Sensorless Control of Electric Drives


Published on

Achieving near perfect torque/speed control of an electrical machine without the use of external shaft sensors is readily achievable at low cost with the latest Texas Instruments™ microprocessors. This presentation examines the challenges associated with achieving this goal and considers a solution based on the Texas Instruments™, ‘InstaSPIN™’ concept which is universally applicable to three-phase PM and Induction machines. A brief outline of this approach will be given where used is made of VisSim™, which is an embedded software control environment that gives the user the ability to develop comprehensive motor drive firmware, without having to be C-code literate.

The ‘InstaSPIN™’ solution has been applied worldwide, to a range of application which includes: washing machines, electrical bicycles, electric vehicles. Furthermore, efficient induction machine based applications for electric fans have been developed to date. Further details on some of the products developed will be discussed in the presentation.

Published in: Engineering, Business, Technology
  • Be the first to comment

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Sensorless Control of Electric Drives

  1. 1. 1 Univ.-Prof. Dr. ir. D.W.J. Pulle RWTH-ISEA Germany CEO EMsynergy, Australia European Altair Technology Conference June 24-26 2014, Munich Germany Sensorless control of Electrical Drives
  2. 2. ■ Introduction □ what is sensorless control? □ Why use it? ■ Challenges to realise sensorless control □ Ability to track the magnetic field in the machine □ Need to identify and track critical motor parameters □ Measurement of the induced voltage ■ Solution □ use of TI InstaSPIN algorithm □ Use of VisSim embedded control tools ■ Application examples ■ Questions ? Content 2Prof. D.W.J. Pulle
  3. 3. 3Prof. D.W.J. Pulle  Schematic representation electrical drive ? Motor Power Electronic Converter with controller and micro-processing unit (MCU) Introduction: typical electrical drive Front wheel drive-train Mercedes SLS Electric vehicle
  4. 4. 4Prof. D.W.J. Pulle  What is 'Field Oriented Control' (FOC) ? + DC Bus Micro Controller Unit (MCU) Timers and PWM Compare Units Capture Unit ADC Serial coms (UART) SPI Serial coms Encoder PWM1 PWM2 PWM3 PWM4 PWM5 PWM6 PWM1 PWM2 PWM3 PWM4 PWM5 PWM6 Motor Power Electronic Converter - Introduction: typical electrical drive
  5. 5. ■ Gives highest torque for the lowest current ■ High dynamic response fully equivalent to DC motor ■ Requires control of the currents in the stator by using the converter ■ Requires knowledge of the PM magnetic field orientation Introduction: What is 'Field Oriented control' ? 5Prof. D.W.J. Pulle A` B C` A B` C N S 90° F F Maintain the angle between stator field and PM field at 90° Principle of 'Field Oriented Control' (FOC) for PM [1], [2] N S Magnetic field due to phase currents Permanent magnet rotor
  6. 6. ■ Shaft encoder: reliable position information, but expensive , increases drive complexity and requires an additional cable to MCU ■ Encoder: requires access to a motor shaft end and increases motor volume ■ EMF sensing required sohisticated algorithm that works at near zero speed Introduction: How can we find rotor position? 6Prof. D.W.J. Pulle A` B C` A B` C N S Use of a shaft encoder S Encoder A` B C` A B` C N S - + Induced voltage in phase windings due to magnet(EMF) Use EMF induced by rotor magnets
  7. 7. ■ EMF amplitude and direction allows (in principle) calculation of the magnetic field amplitude and direction using □ 𝐹𝐹𝐹𝐹 = 𝐸𝐸𝐸 (𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 𝑎𝑎𝑎 𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑) 𝑠𝑠𝑠𝑠𝑠 𝑠𝑠𝑠𝑠𝑠 ■ EMF amplitude proportional to shaft speed □ lower shaft speed means lower EMF voltage amplitude Introduction: How can we find rotor position? 7Prof. D.W.J. Pulle A` B C` A B` C N S - + Induced voltage in phase windings due to magnet(EMF) EMF induced by rotor magnets in stator phase windings Shaft speed
