Implementing Field-Oriented Control of AC Motors with the
Contents
1. Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP SPRAS26 4 Asynchronous Motor Modeling The induction machine is a non linear high order system and for this reason complicated models must be used to control it The dynamic behaviour of the induction motors can be described by a set of differential equations in a rotating frame of reference 14 16 with the angular velocity For the purposes of this work it is advantageous to express these equations in a rotating frame of reference d q with the same angular velocity This means that the general equation of the torque is transformed to an advantageous form If the fluxes expressed by means of currents and the terms i and y are eliminated moreover it is assumed that a stationary reference frame is fixed to the stator o 0 we obtain the general two phase model in a matrix form 18 Figure 4 This model is used in the motor simulation Figure 4 Matrix Form of Two Phased Model La dia dt Va V ae y LnRr OL A q oL OLL oLL oL Qo 9L LR MES E oL oLL OLL oL isl Er g A maduro O a n R V 0 E o E 0 0 L R r The program calculates the voltages corresponding to the necessary current of the motor instead of using a current control loop Because the algorithm running time is less than the rotor time constant the flux can be considered as a constant Consequently we can choose an equivalent circuit2. Hardware Requirements for the Debugger eeeeeeee 31 Description of the Debugger emet th Re ERR ERIS SER IR ERE tes 31 Special Test Feature of the Debugger seeeseeeeeeeeee 31 Experimental Results susana dogs ce sucus Moda ada cai riii diu inde isa o Ligia bia ia EXE AE CORN TIE ala 33 COCU OS e 35 Mp wp 36 Relevant Literature sos in iaceret ia ere o eee d Eee Un a ota v dedu a adress ced een d cesa r o dua CR eaa 37 Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Table 1 Table 2 Figures Fuzzy Logic Controll r oio o epi n o He ER e ERR en 10 Compositional Rules of Inference erre 11 MAMDANI GOnIFGIIBE sis coorta e oot peo O RERO ROO ERR ERG PR dn 12 Matrix Form of Two Phased Model e sete eds rene 15 Control System Block Diagram esta teca deren Ae aun ee be edet ud 17 Hardware Block Scheme ee toe IIb Prae eie sis 20 Block Diagram of the TI TMS320C25 DSP sese 21 Memory Configuration after Reset rrenan 23 Internal Architecture of the XILINX I O Interface iii 24 Memory Configuration without EPROM a 27 Input Output Address Region Map eee 28 IBM PC DSP Target Board Communication sees 30 S
3. Oriented Control of AC Motors with the TMS320C25 DSP X SPRA326 Hardware Requirements for the Debugger There are some restrictions for target boards intended to be used with the DSP Debugger Q Memory requirements the Debugger needs approximately 1 kword ROM program memory and 30 words of data memory For that latter one the DSP s on chip RAM Block 2 is reserved Appropriate serial port described previously O Writeable program memory for the user software if the debugged routine is longer than 256 words O If step by step program execution is required the Debugger uses the TRAP instruction Description of the Debugger The DSP Debugger program has a simple debugger like user interface The user is able to start and stop programs display and modify the contents of the program data memory and the memory mapped registers There is a more powerful user interface that can be used during program development The list of the debugged program is displayed on the screen The breakpoints and the next instruction to be executed are shown with another display attribute Instead of unassembling the program memory contents the Debugger uses a special list file that can be generated from the assembler list file Using this user interface program debugging will be more simple and comfortable Special Test Feature of the Debugger There are some special functions and commands for testing the target board s hardware components
4. The testing method used in the Debugger serves two purposes to explore the manufacturing defects workshop test and to examine the board s operational capability by the user BIST This testing theory is used for testing our non intelligent VMEbus I O boards too 12 The test programs can run in two modes in workshop test mode and in BIST mode There are three built in macros that can be executed from a test program Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP 31 4 SPRA326 Figure 13 Sample Test Program 32 write read compare branch HALT SCOPE test check branch HALT SCOPE ENDTST if if 1 This is a short sample test paragraph Two hardware related checking is made eg a register read and write Each of them can be forced in a loop with the SCOPE macro First SCOPE loop is between labell and ok1 labels the second one is between label2 and ok2 labels The program is executed in an endless loop if it is enabled in the test Debugger The instructions between the conditional branch instruction and the SCOPE macro are executed in the case only when an error is detected The last instruction in the test program is an ENDTST macro call Using these macros the hardware can be tested efficiently For example if an error detected in loop 2 the program execution stops at the second HALT macro and displays the error number and messages Using an oscilloscope the user can
