pull, the thermostat short-circuit, end of the limit the current starting. Limit the current starting circuit shown in Figure 6.
R1 +5V C1
+3.3V
C4
C3 DCP
D1
4
R2
3
TLV2374ID +
+5V
2
K
ADC
1
-
R5
11
R3
D2 A
R4
D1 PTC
C2
Figure 8. Voltage Detection Circuit Diagrams
I. Dynamic Braking Circuit
G.DC Bus Over-current Protection Circuit When the DC bus current is too large, IPM module can easily be burned, therefore designed a DC bus over-current protection circuit, Shown in Figure 6. In the filter circuit and IPM modules in series one 0.02Ω sampling resistor, the over-current protection circuit to monitor the pressure drop across the sampling resistor. Over-current detection signal is active low, when the DC bus current is normal, the transistor Q5 is turned off, the comparator LM2903 inverting input is grounded, +15 V power input access the same phase, LM2903 output invalid high signal; When the DC bus over-current, transistor Q5 turns on, comparator LM2903 output low flow signal through opto-coupler to PDPINT pins. Immediately cut off the PWM signal output and LED alarm instructions after DSP receive over-current signal. +5V
When the servo driver brakes frequently with a large inertia load, servo motor will in the power generation state, but rectifier part of the system is uncontrollable rectifier circuit, the energy can’t be fed back into the grid, DC bus voltage of the system will rapidly increase. It is easier to be damaged for the energy storage capacitors, power module, so the dynamic braking circuit must be added for the servo drive system. The TMS320F2812 chip GPIOA pins sends out the brake signal, first through opto-isolation, and then through diode NPVIDT2227 to be amplified 15V level signal, it drives the power transistor Q4 to be turn on, the power is consumed in the resistor R10. When the DC bus voltage is less than 360v, the driven power transistor Q4 is turn off, the braking process is ended The brake circuit is shown as in Figure 9. +3.3V
C2
C3
C4 R4
R5
C3
C4
R2 2 3 NC
GPIOA
R7 +
1
D1 DCP
R6
C1
LM2903
D1
5 3
Q2
1
MMDT2227
PDPINT
2
-
4
R3
4N25
R4 R2
4
R5
2
2
R3 Q1
DCP
+5V 6
5
R1
+3.3V
Q1
+5V
6
1 C1
3 3
Figure 6. Limit the Current Starting Circuit
Figure 9. Dynamic Braking Circuit
C2
R1
J. Power Management Circuit Figure 7. DC Bus Over-current Protection Circuit
H. Voltage Detection Circuit The DC bus voltage detection circuit uses a closed-loop hall voltage sensor, the detected voltage signal is connected to voltage follower circuit through the TLV2374 of precision op amp. It is sent into DSP chip by the A/D converter. When the bus voltage is larger than 380V, the controller sends a brake signal. When the bus voltage is less than 120V, DSP takes as under voltage state, its principle is same as over voltage. The voltage detection circuit is shown as in Figure 8.
+5V
C1
C2
1 2 3 4 5 6 7 8 9 10 11 12 13 14
TPS767D301
28 27 26 25 24 23 22 21 20 19 18 17 16 15
DSP_RST1
R1 R2
SENSE
1.8V DSP_RST2
C3
C4
C5
C6
3.3V
Figure 10. Power Management Circuit
The circuit voltage is 3.3V for TMS320F2812 peripheral interface, its core voltage is 1.8V. In the Paper, power management chip TPS767D301is selected, output © 2013 ACADEMY PUBLISHER
928
JOURNAL OF NETWORKS, VOL. 8, NO. 4, APRIL 2013
two-way power are 1.8V and 3.3V, so that the DSP chip can normal work. DSP has ADC module, the voltage output must distinguish the digital voltage and analog voltage, so the digital ground and analog ground are connected by beads. The circuit is shown as in Figure 10. K. Serial Communication Interface Circuit TMS320F2812 chip has a serial communication interface (SCI) module, the serial port is configured as RS232 port to connect with the PC easy. Because the DSP level and RS232 level cannot match, a external circuit must be constituted by using power level conversion chip MAX232D to meet the communication functions between the control panel and PC. MAX232D uses a 3.3V supply voltage, its external four small capacitor rise the voltage from 11V to 15V. The serial communication interface circuit is shown as in Figure 11.
after the stage circuit electrical isolation, but also has a level conversion function. The IPM module fault detection circuit diagram is shown in Figure 12. M. Temperature detection protection circuit IPM module PS21964 controller uses normal working temperature does not exceed 100, the heat sink is not only to the design of the controller, but also the design temperature detection and protection circuit, to prevent overheating burned chip and other circuit. Paper, the temperature change of the temperature sensor T255 detecting controller, the temperature sensor T255 is reflected by the change in resistance to temperature change, the higher the temperature, the smaller the resistance of the temperature sensor T255.The typical o
value of the temperature sensor T255: R(25 C)=5.0KΩ .