  8. 8. ■ Need to find EMF 'vector' in amplitude and orientation using □ 𝐹𝐹𝐹𝐹 = 𝐸𝐸𝐸 (𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 𝑎𝑎𝑎 𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑) 𝑠𝑠𝑠𝑠𝑠 𝑠𝑠𝑠𝑠𝑠 □ 𝐸𝐸𝐸 ∝ (𝑓𝑓𝑓𝑓 x 𝑠𝑠𝑠𝑠𝑠 𝑠𝑠𝑠𝑠𝑠) ■ Calculating the flux from EMF means solving the equation 𝐹𝐹𝐹𝐹 = 𝐸𝐸𝐸 ∝ (𝑓𝑓𝑓𝑓 x 𝑠𝑠𝑠𝑠𝑠 𝑠𝑠𝑠𝑠𝑠) 𝑠𝑠𝑠𝑠𝑠 𝑠𝑠𝑠𝑠𝑠 equation (1) Challenges to achieving sensorless control 8Prof. D.W.J. Pulle  Complication: as speed reduces both numerator and denominator terms of equation (1) reduce to zero
  9. 9. ■ Need to find EMF 'vector' in amplitude and orientation using □ 𝐸𝐸𝐸 = 𝑉𝑚𝑚𝑚𝑚𝑚 − 𝑉𝑅 − 𝑉𝐿 □ Approach requires knowledge of the motor parameters R, L and currents Challenges to achieving sensorless control 9Prof. D.W.J. Pulle L  Complication: motor voltage (𝑉𝑚𝑚𝑚𝑚𝑚) must be reconstructed from converter voltage (V_converter) which has a value V_DC or 0 V + − - + Converter R EMF n V_converter V_L + - + V_R - + - DC Motor
  10. 10. ■ Estimate the rotor flux in terms of amplitude AND orientation □ Undertake this at near zero speed and remain STABLE at zero speed □ Estimate the motor parameters in order to 'reconstruct' the EMF of the motor □ Be used universally for all three-phase machines (PM and INDUCTION ) □ Accurate measurement of the motor voltages and currents ■ Require a software package than can communicate with the algorithm at a 'high level' so that inexperienced users can use this technology Challenges to achieving sensorless control 10Prof. D.W.J. Pulle  Solution??
  11. 11. ■ Use of Texas Instruments InstaSPIN™-FOC □ A new advanced field oriented control technique for sensorless control of permanent magnet (salient and non-salient) and Induction motors □ Comprehensive self- commissioning capabilities □ On-line estimation of key variables □ Relatively easy to use by inexperienced users ■ What are the components ? □ FAST algorithm that provides:  Flux amplitude , Angle and Speed of the flux vector and machine Torque □ Motor ID:  Identification of motor parameters □ EPL (PowerWarp) for energy efficient induction machine operation under partial load Solution to realizing sensorless control 11Prof. D.W.J. Pulle [3]  System architecture using InstaSPIN-FOC? InstaSPIN- FOC FAST Motor ID EPL InstaSPIN- FOC FAST Motor ID EPL
  12. 12. ■ System architecture of InstaSPIN-FOC located in MCU: TMS320F280xF □ FAST module: provides flux amplitude/angle/speed and torque □ required speed and current control algorithms to achieve FOC Solution to realizing sensorless control 12Prof. D.W.J. Pulle  System implementation example ?
  13. 13. ■ Example of implementation ■ System components: □ PM motor □ Micro Controller Unit (MCU) with InstaSPIN-FOC □ Power electronic converter Solution to realizing sensorless control 13Prof. D.W.J. Pulle  How do we interface/communicate with the MCU ?
  14. 14. ■ Use of VisSim [4] 'embedded control' development software ■ Meaning 'Embedded control' ? □ Ability to represent and develop a control algorithm for the MCU using graphical modules instead of C-code □ Example where we implement z=(x+y)q in scaled fixed point format  Use of VisSim  Use of C-code Solution to realizing sensorless control 14Prof. D.W.J. Pulle  VisSim approach simplifies the development of complex control structures by using a graphical interface and does efficient C code generation
  15. 15. ■ Example of VisSim [4] 'embedded control' development software □ Ability to develop and veritfy a fixed point control algorithm for the MCU and test this with a simulated motor model first Solution to realizing sensorless control 15Prof. D.W.J. Pulle  VisSim for sensorless control using InstaSPIN? Analysis controller details Motor model
  16. 16. ■ Use of VisSim to develop complete sensorless embedded controller structure ■ Modules present: □ InstaSPIN-FOC : specially designed VisSim module [5] which executes the FAST algorithm and all control tasks □ ADC-PWM: module used to obtain the measured voltage/currents and controls the power electronic converter Solution to realizing sensorless control 16Prof. D.W.J. Pulle  Compilation of this module to C-code generates an 'outfile' that runs the drive