5. circuit TMS320C25 DSP Bus interface PWM amp WD D A converter A D converter Signal generation control circuit control circuit circuit The internal architecture of the XILINX chip is sown in Figure 9 The DSP bus interface circuit receive the control data and address signals from the DSP and together with the address decoding circuit doing the write and read tasks of the chip internal registers The internal registers of the XILINX circuit are enumerated in Table 1 24 Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP X SPRA326 Table 1 XILINX Circuit Internal Register Values NAME ADDRESS ACCESS MOD PWMA F800h RW PWMB F801h RW PWMC F802h RW SETWD F803h WR DACCTRL F804h RW DADATA F805h WR ADDATA F805h RD ADCSTART F806h WR The PWMA PWMB and PWMC registers contain three data defining the actual PWM values The dates from the registers are written in the same cycle in the internal readable registers The data written into the SETWD register define the status of the watchdog signal The DACCTRL register deserve for setting of desired channel as a chanel which can be write and for setting the functionality of the 4 channel D A converter In the here selected channel or the selected ones the data is written to the DADATA register By writing in the ADCSTART register the circuit measures the 6 channels of the A D converter and with the help of an inter
6. dnce dei 7 Product Support on the World Wide Web eee 8 oligo ls CUNO eM 9 Fuzzy Logic Gontrollgr ehe ce nto cde teca ax iiaee te sese laxare 10 Pre Processing and FUzziflCatlODi esce socie roter rex mex erede abs daa eeu 10 iai esame CU 12 Defuzziflcation and POSIEPrOGCOSSIDQ uctor dean he cee Saiba 13 Application to an Induction Motor Drive eeeeeerereneeeeeennnnnnn 14 Asynchronous Motor Modeling 1eeeeeeeeeeeeee eene 15 Vector Control of the Asynchronous Motor eene 16 Description of Experimental System cessere nennen nnns 17 Realized Hardware Setup in the Development Stage 20 The DSP Single Board Computer cccsseeeceseseeeeeeeseeneeseessneeeeeeseeeeeeeeseeeaeeeseseeeeeenss 21 General DESCripuOn ene 21 Features of the Board sussa eer oe bobo RSS eSc uae qai tcs Meets 21 iuo 22 MEMO ECTS 22 Eoee lke s 22 DVIS AY c MC al pas Ed aC EDS aa elas 22 Serial OPUS REERERNM MON Cc Oc 24 XILINX VO Internac S e RO HE 24 Memory m Grenioie uo mem 25 Burned In Firmware acd aieo trei e RO vbt ise Dead AA SR ea dare 29 Physical SOLID sans nesses DORES p e Ub p MEE RS eee 29 DSP DE DUGG sects oat Tp HT 29 Main Features of the Debugger ec ir De neebe eiecit de 29
7. vector in the stationary reference frame reference value of the flux d q axis of the reference system fixed to the rotor flux rotor flux angular velocity position of the flux from the fixed slip position of the stator current vector from the fixed reference axis measured angular speed of the rotor slip frequency stator current vector angular speed reference value of the angular velocity 36 Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP X SPRA326 Ea l a reference values of isa Isq l a the new reference value of the flux generating current Relevant Literature 1 TMS320C25 User s Guide Texas Instruments 2 MULTIBUS Il Bus Architecture Specification Handbook Intel Order No 146077 3 MULTIBUS II Transport Protocol Specification and Designer s Guide Intel Order No 453508 4 ANSI IEEE Std1014 1987 IEEE Standard for A Versatile Bus VMEbus 6 MBRT 4 05 01 0 8 channel Sample amp Hold Input Analog to Digital Converter Operating Manual KFKI MSZKI July 1992 7 MBRT 4 10 01 0 8 channel Digital to Analog Converter Operating Manual KFKI MSZKI July 1993 8 ERT 2 01 01 0 Quad Scaler User s Manual KEKI MSZKI 1992 9 ERT 2 01 01 0 Quad Scaler Programmers Guide KFKI MSZKI 1992 10 ERT 2 23 01 0 Status Change Monitor Users Manual KFKI MSZKI 1992 11 TMS320 Family Development Support Reference Guide Texas Instruments 12 M Barkaszi Z Katona UBI Speci
8. where the rotor leakage impedance is included into the stator transient impedance L U I R eL L jo L I J 5 5 sm sd After reducing the motor s physical parameters and transforming the equations mathematically we obtained the stator voltage vector equation 15 Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP 15 4 SPRA326 Vector Control of the Asynchronous Motor The field oriented theory 2 5 6 is the base of a special control method for induction motor drives With this control method induction motors can successfully replace the expensive DC motors Nowadays this method has become general in induction motor drives of high precision The main advantages of induction motors are their simplicity and price as well as greater reliability especially in harsh industrial environments Induction motors require very complex control algorithms because there is no linear relationship between the stator current and either the torque or the flux This means that it is difficult to control the speed or the torque because of the transients until the motor reaches its new stationary state The problem can be solved by controlling the rotor flux since it cannot be measured only computed The purpose of he controller is to keep the amplitude of the rotor flux at a constant value so the only its direction is changed The field oriented theory offers a suitable method for optimally control of the inud
9. D TO BE SUITABLE FOR USE IN LIFE SUPPORT APPLICATIONS DEVICES OR SYSTEMS OR OTHER CRITICAL APPLICATIONS Inclusion of TI products in such applications is understood to be fully at the risk of the customer Use of TI products in such applications requires the written approval of an appropriate TI officer Questions concerning potential risk applications should be directed to TI through a local SC sales office In order to minimize risks associated with the customer s applications adequate design and operating safeguards should be provided by the customer to minimize inherent or procedural hazards TI assumes no liability for applications assistance customer product design software performance or infringement of patents or services described herein Nor does TI warrant or represent that any license either express or implied is granted under any patent right copyright mask work right or other intellectual property right of TI covering or relating to any combination machine or process in which such semiconductor products or services might be or are used Copyright O 1997 Texas Instruments Incorporated TRADEMARKS Tlis a trademark of Texas Instruments Incorporated Other brands and names are the property of their respective owners CONTACT INFORMATION US TMS320 HOTLINE US TMS320 FAX US TMS320 BBS US TMS320 email 281 274 2320 281 274 2324 281 274 2323 dsph ti com Contents PRD SU ACU asia a ciam teer a qu t t a O a
