Vector Control of Permanent Magnet Synchronous Motor for Fan of New Energy Vehicle

Transcription

1 Vector Control of Permanent Magnet Synchronous Motor for Fan of New Energy Vehicle Liu Yong* Yancheng Teachers University, school of mathematics an statistics, China Abstract:In this paper, base on the analysis of the mathematical moel of permanent magnet synchronous motor (PMSM) an the common control strategy, an the overall system architecture of the control system is analyze base on the specific control algorithm.the control system uses NXP company's latest special motor control chip MC9S12ZVMC128 as the system controller an uses vector control (FOC) as the control algorithm.the esign of controller software architecture is escribe in etail, using the state machine moel to control the motor running in ifferent stages, incluing start-up stage, motor open etection an treatment of close loop phase an fault ring stage, etc...the experimental results show that,with the NXP MC9S12ZVMC128 harware platform, using vector control algorithm,permanent magnet synchronous motor can operate efficiently an stably, give full play to the role of MC9S12ZVMC128 in the electronic fiel of the avantages of new energy vehicles. Key wors:permanent magnet synchronous motor; new energy vehicle; vector control; MC9S12ZVMC Introuction In the environment an the energy eman increasingly higher toay, energy consumption has become a serious problem, which resulte in the popularization an evelopment of automobile energy consumption into explosive evelopment, energy-saving an emission-reuction has become a major tren, new energy vehicles came into being [1].There are a large number of motors in the vehicle,which makes it the most energy consumption part in the vehicle.such as automotive air conitioning, fans, pumps, oil pumps, wipers an other parts of the motor control [2].At present, this part of the motor is still a lot of traitional brush motor or low efficiency of DC motor. In the context of energy-saving emission reuction, reuire the use of power can achieve higher efficiency,increase the number of the evelopment an use of more efficient spee control systems,permanent magnet synchronous motor (PMSM) control system has great potential in prouction an life because of its goo control performance an operation performance[3].permanent magnet synchronous motor (PMSM) has been wiely use in this kin of backgroun, an the brush motor on the vehicle is being replace by brushless permanent magnet synchronous motor (PMSM) step by step. Because of the goo performance, PMSM is wiely use 302

2 in many fiels, an the control is also showing more efficient an stable tren, a variety of control theory an control algorithms have emerge [4-5]. Due to the many avantages of PMSM, many scholars an experts put the research irection to the PMSM, mainly ivie into the following three irections[6]: one is the stuy of control strategy, that is, software an control algorithm. One is the esign of the motor, incluing the use of higher performance motor esign, improve its power ensity. Another is the esign of the motor controller, the controller harware circuit esign.the most of them is that researching the controller an control algorithm, at present, it has become a hot research topic in major universities an enterprises. Stuying on the controller, uses MCU as the controller, three-phase brige as the riving circuit, low voltage control high pressure an SVPWM technology, because the MOSFET turn-on voltage is a little high, The PWM generate by the MCU is not sufficient to rive the MOSFET or IGBT on, often on many occasions nee to a aitional riving circuit.furthermore, using the FOC algorithm nee to sample the two-phase current, an when the bus voltage is over voltage or uner voltage, it is necessary to etect the fault an protect control system. In aition, the controller to eal with the FOC algorithm, must have a certain amount of computing power.therefore, now there are more an more PWM riving circuit an operational amplifier integrate in a MCU,which makes harware more simplifie[7]. There are a lot of PMSM control system usually using the encoer as the position sensor, but more an more occasion nee the harware system more simplifie, slowly,pmsm position sensorless control strategy graually gets more an more wiel [8].In the fiel of automotive electronics, PMSM has been wiely use, the major semiconuctor companies were launche its own eicate motor control MCU, such as Infineon (Infineon), ELMOs En Zhipu (NXP), etc.. In this paper, NXP is use in the fiel of automotive electronic control of motor MCU. The software, for the control algorithm of PMSM currently has the following three kins[9]:vector control (Fiel Oriente Control, referre to as FOC), VVVF control (Variable Voltage Variable Freuency, referre to as VVVF or VF) an irect torue (Direct Torue Control, referre to as DTC). FOC has been wiely use in many PMSM control systems because its goo 303