  17. 17. ■ Example of a VisSim based controller which operates a sensorless drive Solution to realizing sensorless control 17Prof. D.W.J. Pulle  Example from Texas Instruments InstaSPIN- VisSim workshop program [5]
  18. 18. Applications of sensorless control 18Prof. D.W.J. Pulle  Application examples ■ Why customers change to Sensorless control using InstaSPIN? □ To reduce product cost : removal of the encoder leads to significant savings □ To reduce product development time □ To increase reliability: to avoid encoder alignment and breakdown problems □ To enhance their product: having access to instantaneous shaft torque and speed without requiring additional (to the power leads) is attractive □ Ability to measure and track key motor parameters: measurement of, for example, resistance gives the ability to monitor temperature of the motor □ Need for sophisticated motion control: use of SpinTAC [6] control suite □ To provide back up to encoder based drives: provides the ability to keep the drive in operation if an encoder related problem occurs □ To improve energy efficiency: for drive which use an induction machine, which allows field weakening during partial load operation
  19. 19. Applications of sensorless control 24.06.2014 19 Prof. D.W.J. Pulle  Application examples ■ Application examples: currently in use or being developed PUMPS •Transmission •Brake/Boost •Oil •Turbo •Fuel/Water •Constant pressure •Water/Waste/Chemical •Spa/pool pump •Geothermal pump •Dishwashers Automotive Industrial/Consumer COMPRESSORS •Refrigeration •Air/Con •Refrigeration Automotive Industrial/Consumer BLOWERS/FANS •Air/Con Blowers •Cooling Fan •Respiratory •Vacuum •Fans •Air/Con Blowers •Exhaust Automotive Industrial/Consumer •Washers •Dryers LAUNDRY HIGH TORQUE •Traction •eBike/Moped/Scooter •Off-highway Vehicles •Carts, Transport •Fork lifts •Wheel chairs •Escalators •Elevators •Treadmill •Tools •AC Drive / Inverter •Assembly Line Transit Conveyors
  20. 20. Applications of sensorless control Prof. D.W.J. Pulle ■ Specific application examples: Traction drive for Electric Vehicle □ Use of InstaSPIN-FOC to realise a highly responsive & efficient torque machine like this high performance 'Tesla' 20
  21. 21. Applications of sensorless control Prof. D.W.J. Pulle ■ Specific application examples: Induction machine drives using PowerWarp Algorithm is based on reducing motor copper losses in the stator AND the rotor! PowerWarp Savings 80% of energy vs. Traditional Triac 45% of energy vs. IS-FOC Real World Field Trial Induction Motors used for Agriculture Air & Humidity Control Adaptively reduce magnetizing current to only induce the field required for the torque required!
  22. 22. Applications of sensorless control Prof. D.W.J. Pulle ■ Specific application examples: Bow thruster for yacht Max power: 3500W Speed: 6000 rpm Max current: 15 A Efficiency : 90% Weight: 3.5 kg Max power: 2200W Speed: 4100 rpm Max current: 280 A Efficiency : 75% Weight: 8.9 kg Bow Thruster propeller DC brushed motor Bow Thruster motor inside yacht Three-phase PM motor
  23. 23. ■ Introduction on sensorless electrical drive technology: ■ Challenges faced when trying to implement a sensorless drive ■ Introduction on the InstaSPIN Sensorless solution ■ Use of VisSim Embedded software that can be used to speed up and simplify the development of MCU based control in general and InstaSPIN in particular ■ Overview of InstaSPIN based applications currently in placed and those being developed world wide Summary 23Prof. D.W.J. Pulle Thank you for your attention !
  24. 24. 1. Fundamentals of Electrical Drives, Veltman, A. , Pulle, D.W,J. and De Doncker R. , Springer 2007 2. Advanced Electrical Drives,, De Doncker R. , Pulle, D.W,J. and Veltman, A. , Springer 2010 3. Texas Instruments InstaSPIN: 4. VisSim: 5. C2000 based 2-Day Hands-On Motor Control Workshops 6. SpinTAC motion control suite: part of InstaSPIN-Motion control References 24Prof. D.W.J. Pulle