10. Disclaimer This document was part of the First European DSP Education and Research Conference lt may have been written by someone whose native language is not English Tl assumes no liability for the quality of writing and or the accuracy of the information contained herein Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP Authors D Fodor Jozsef Vass Z Katona ESIEE Paris September 1996 SPRA326 vis TEXAS INSTRUMENTS IMPORTANT NOTICE Texas Instruments TITM reserves the right to make changes to its products or to discontinue any semiconductor product or service without notice and advises its customers to obtain the latest version of relevant information to verify before placing orders that the information being relied on is current TI warrants performance of its semiconductor products and related software to the specifications applicable at the time of sale in accordance with TI s standard warranty Testing and other quality control techniques are utilized to the extent TI deems necessary to support this warranty Specific testing of all parameters of each device is not necessarily performed except those mandated by government requirements Certain application using semiconductor products may involve potential risks of death personal injury or severe property or environmental damage Critical Applications TI SEMICONDUCTOR PRODUCTS ARE NOT DESIGNED INTENDED AUTHORIZED OR WARRANTE
11. PROM s After these steps the monitor software is started that aids in the downloading and debugging of the software via the serial interface The board size 155 mm x 155 mm The mechanical placement of connectors for different type of I O devices are designed in the way that they can be placed bellow of the DSP board TI offers a wide range of development tools for the TMS320C25 DSP 1 including an evaluation module EVM a software development system SWDS and an emulator XDS 22 Atthe beginning we used an SWDS for software development This is a useful tool if no hardware related instructions are used but the handling of memory mapped I O chips or circuitry is not possible because the I O address range of the target system can be accessed through the emulator cable only So we developed a program package to provide a debugging tool for these I O devices DSP Debugger The DSP Debugger can be used to debug programs written in TMS320C25 assembly language in the target system The Debugger consists of two parts one of them runs on an IBM PC as a user interface while the other runs on the target system burnt in The IBM PC can be used for software development assembler linker also The communications between the two parts is realized via a serial cable Either the DSP s serial port can be used in this case a special serial card is required for the PC or the RS 232C serial connection can be used if a serial chip UART and lin
12. SP SPRAS26 Figure 10 Memory Configuration without EPROM Program memory Data memory 0000h 0000h Internal data 80r32k 0400h First 8k block RAM chips 80r32k 600k Sea 2000h RAM chips e block BANE CMDS RAM chips 4000h um 4000h L J j k t 32k dcus chips RAM chips 6000h 4 6000h IRE k 32k F i sa bloc chips RAM chips 8000h 8000h ies A000h AO00h l coooh C000h 4 E000h BOOG sda i F000h On board HW FFFFh BEE FFFFh ETO Sa Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP 27 X SPRA326 Figure 11 Input Output Address Region Map F000h F800h FAO0h FC00h I O devices pm mo FEOOh e Slow I O devices FFFFh NRI AS ae ed Table 2 Input Output Device Addresses fcOOh NVRAM non volatile read or write fcO1h KEYBOARD register read fcO2h DISPLAY data register fcO3h DISPLAY control register fe00h Serial chip 28 Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP SPRAS26 4 Burned In Firmware Physical Set Up DSP Debugger The board burned on firmware contain the basic programs After Rreset the programme check the RAM memory on the board copy the contain of EPROM into the RAM without changing the programme memory address after turning off the E
13. ample Test P OG STI doute teni a ta Rd 32 Characteristic Control System Results 34 Tables XILINX Circuit Internal Register Values seeeeeeseeeeneneeeee 25 Input Output Device Addresses eeesessseeeeneeenennne 28 Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP Abstract The field oriented theory is the base of a special control method for induction motor drives With this control method induction motors can successfully replace expensive DC motors Induction motors require complex control algorithms because there is no linear relationship between the stator current and either the torque or the flux This means that it is difficult to control the speed or the torque because of the transients until the motor reaches its new stationary state The problem can be solved by controlling the rotor flux since it cannot be measured only computed Because of the complexity and nonlinearity of control equations it is useful to implement a part of the control system by a fuzzy logic controller By using linguistic variables in place of numerical variables that approach represents a substantive departure from the conventional quantitative techniques of system analysis and control In the present paper we use fuzzy logic based speed control for field oriented AC motor For implementing the control algorithm we developed a digital signal processor DSP based sing
14. ant current Of course if the stator voltage changes the reference current vector must be recalculated with this new value The resulting stator voltage vector is rotated back and transformed and used to generate the PWM values Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP X SPRA326 The program uses 16 bit integer arithmetic and was written primarily in C language Lookup tables are used for limiting the stator voltage for storing the trigonometric values to co ordinate rotation Lookup table was used too to storing the control information for the fuzzy controller In the realized solution of the algorithm the control system works on a small power motor using a sampling period of 1000 us Although the time constant of the motor is small the slow sampling rate gives a satisfactory result The dynamic behavior of the algorithm speed reversal torque changing is good despite some torque fluctuation during the speed reversal Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP 19 X SPRA326 Realized Hardware Setup in the Development Stage The field oriented control drive consists of an electronic control inverter current sensors and speed sensor Figure 6 An inverter was developed with power MOSFETs which is suitable for field oriented control of the motor The motor used in our experiments is a simple three phase squirrel cage induction machine 380 220 V 370 W The PWM switchi
15. ard In this way it is suitable for an internal controller task where the high computing capacity of DSP is required The block diagram of the board are shown in Figure 7 Figure 7 Block Diagram of the TI TMS320C25 DSP 7 segment LED display and LEDs XILINX interface Features of the Board m m Texas Instruments TMS320C25 Digital Signal Processor 40 MHz clock signal Memory configuration is based upon the necessities 8 or 32 kword EPROM 3 wait state 8 or 32 kword static RAM 0 or 1 wait state 2 Kbytes nonvolative memory 1 wait state RS 232C serial port Keyboard consisting 6 tastes which can be controlled by burned in software 4 pieces of 7 segments LED display and 4 LED s controlled by the same burned in internal software Configurable XILINX FPGA interface circuit for different I O devices dedicated mostly for control and measurement device connectivity Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP 21 4 The Processor Memory Keyboard Displays SPRA326 The Texas Instruments TMS320C25 16 bits fixed point Digital Signal Processor DSP with 40 MHz clock rate give rise for a 100 ns instruction cycle execution time An internal timer and external interrupt connection are available Considering the internal structure of the processor it was built with Harvard architecture which means the program and data memory phisically and logically are s
16. better controlled output The fuzzification relates the input values to the fuzzy sets Xi X which will be composed with the fuzzy algorithm according to compositional rules of inference shown in Figure 2 All the designers agree that the numerical values may be well represented by fuzzy singletons the membership functions of which are equal to zero everywhere except at the measured values x and xz and where they equal one This choice greatly simplifies the expression of the composition rule of inference Figure 2 Compositional Rules of Inference NEM NS AZ PS PMP SEMANTICS RULE BASE y X1 NL NM NS AZ PS PM PL Fy Ga a Lal ad pt Je eu em ps az Ns pu et fem rs az s nm e femp oz no jr if x jis NS and x is NM then y is PL Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP 11 4 Figure 3 MAMDANI Controller E EU 1 SPRAS26 Fu ier singleton i Fuzzy inference engine also and A NGS Defuzzyfier Center of gravity Mean of Maxima etc Inference The second block in Figure 3 constitutes the fuzzy algorithm According to the logical or which joins together the rules the fuzzy output set C is established by the union of all the fuzzy sets inferred by each rule C C C2 C3 12 Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP X SPRA326 Defuzzification and Post Processing The input of the control system is a sin
17. cation may be also used Since the universes of the fuzzy sets representing the linguistic values of the input variables are finite limitation is usually performed In addition if the universes are discrete a quantification concludes the pre processing delivering the input numerical values The choice of the shape of the fuzzy sets is another important aspect in designing FLC Triangular fuzzy sets are often utilised but trapezoidal exponential or monotone forms have been experimented A general criterion in choosing the form of fuzzy set related to the input variables is that they must cover the whole universe X with some overlapping between adjacent fuzzy sets in order to avoid high discontinuities of the control action as a consequence of a small change in the input variables This is of particular importance in Flocs for motor drives because they usually exhibit small time constant 10 Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP SPRAS26 4 In this work the compositional rule of inference used to relate uc to La He and uris is defined by min max composition He Z max min ua X UB Y Ur X Y Z XY However there are also different definitions For instance those based on the max product model use the algebraic product in place of the min operator for the logical and operational and or the implication and or the inference with the advantage in some cases to produce a smoother and