3 ynamic response an high stability [10]. The esign in this paper mainly accoring to the relate technical ocuments of NXP, base on a large number of omestic an international materials, esigns a PMSM-FOC system base on MC9S12ZVMC128, after repeate experiments, the new energy automobile fan with PMSM-FOC system, the test run results up to the reuirements of the project. 2. Principles of PMSM vector control 2.1 mathematical moel of PMSM Permanent magnet synchronous motor (PMSM) is a kin of strongly couple non-linear system. Its mathematical moel is the theoretical basis of the analysis of motor performance an the realization of torue an spee control. The Clark transform an the Park transform are use to transform the motor from the three-phase static coorinate system to the two-phase rotating coorinate system to realize the ecoupling of the torue an flux. Voltage moel in - coorinate system is: i u ri L we L i (1) t i u ri L we ( L i f ) t The expression of electromagnetic torue prouce by stator wining is: 3 3 Te N ( L ) p L ii N pi f (2) 2 2 Motor motion euation is: 3 wm Te TL N p ( i i ) TL J (3) 2 t i u L u L i f T e T L In the formula: is straight shaft voltage, is uarature axis voltage; is irect axis current, is uarature axis current; is irect axis inuctance, is uarature axis inuctance; is the rotor flux reference value; is motor output electromagnetic torue ; is loa torue; 304

4 N is motor pole pairs; an respectively are electric angular velocity an angular velocity. p w e w m 2.2 space vector pulse with moulation The first step of space vector moulation (SVPWM) is to analyze the motor ieally,then establish the mathematical moel of the motor in each state.an accoring to the coorinate transformation of the motor space vector combine with the moel.accoring to the switching state of the inverter, an appropriate pulse with moulation is generate, so that the actual flux of the permanent magnet synchronous motor is infinitely close to the ieal flux circle. Space vector pulse with moulation (PWM) technology has a very obvious avantage, it can prouce stanar sine wave an can provie high voltage an ecrease low current istortion. The inverter is compose of three half brige,in total,six power tubes, we set "1" as turn on an "0" as turn off, six power tubes have eight kins of combination state, namely 000, 001, 010, 011, 100, 101, 110, 111. Schematic as shown in figure (1). Fig. 1 principle of space vector pulse with moulation inverter 305

5 The power tube of top an bottom work in a complementary moe, if the upper tube S At is "1",the S Ab corresponing is "0"; in the space vector pulse with moulation, eight voltage vector synthesize a resultant vector, of which there are two zero vector. As shown in Figure 2, a perio can be ivie into six sectors, each of the ajacent two sectors between the angle is 60 o, the mile point of sector,(000) an (111) represent the zero vector. (011) U180 [-2/ 3,0] (010) U120 β U60 (110) [1/ 3,-1] [1/ 3,1] (111) (000) o o Us U0 (100) [2/ 3,0] α (001) U240 U300 (101) [-1/ 3,-1] [-1/ 3,1] Fig. 2 Diagram of voltage vector 2.3 construction of vector control system for permanent magnet synchronous motor The characteristic of the vector control algorithm is to control the stator current(i.e.,control i i an ) to realize the torue control. The system control i an i by two PI controllers an the system structure is shown in figure

6 i i PI u Park 反变换 u SVPWM 三相桥 PMSM i PI u u i i i i i ABC i A i B i C A/D Park 变换 Clark 变换 转子位置 Fig. 3 FOC control s structure It can be seen from Figure 3, to achieve permanent magnet synchronous motor vector control nee the AD sampling to obtain two-phase current, an through the back electromotive force to estimate the rotor position an spee, then accoring to the location of the rotor etermine i an i,an through the two PI regulator to ajust. 3. Design of motor control state machine The whole process that the motor from starting to the open loop,then to the close-loop is all base on the state machine to achieve, the state machine uses a two-imensional array function pointer state_table[][] (). The first parameter escribes the current event, an the secon parameter escribes the program state. Figure 4 shows the principle of the state machine. 307