3.3V 16 C1
C2
+3.3V
+3.3V 1
4
2
3
C5
C3 3 4
2 V+ 6 VC4 GND
fault
JP1 1 6 2 7 3 8 4 9 5
5
11 T1IN 12 R1OUT 10 T2IN 9 R2OUT
R2 1K
T255 C6
15 SCITXDA SCIRXDA SCITXDB SCIRXDB
+5V
1
VCC
14 T1OUT 13 R1IN 7 T2OUT 8 R2IN
R1 1.2K
PC181
PDPINT
R3 5.1K
Figure 13. Temperature Detection Protection Circuit
MAX232
N. The Software Structure of the Controller
Figure 11. Communication Interface Circuit
Start
L.IPM module fault detection circuit
The initialization of the system clock
+5V
+3.3V
4
1 R1 10K FO
R2 1K Q1 NPN
The timer is initialized
Initialize the system parameters
fault PDPINT
2
R3 330
PC181
3 The software module initialization
R4 5.1K
Figure 12. Temperature Detection Protection Circuit
This article uses the IPM module PS21964, internal integrated overvoltage, overcurrent, overheating protection function, IPM modules work properly, the PS21964 No.14 pin fault detection signal output terminal of the high level of +5 V,NPN type transistor C9013 and opto-coupler PC181 in the conduction state, the opto-coupler output of +3.3V high. When the IPM module fails, the fault detection signal output terminal output current of 10mA, low pulse width of 1.8ms signal, the NPN type transistor C9013 are in the OFF state and opto-coupler PC181 output is low, that is, a fault signal is active low. The opto-coupler used not only before and
© 2013 ACADEMY PUBLISHER
Timer overflow interrupt and capture interrupt enable
To detect the position of the rotor initialization
Y
Rotor initial position detection subroutine
N Interrupt latency
Y
Interrupt service routine
N
Figure 14. Main Program Flow Chart
A complete software system should include function subroutine, system initialization procedure, the timer main interrupt program, position detection interrupt program and protection interrupt program. Software
JOURNAL OF NETWORKS, VOL. 8, NO. 4, APRIL 2013
is set to 400rad, the simulation time is set to 0.5s.The rotating speed response waveform is shown in Figure 16,can see motor no-load can quickly reach the maximum speed, after slight fluctuations stabilized rated speed in 1000rad/s, when 0.1s load 4 N ⋅ m , speed fluctuate somewhat, but soon was stable in 1000rad/s. When arrived at a set position 400rad, motor speed decreased rapidly, while unload 4 N ⋅ m load, during descending process the speed is corresponding rapidly , with slight fluctuations, finally speed stabilized in 0rad/s and parking at 0.5s. 1200 1200
1000 1000
800800 n(rad/s)
mainly achieve SVPWM control algorithm, the coordinate transformation, digitized PI regulator and current / position signal detection, they are all implemented in the main interrupt program [12]. The first three do not need to control the hardware, the rest requires software and hardware properly fit in order to achieve accurate control effect. The introduction of interference in the signal detection will directly affect the control performance of the system. After control system start plus power, the first need to do is the initialization of system input and output module classes and data processing module class, and system resources, and then is the detection of the initial position angle of the PMSM. After get the initial position angle information of the rotor and then carry on backstage processing program, and wait for interrupt handling signal is used to access the system subroutine in order to control PMSM [13]. The system main program flow chart is shown in Figure 14.
929
600600
400400
IV . SIMULATION RESULTS
200200
Matlab/simulink can accurately establish PMSM vector control system simulation model which based on magnetic field orientation. Three closed-loop control system control process: through position regulator seek the velocity command value after position instruction value and feedback value compared; through velocity regulator get current (vector) size of the command value after the speed instruction value and the speed feedback value [14-15]. According to the current command value and the actual position value calculated three-phase instantaneous current command value, the stator current is close to the command value after the current closed loop control. Current regulator output value through SVPWM module, output of 6 PWM wave controlled inverter module, produce the desired phase current supply PMSM. PMSM three closed-loop vector control simulation model system diagram as shown in figure 15. This paper chooses the rated speed of the motor is 1000rad/s, pole number is 4, the position reference value
0
0
0
0
0.05
0.05
0.1
0.1
0.15
0.15
0.25
0.25 t/s
0.3
0.3
0.35
0.35
0.4
0.4
0.45
0.45
0.5
0.5
Figure 16. Response of Speed
Torque and excitation component of current waveforms are shown in Figure 17, Torque and excitation component of the stator have fluctuation when the motor start and stop, when the motor operated stably ,the fluctuation is small; At 0.1s joined the load, torque component corresponding rapidly, operated stably after a greater fluctuation; the motor speed decreases and eventually stop the rotation, during the motor brake process the torque and the excitation component of current has a bigger wave motion, finally the two components are reduced to 0 A.
Figure 15. The Block Diagram of the Control System of PMSM
© 2013 ACADEMY PUBLISHER
0.2
0.2
930
JOURNAL OF NETWORKS, VOL. 8, NO. 4, APRIL 2013
4040 3030 2020 Iq
Id Iq/A
1010 00 -10-10
Id
-20-20 -30
-30
-40-400 0
0.05
0.05
0.1
0.1
0.15
0.15
0.2
0.2
0.25
0.3
0.25 t/s
0.3
0.35
0.4
0.35
0.45
0.4
0.5
0.45
0.5
Figure 17. The Torque and the Excitation Component of Current
Electromagnetic torque and lode torque waveforms are shown in Figure 18. 4040
operation, 0.1s added load, three-phase current value rose rapidly; when the speed drop suddenly, and eventually reduced to 0 r/s, the current of each phase of the stator, first there are small fluctuations, and then stabilized and, ultimately, the phase current values down to 0. By the simulation waveforms visible, motor starts fast, and can quickly and accurately track a given speed. In the case of load increase, speed after short-term volatility can track the given speed, the speed fluctuation is very small. In the debugging system, the speed of the motor starter - stop process curve shows that motor by the rated speed quickly arrived at the designated location and is able to accurately locate parking. The control system can rapidly reach steady state, overshoot, and steady-state error is very small, and the experimental results show that the controller design is reasonable and has a good dynamic and static performance.
30
30
V. CONCLUSION
2020 TL
Te TL(N*m)
1010 00 -10-10
Te -20-20 -30-30 -40-400 0
0.05
0.05
0.1
0.1
0.15
0.15
0.2
0.2
0.25
0.25 t/s
0.3
0.3
0.35
0.35
0.4
0.4
0.45
0.45
0.5
0.5
Figure 18. Electromagnetic Torque and Lode Torque
We can see that electromagnetic torque of the motor changed always around the load torque; electromagnetic torque has greater fluctuating produced starting torque as motor start, at 0.1s joined 4 N ⋅ m , electromagnetic torque rapidly achieved 4 N ⋅ m , motor speed decreases and eventually stops rotating, electromagnetic torque drops rapidly and produce a reverse braking torque, eventually stabilizing at 0 N ⋅ m . 30
30
2020
This article used advanced digital signal processing chip and designed a PMSM digital controller. By analyzing the model and control strategy of PMSM, Proposed the main hardware structure part and the software program of the PMSM controller, and built a simulation model of the controller. Through the simulation experimental study on PMSM servo system startup, operation and parking, you can get the conclusion: the whole process of the system, speed is running smoothly, the torque ripple is small, small overshoot and precise positioning, and controller designed reasonably and has good dynamic and static performance. ACKNOWLEDGEMENTS The paper is supported by Key Projects of Henan Provincial Department of Education Science and Technology Research (12A47000), Laboratory of Fund Colleges Control Engineering Key Disciplines of Henan Province (KG2011-12), Science and Technology Research of China Coal Industry Association (MTKJ2012-376) and Dr. Foundation of Henan Polytechnic University under Grand (B2011-104).
10
10 I/A
REFERENCES 00
[1] -10
-10
[2]
-20-20 -30-30 0 0
0.05
0.05
0.1
0.1
0.15
0.15
0.2
0.2
0.25
0.25 t/s
0.3
0.3
0.35
0.35
0.4
0.4
0.45
0.45
0.5
0.5
[3]
Figure 19. Stator Current Waveform
The three-phase stator current waveform as shown in Figure 19, as can be seen from the graph, when starting the motor current has fluctuation; the current value is very small close to 0 A when the motor unloaded stable
© 2013 ACADEMY PUBLISHER
[4]
Miao H, Zhenhua L,“Design of Sensorless PMSM Controllers Based on DSP”. Small & Special Electrical Machines,vol. 38, no. 10, pp.63-65, 2010. Wen D, Lin G, Deliang L, Ping S,“Modeling and Simulation of Vector Control System for Permanent Magnet Synchronous Motors”. Micromotors,vol. 43, no. 12, pp.66-71, 2009. Bin Chen, Hong-zhen Yang, Li-wen Wang, “Single Neuron PID Control of Aircraft Deicing Fluids Rapid Heating System”. Journal of Networks. vol. 8, no. 2, pp.405-412, 2013. Tengyue Mao, Zhengquan Xu, Rui Hou, Min Peng, “Efficient Satellite Scheduling Based on Improved Vector Evaluated Genetic Algorithm”. Journal of Networks. Vol.7, no.3, 2012
JOURNAL OF NETWORKS, VOL. 8, NO. 4, APRIL 2013
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
Jianhong L, Yangzhong Z,“The Design for a Permanent Magnet Synchronous Motor Servo Control System Based on DSP”. Power Electronics, vol. 46, no. 1, pp.79-81, 2012. Xiaoli C, Linna H,“The Design of Drives System for PMSM Based on DSP”. Power Electronics, vol. 45, no.11, pp.115-117, 2011. Bingqiang L, Hui L,“Direct Control of Current Vector for Surface-mounted Permanent Magnet Synchronous Motor”. Proceedings of the CSEE, vol. 31, no.13, pp.288-294, 2011. Tiebing L, Chunjiang L, Yuejin Z, Huichen M,“The Study of PMSM Vector Control System Based on DSP”. Micromotors, vol. 43, no. 12, pp.34-38, 2010. Liu Yanping, Liu Shuhong, Wang Huajun. “DSP control implement of permanent magnet synchronous AC servo motor based on vector control”. 2009 IEEE International Conference on Mechatronics and Automation, pp. 866-870, 2009. Hui Z, Chao L,“The Research of Control Strategy for PMSM Based on SVPWM”. Electrical Measurement & Instrumentation, vol. 46, no. 7, pp.34-38, 2009. Naiqing X, Jishen P, Jian L,“Servo Drive Control of PMSM Research Based on IPM”. Control & Automation, vol. 27, no. 5, pp.51-54, 2011. Junfeng J, Xuejuan T,“Digital Vector Control System of Permanent Magnet Synchronous Motor Based on DSP”. Electric Machines & Control Application, vol. 37, no. 2, pp.43-46, 2010. Yannan X, “Research on Servo Vector Control System Based on TMS320F2812”. Micromotors, vol. 44, no. 1, pp.74-77, 2011. Junmei Guo,Yiwei Zhao, “DSP Practical Solutions for Motor Control Using DSP-Controller”. 2010 IEEE International Conference on Networking and Digital Society, pp. 629-634,2010. Murat Karabacak,Halil Ibrahim,“Design, modelling and simulation of a new nonlinear and full adaptive backstepping speed tracking controller for uncertain PMSM”.Applied Mathenatical Modelling, vol. 27, no. 21, pp.1-15, 2011.
Jun Zhu, male, born in Wulanchabu City, Inner Mongolia Autonomous Region, China, in October, 1984. He obtained PH.D degree in college of Mechanical and Electrical Engineering, Inner Mongolia Agricultural University, China, in 2010. His research fields are special motor drive, nonlinear motor control and motion control. He has been working at Henan Polytechnic University since 2001, and is interesting in teaching and scientific research. He is a researcher of Special Electric Machines & Drives (SEMD) in Henan Polytechnic University. His research works mainly engaged in novel control theory research and the new control system development of special motor. Dr. Zhu is a member of Henan Province institute of special electric machine and drives. In recent years, he presided 3 provincial and one country research projects, and published more than 20 academic papers. He obtained 3 national invention patents, and 2 provincial academic rewards.
© 2013 ACADEMY PUBLISHER
931
Li Wankui, male, born in Xinyang City, Henan Province, China, in August, 1988. He is studying at the college of Electrical Henan Engineering and Automation, Polytechnic University, China, since 2011. His research fields are special motor drive, PMSM control and motion control. He obtained bachelor's degree in college of Electrical Engineering and Automation, Henan Polytechnic University, China, in 2011.His research works mainly engaged in novel control theory research and the new control system development of PMSM.In recent years, he published one academic paper. Han Lili, female, born in Puyang City, Henan Province, China, in August, 1987. She is a postgraduate student at Henan Polytechnic University. Her research fields are Permanent magnet synchronous motor control Kalman filter algorithm.