18. e The TMS320C25 16 bit fixed point Digital Signal Processor used in the control algorithm implementation gives a relatively long running time Consequently the program takes into account the dead times caused by the current filters and program running Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP 17 SPRAS26 The algorithm implements indirect field oriented control because the rotor flux is not computed exactly The flux is considered a constant by the algorithm in the rotating d q co ordinate system and the currents are controlled according to this condition would be true Knowing the maximum output stator voltage limited by the inverter the program modifies the active current l if necessary The passive current la remains invariable so the flux remains invariable too Form the measured currents isa is isc after co ordinate transformation and rotation the control program estimates the current vector Istator at the beginning of the control cycle taking into account the current changes in the previous two cycles dead time compensation This estimation is essential because of the significant difference between the measured currents and the real currents at the beginning of the next cycle The first reason of it is the delay caused by the measurement of stator currents Approximately a 150 ms delays is caused by the filters and the A D conversion takes some time too The second reason is that the stator c
19. e speed from 3000 min to 3000 min The speed the reference speed and one phase current are shown On the left side the results of PI type controllers are shown while on the right side the results of fuzzy type controllers There is 50 ms between two ticks on the time axes The bottom two figures show the dynamic behaviour of the motor with linear reference from 0 to 3000 min The speed the reference speed and one phase current are shown The drive shows good dynamic behaviour Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP 33 X SPRA326 Figure 14 Characteristic Control System Results 6000 6000 4000 4 4000 lt i E 2000 1 amp 2000 E E 0 E E E 2000 E i 2000 Time 300 msec aoo i sa H 4000 6000 t 6000 8000 6000 4000 2000 0 2000 4000 6000 8000 5000 5000 7 4000 4000 3000 300 4 endi g 007 1000 E 1000 3 0 1000 100 1 2000 2000 wy 3000 H 3000 4000 Time 300 msec 4000 jl 5000 5000 34 Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP X SPRA326 Conclusions The availability of low cost microelectronics means that the field oriented theory can be widely used in industrial applications The software and hardware configurations are presented with experimental results in case of a conventional PI type controller and a
20. e transceivers are mounted on the target board Main Features of the Debugger 1 Programs can be downloaded from the IBM PC into the target board s memory After assembling the source code and linking the object codes the program is converted to Intel hex file format All of the required software development tools are available from Texas Instruments If the target board contains no writeable program memory the internal program memory block 0 can be used In this case the downloaded program can be as long as 256 words Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP 29 SPRAS26 Program starting break point setting and step by step execution of commands are available in the Debugger Break point setting and step by step executing can be done only if the debugged program is loaded into a writeable program memory area The contents of program and data memory and the DSP s on chip registers can be examined and modified by menu driven commands The Debugger can read and write the DSP s input output ports There are some special functions and commands for testing the target board s hardware components The communication between the IBM PC and the DSP target board uses one character control commands and small packets based on Intel hex format This simple communication method requires a small number of packet types only Figure 12 IBM PC DSP Target Board Communication 30 Implementing Field
21. easily find where the hardware error happened HALT macro A test program calls this macro if an error is detected In workshop test mode the Debugger prints an error message to the IBM PC display with a unique error identification number and some parameters The test program is suspended and the Debugger takes over the control so the user can examine the contents of the memory registers etc The printing of error messages and suspending the test program can be disabled by debugger commands As no user intervention is expected in BIST mode the testing is suspended and the error status is signaled via the BIST status SCOPE macro In workshop test mode the SCOPE macro is used to force the test program to execute a test loop For example a register read write verify cycle can be executed in a loop so the hardware can be checked with an oscilloscope In BIST mode this macro is ignored ENDTST macro This is the last instruction executed by the test program In workshop test mode the Debugger takes over the control In BIST mode the test handler starts the next test paragraph Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP X SPRA326 Experimental Results Figure 14 shows some characteristic results of the control system for different types of controllers Pl fuzzy and different reference speeds The upper two figures show the signal response of the control system to a step change in the referenc
22. eparated In this way they can be accessed parallely in the same time allowing a quick instruction execution time In the case of external memory the logical separation exists but not the phisically not only one address and data bus exist and parallel memory access is not possible On the board there are three kinds of memory EPROM SRAM and NVRAM The EPROM memory consists of 2 pieces of 8 Kbytes or 2 pieces of 32 Kbytes chips This memory contain the program code and the initialised data The RAM memory which can be used as program or data memory is consists of 2 pieces of 8 Kbytes or 2 pieces of 32 Kbytes chips The NVRAM which memory preserve the date after the turn off is consisted by one pieces of 2 Kbytes chip There are 6 pieces of Hall generator type tastemounted on the board The status of these keys can be read by the processor via a dedicated register One can find 4 pieces of 7 segments LED s display and 4 traditional LED s on the board The processor accesses the displays via two registers and displays hexadecimal numbers In the first one must be written the values of the desired numbers while in the second controls the LED s operation for fraction dots and the validation of the numbers displayed by indicators 22 Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP SPRAS26 Figure 8 Memory Configuration after Reset 0000h 2000h 4000h 6000h 8000h A000h C000h E000h FFFF
23. fic Test Software Environment XV International Symposium on Nuclear Electronics and International Seminar CAMAC 92 Warsaw 29 September 2 October 1992 pp 204 222 13 IBM PC AT VME Adaptor Model 403 B1T3 Computer Corporation pub no 8200301 0 14 Ben Brahim L Kawamura A A Fully Digitized Field Oriented Controlled Induction Motor Drive Using Only Current Sensors IEEE Transactions on Industrial Electronics Vol 39 No 3 June 1992 pp 241 249 15 Blaschke F The Principle of Field Orientation as Applied to the New TRANSVECTOR Closed Loop Control System for Rotating Field Machines Siemens Rev 1972 pp 217 226 16 Hainemann G Leonard W Self Tuning Field Oriented Control of an Induction Motor Drive International Power Electronics Conference Tokyo 1990 pp 465 472 14 Halasz S Automatizalt Villamos Hajtasok Tankonyvkiad Budapest 1989 15 Kelemen A Imecs M Vector Control of AC Drives OMIKK Publisher Budapest 1991 Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP 37 38 SPRAS26 16 Leonard W Control of electrical drives Springer Verlag Berlin 1985 18 Lessmeier R Schumacher W Leonard W Microprocessor Controlled AC Servo Drives with Synchronous or Induction Motors Which is Preferable IEEE Transactions on Industry Applications Vol IA 22 No 5 Sept Oct 1986 pp 812 819 19 Rodriguez J Kastner 0 Non linear current cont
24. fuzzy type controller The FLC can be successfully used in the field of induction motor control due to the robustness of the controller 1 This solution can be used in variable speed servo drives A low cost DSP was used 2 The algorithm considers the dead time caused by the current filters and measurements and the program running time 3 No external hardware current control loop is required the program calculates the stator voltages corresponding to the necessary currents of the motor 4 If the calculated stator voltage is greater than the output voltage of the inverter the program limits only the q axis component of the stator voltage if necessary 5 The simulation and experimental results show good dynamic behaviour in both cases Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP 35 Symbols Usa Usq Yra Yra Rs Ri R Ls L Ln m lsd Isq los Ip Wret d Ar q Or Oslip Os Orer SPRAS26 stator rotor voltage vector stator rotor current vector stator rotor flux vector components of the stator voltage in the d q frame of reference the d q frame of reference stator and rotor resistance stator and rotor resistance equivalent resistance of the motor stator rotor and magnetizing inductance respectively total leakage factor moment of inertia load torque components of the stator current vector in the d q frame of reference components of the stator current
25. gle numerical value and therefore the defuzzification block extracts from C the numerical value y which may be considered the best representative element of the fuzzy output set Alternatively the entire form of C which would mean the contribution of all rules is taken into account This is obtained by the center of gravity method which consists of averaging all the elements of C weighted with their own membership grade A scaling factor is applied to the numerical value y in the post processing part of the block to obtain the output variable y Finally this is manipulated to generate the control command to the plant Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP 13 4 SPRA326 Application to an Induction Motor Drive We make different measurements with PI speed controller and FLCs For example a signal response of the control system under the same working condition to a step change in the reference speed from 3000 min to 3000 min in the case of the PI speed controller is about 160 ms and in the case of FLC is about 80 ms This means a better dynamic behaviour in the case of fuzzy control but because a small look up table was used a better reference speed accuracy is achieved by PI type control The speed overshoot has disappeared and the current swings have been significantly reduced as well as the effect on the speed of the load torque disturbance The robustness of the FLC is recognised
26. h Program memory 80r32k EPROM chips SEE Ged 32 k EPROM chips pee ete 0000h 0400h 2000h 4000h 6000h 8000h A000h C000h E000h F000h FFFFh Data memory 1 Internal data Second 8k block 32k RAM chips Third 8k block 32k RAM chips 4 Fourth 8k block 32k RAM chips First 8k block 80r32k RAM chips a Ee C On board HW Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP 23 X SPRA326 Serial Ports The board has two serial ports one of them is the DSP own serial port and the other one is a standard RS 232C serial interface XILINX I O Interface This interface is dedicated for connection to different types of I O devices mainly measurement and control devices With the reconfiguration of the XILINX FPGA circuit the board can be easily adapted for a new task In the present configuration the board is set up for the control of a vector controlled AC motor That means that the XILINX FPGA chip is configured to control an ERT 4 05 01 0 type 8 channel analogue digital converter board and a 4 channel digital analogue ERT 4 10 01 0 converter board More over the XILINX chip is responsible for the generation of the PWM Pulse Width Modulation and watchdog signals Figure 9 Internal Architecture of the XILINX I O Interface Address decoding
27. le board controller based on the Texas Instruments TIM TMS320C25 DSP The board contains a configurable digital interface which can be used to extend the functionality of the board by connecting analog and digital I O peripherals It has a flexible memory subsystem with EPROM SRAM and nonvolatile SRAM support an RS 232C serial port a keypad with 6 keys and four 7 segments LED display In this paper we present the state of the art of field oriented control of AC motors and fuzzy logic controllers We describe our control system from both software and hardware point of view The experimental results have been compared by in the case of traditional P1 type speed control and shows good dynamical behavior Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP 7 X SPRA326 This document was part of the first European DSP Education and Research Conference that took place September 26 and 27 1996 in Paris For information on how TI encourages students from around the world to find innovative ways to use DSPs see TI s World Wide Web site at www ti com Product Support on the World Wide Web Our World Wide Web site at www ti com contains the most up to date product information revisions and additions Users registering with TI amp ME can build custom information pages and receive new product updates automatically via email 8 Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP SPRAS26 Introduc
28. ng frequency is 10 kHz Galvanic separation and stator current detection are realised through Hall type sensors Stator currents are filtered by passive LC low pass filters and sampled and held at every sampling instant then converted with 12 bit accuracy Figure 6 Hardware Block Scheme MULTIBUS II The experimental system was built in a MULTIBUS II crate In the crate there were two DSP boards based on TMS320C25 processor to control and compute the algorithm and a development system based on an Intel 80386 processor The first DSP generates the PWM pulses and measures and pre processes the stator currents the speed and the line voltage The second DSP computes the field oriented algorithm and handles a 4 channel D A converter to visualise the selected variables during development The reference signals and the process parameters of the control algorithm are interactively controlled from the development system Because the development system presented above is too expensive for a real industrial application we developed and executed a single board TMS320C25 DSP based computer system 20 Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP SPRAS26 The DSP Single Board Computer General Description The DSPSBC board based on Texas Instrument s TMS320C25 Digital Signal Processor DSP According to the configurable interface different kind of measurement and control I O instruments can be connected to the bo
29. rol of an inverter fed induction machine Etz Archive Bd 9 1987 H 8 pp 245 250 21 Vainio 0 Ovasaka S P A Digital Signal Processing Approach to Real Time AC Motor Modeling IEEE Transactions on Industrial Electronics Vol 39 No 1 February 1992 pp 36 45 22 0 Klir and Tina A Folger Fuzzy Sets Uncertainty and Information Prentice Hall 1988 23 H J Zimmermann Fuzzy Set Theory and its Applications Kluwer Academic Publishers Boston Dordrecht London 1990 24 D Naunin C Karaali Fuzzy Controller and Digital Cascaded State Controller for an Intelligent Synchronous Servodrive Proceedings of IECON 94 Conference Vol Il pp 1298 1303 25 D Fodor Z Katona and E Szesztay Digitized Vector Control of Induction Motor with DSP Proceedings of IECON 94 Conference Vol III pp 2057 2062 26 B Kosko Neural Networks and Fuzzy Systems Prentice Hall 1992 27 D Fodor Z Katona E Szesztay Field Oriented Control of Induction Motors Using DSP Computing amp Control Engineering Journal Vol 5 no 2 pp 61 65 Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP
30. rupt signal inform the DSP processor of the channel s measurement termination The data s stored in the internal FIFO can be read from the ADDATA register Memory Configuration The memory map of the processor can be viewed in Figure 10 The memory state after the reset can be viewed in Figure 7 while the EPROM chips state after power off can be viewed in the Figure 10 Because of the low speed of the EPROM memories is recommended to copy the contents into the RAM memory running the programs from here The board allow this operation through the turn off possibility of EPROM s when the RAM memory serve both as program and data memory It was previously mentioned before in the processor description that the external memories can not be accessed parallely in the same time For this reason after removing power off from the EPROM s chips the physical separation of program and data memory can not take place This mean that with the same logical address in the programme and data memory physically the same memory is addressed This memory configuration problem must be taken into account in the software linking Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP 25 26 SPRAS26 The I O devices on the board can be accessed via data memory address The F000h FFFFh address region map is shown in Figure 11 The I O devices address can be seen in Table 2 Implementing Field Oriented Control of AC Motors with the TMS320C25 D