7 Fig. 4 principle of the state machine As shown in Figure 4, the FAULT state can be entere from any other state, an when the program etects an error, it will jump to the FAULT an shutown PWM. When the program etects over voltage, uner voltage, over current or over temperature, it will enter this state. In INIT state, it will initialize the state machine an some parameters of vector control algorithm, incluing the parameters of the PI regulator. After the completion of INIT,READY will be execute as the next state. After entering the READY state, through the etection of the PWM's uty cycle, to etermine whether the motor is starte, if the start is effective,then set the corresponing spee. It will set to the corresponing flag after the en of READY's execution, an enter the next state,namely CALIB. In the CALIB state, the PWM is output accoring to 50% uty cycle, ADC0 an ADC1 starte to work, it can calibrate the sampling of ADC in this state, after the calibration will enter a state of ALIGN an set the corresponing flag.since the permanent magnet synchronous motor in use is without position sensor, on't know the specific position of the motor at first. In the ALIGN state, the voltage is applie irectly to the A phase for a perio of time, the rotor will rotate until the rotor flux an the stator flux are aligne.when the alignment status is complete, the corresponing flag is set an enter the running status. 308

8 Figure 5 shows the flow chart of the running state of the permanent magnet synchronous motor, the corresponing flag is reset when the program is in the RUN state, just enter the RUN state rate is too low, can not prouce enough the back EMF to recognize, so the motor will enter into an open loop moe, its location in this moe is not known, only use the corresponing acceleration formulas to make motor reaches a certain spee to ientify the information of specific position, when it is enough to ientify the information of specific location,the motor convert from open-loop to close-loop,then start to execute the FOC algorithm processing an back EMF observation Device in the close-loop moe. Fig. 5 flow chart in RUN state 309

9 In Figure 5, you can choose a ifferent way of working,when choosing manual moe,the each process of motor's starting can use FreeMASTER to sen commans to achieve the control of close-loop,if select the automatic moe,it will control the whole process automatically by software, so choose the automatic moe in practical application. 4. Position etection base on back EMF When the rotor rises to a certain spee, it can etect the back EMF,the rotor position information of permanent magnet synchronous motor is containe in the motor's stator voltage an current signals, so by accurately etecting the stator voltage an current signal an using the mathematical moel an vector transform, can get the rotor position information, an then obtain the information of spee estimation. In this paper, the position an velocity of the rotor are estimate by referring to the Luenberger observer. The estimation of back EMF as shown in Figure 6, Luenberger stator current observer is the back EMF (BEMF) filter. u i L 1 sl Rs L F c ( s) estim e estim e i L L 1 sl Rs F c ( s) estim e u Fig. 6 Luenberger current observer 5. Achievement of FreeMASTER through the calibration of CAN bus The calibration is through the PC software to moify the controller parameters online to confirm the optimal parameters, in automotive electronics, commonly use CAN bus as the communication moe of the calibration, use in this stuy is CAN communication an FreeMASTER 2.0 as the calibration software on the PC sie, in aition, the FreeMASTERr 310

10 bottom-riven nee integrating with the software. So that you can rea an write any variables in the microcontroller, incluing the value of the register, etc.. Fig. 7 FreeMASTER 2.0 PC ebugging interface Figure 7 is a FreeMASTER 2 PC ebugging interface,in orer to achieve online calibration not only nee the software, but also nee to integrate the unerlying coe of FreeMASTER into the project, the unerlying coe of FreeMASTER support SCI, CAN, BDM communication, by configuring the corresponing interface function an initializing parameter function can achieve online calibration through CAN bus. 6. experimental results an analysis As shown in Figure 8 is FreeMASTER2.0 ebugging interface, through the set, it can isplay all the ata variables use in motor ebugging, can also put specific variables accoring to the proportion into the curve tool to view the variables' change so that make it convenient that observing PI ajustment an other variables change. The interface shows whether the motor break own when it is running, as well as the change of the bus voltage, the analog ial isplay spee, you can see the bus voltage is 12V, the spee as the same as the set 944 RPM. 311