31. the current filters and measurement and the program running time In the P1 controllers we gain a program running time about 1000 us In the case of fuzzy logic control the inference and composition block together with the fuzzification and defuzzification blocks have been implemented by means of a look up table The content of the table is off line computed and thus it is greatly reduced the demand to the digital system of on line computation This allows the control cycle to be as short as required by field oriented AC motor drive application A disadvantage of the look up table technique is the low resolution in the input and output variables unless a large and memory consuming table is used Xa SPRA326 Fuzzy Logic Controller The basic configuration of a FLC with three linguistic variables two inputs and one output is shown in Figure 1 The implementation of FLC needs a digital hardware in our case a TMS320C25 fixed point DSP From the Figure 1 three main blocks can be distinguish whose function is hereafter briefly explained Some further details will be given in the next Section Figure 1 Fuzzy Logic Controller Knowledge Base Rule Base Fuzzy Inference Engine Pre Processing and Fuzzification This block receives the reference and feedback signals and determines the variables x and x which are chosen in linguistic form like antecedents by the control rules Same linear and non linear scaling factors amplifi
32. tion Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP 4 By using linguistic variables in place of numerical variables that approach represents a substantive departure from the conventional quantitative techniques of system analysis and control The fuzzy set theory gives the tools to represent and manipulate the linguistic variables This approach provides an effective mean for describing systems which are too complex or too ill defined to admit a precise mathematical model Fuzzy set theory fundamentals have been investigated in detail and applications have been proposed in economics artificial intelligence information retrieval diagnostics and so on Fuzzy Logic Control is one of the most interesting fields to which the fuzzy theory can effectively applied Recently some applications of FLCs to motor drives have been also reported In our cases an asynchronous motor has been chosen as a bench to explore the design of a FLC drive system and to investigate by simulation and experiment its performance The drive is preliminary simulated with conventional digital P1 speed regulator in order to establish a term of comparison The current control is performed according to an instantaneous voltage equation of d q model of the asynchronous motor The implemented drive is fully digitised the control algorithm is supported by TMS320C25 fixed point digital signal processor The program considers the dead time caused by
33. uction motors The complexity of this method is compensated by its advantages The most often used method is the one with the rotor flux orientation because of the simple structure of the control loops and command variable calculation The space phasor of the stator current is split into two components which become control variables Vector rotation techniques are used to transform three phase axes into rotating two phase d q axes This two phase rotation technique greatly simplifies the analysis making it equivalent to analyzing separately excited DC motors because in this case there are two independently controllable currents the field current and the armature current Implementing Field Oriented Control of AC Motors with the TMS320C25 DSP X SPRA326 Description of Experimental System Figure 5 shows a block diagram of the control system for a vector controlled induction motor in a velocity loop using a fuzzy controller Figure 5 Control System Block Diagram irj E POWER ELECTRONICS AC line RECTIFIER MOSTE EWM ae nu pu Ta m C 14 cO 3 om Speed Sensing ik E E 4 The program calculates the voltages corresponding to the necessary current of the motor All calculations in the block diagram the co ordinate transformation and rotation fuzzy controller implementation control calculation and so on are done by softwar
34. urrents are changing during the running time of the program The active reference current I is generated from the angular velocity error signal by a PI Proportional Integral controller The passive reference current I y can be controlled by the velocity in the field weakening region taking into account the time constant of the motor The difference between the estimated stator current and the reference current gives the current vector dl As the real current vector is behind the reference current because the current reaches its reference value at the end of the cycle so its average value differs from the required one and the program modifies turns forward the reference current vector half step forward modification to generate the corresponding value The stator voltage Ustator is calculated so that the current reaches its reference value in one step The calculation is based on a simple physical single phased equivalent model of the induction motor This model can be used because the running time of the program is far less than the rotor s time constant so the rotor flux can be considered a constant When the calculated stator voltage is greater than the output voltage of the inverter the corresponding current vector will not be as expected consequently the flux will not be invariable Therefore the program limits the stator voltage Only the q axis component of the stator voltage is modified to keep the flux generating const
Download Pdf Manuals
Related Search