11 Fig. 8 FreeMASTER2.0 ebugging interface of permanent magnet synchronous motor Fig. 9 curve of reaction time In Figure 9, the orange curve is a given spee, the brown is the actual spee, the gray is the spee error, it can be seen from the figure, the error of the spee is relatively small, the given spee an the actual spee is basically synchronize. Figure 10, 11 are respectively experimental fan PMSM an controller, the controller eliminates the traitional gate rive, simple structure, volume is shrunk. Fig. 10 fan use in experiment Fig. 11 the controller of the fan As can be seen from the figure, MC9S12ZVMC128 in the automotive electronic PMSM 312

12 control system, making the entire system circuit is more simple. In aition, the system can not only realize the vector control of non-position sensor PMSM, can also use BLDC six step suare wave control metho to control non-position sensor Brushless DC motor, with the hans of the resources,successfully achieve the control of oil pump, water pump, fan brushless motor (PMSM or BLDC). 7. Conclusion Accoring to the relevant technical ocuments of NXP, base on a large number of omestic an foreign relate ata, esign a vector control system of permanent magnet synchronous motor base on MC9S12ZVMC128, after repeate failures an try, achieve vector control system of new energy automobile fan with permanent magnet synchronous motor, the test run results achieve the reuirements of the subject. The content complete in the esign process as the following: (1) come to know the evelopment of permanent magnet synchronous motor (PMSM) an the application of permanent magnet synchronous motor (PMSM) in the fiel of automotive electronics at the present stage by some omestic an foreign references. By stuying the basic principle an mathematical moel of permanent magnet synchronous motor, the control strategy of permanent magnet synchronous motor (PMSM) is more eeply unerstoo. (2) The esign of controller software architecture is escribe in etail, using the state machine moel to control the motor running in ifferent stages, incluing start-up stage, motor open etection an treatment of close loop phase an fault ring stage, etc...the experimental results show that, with the NXP MC9S12ZVMC128 harware platform, using vector control algorithm, permanent magnet synchronous motor can operate efficiently an stably, give full play to the role of MC9S12ZVMC128 in the electronic fiel of the avantages of new energy vehicles. BIBLIOGRAPHY [1] Alexanrou A D, Aamopoulos N K, Klaas A G. Development of a Constant Switching Freuency Deabeat Preictive Control Techniue for Fiel Oriente Synchronous Permanent Magnet Motor Drive[J]. [2] Kangkang Z, Jianiu L, Minggao O, et al. Electric braking performance analysis of PMSM for electric vehicle applications[c]//electronic an Mechanical Engineering an Information 313

13 Technology (EMEIT), 2011 International Conference on. IEEE, 2011, 5: [3] Lara J, Xu J, Chanra A. Effects of Rotor Position Error in the Performance of Fiel Oriente Controlle PMSM Drives for Electric Vehicle Traction Applications[J]. [4] Lee S, Song B M, Won T H. Evaluation of a software configurable igital controller for the permanent magnet synchronous motor using fiel-oriente control[c]// n Southeastern Symposium on System Theory (SSST). IEEE, 2010: [5] Antony M S, Raj R S P. Four uarant operation of vector control of PMSM with ynamic braking[c]//2015 International Conference on Control Communication & Computing Inia (ICCC). IEEE, 2015: [6] Shinnaka S, Amano Y. Elliptical Trajectory-Oriente Vector Control for Energy-Efficient/Wie-Spee-Range Drives of Sensorless PMSM[J]. IEEE Transactions on Inustry Applications, 2015, 51(4): [7] Hafez B, Abel-Khalik A S, Massou A M, et al. Single-sensor-base three-phase permanent-magnet synchronous motor rive system with luenberger observers for motor line current reconstruction[j]. IEEE Transactions on Inustry Applications, 2014, 50(4): [8] Wang Z, Chen J, Cheng M, et al. Fiel-Oriente Control an Direct Torue Control for Parallele VSIs Fe PMSM Drives With Variable Switching Freuencies[J]. IEEE Transactions on Power Electronics, 2016, 31(3): [9] Sreejeth M, Singh M, Kumar P. Particle swarm optimisation in efficiency improvement of vector controlle surface mounte permanent magnet synchronous motor rive[j]. IET Power Electronics, 2015, 8(5): [10] Chen T. Moeling an simulation of PMSM servo system base on SVPWM[C]//Control Conference (CCC), r Chinese. IEEE, 2014: