Vector Control of Permanent Magnet
Synchronous Motors (PMSMs)
1. Introduction
Permanent Magnet Synchronous Motors (PMSMs) are widely used in electric vehicles,
robotics, and industrial automation due to their high efficiency, high torque density, and
excellent performance at varying speeds.
Key Advantages:
- High efficiency
- High power density
- High torque/inertia ratio
- No rotor current losses
To fully exploit these features, Vector Control (also called Field-Oriented Control, FOC) is
used to regulate PMSMs like a DC motor—independently controlling torque and flux.
2. Structure of PMSM
A PMSM consists of:
- Stator: Wound with a 3-phase winding, similar to an induction motor.
- Rotor: Contains permanent magnets (surface-mounted or embedded).
3. Dynamic Model of PMSM in d-q Reference Frame
3.1 Assumptions:
- Sinusoidal back EMF
- Negligible iron losses
- Balanced 3-phase system
- No saliency for surface-mounted PMSMs (i.e., Ld = Lq)
3.2 Voltage Equations:
vd = Rs * id + Ld * (did/dt) - ωe * Lq * iq
vq = Rs * iq + Lq * (diq/dt) + ωe * Ld * id + ωe * λm
4. Electromagnetic Torque Equation
Te = (3/2) * p * [λm * iq + (Ld - Lq) * id * iq]
For surface-mounted PMSM (Ld = Lq):
Te = (3/2) * p * λm * iq
This shows torque is controlled by iq and flux is controlled by id — the basis of vector
control.
5. Principle of Vector Control (Field-Oriented Control - FOC)
5.1 Objectives:
- Decouple control of torque and flux
- Control PMSM like a DC motor
5.2 Basic Steps:
1. Measure stator currents ia, ib, ic
2. Transform to αβ-frame using Clarke Transformation
3. Transform to dq-frame using Park Transformation
4. Control id and iq using PI controllers
5. Inverse Park and Clarke Transformations to generate PWM signals
6. Transformations
6.1 Clarke Transformation (abc to αβ):
[iα, iβ]ᵗ = (2/3) * [[1, -1/2, -1/2], [0, √3/2, -√3/2]] * [ia, ib, ic]ᵗ
6.2 Park Transformation (αβ to dq):
[id, iq]ᵗ = [[cosθe, sinθe], [-sinθe, cosθe]] * [iα, iβ]ᵗ
7. Control Structure
The vector control system consists of:
1. Current Controllers (PI): For id* and iq*
2. Speed Controller (PI): Outputs iq*
3. Flux Control: Typically, id* = 0 for surface PMSM
4. Space Vector PWM (SVPWM): Generates inverter gating signals
8. Rotor Position Detection
Accurate rotor position is essential for proper Park/Clarke transformations. It can be
achieved by:
- Encoder (incremental or absolute)
- Resolver
- Sensorless Estimation (using voltage and current feedback + motor model)
9. Simulation and Implementation Notes
- Control Loop Frequencies:
- Current loop: ~10 kHz
- Speed loop: ~1 kHz
- PWM frequency: 10–20 kHz
- Anti-windup and feedforward terms may improve performance
10. Summary of Key Equations
PMSM Equations in d-q Frame:
vd = Rs * id + Ld * (did/dt) - ωe * Lq * iq
vq = Rs * iq + Lq * (diq/dt) + ωe * Ld * id + ωe * λm
Te = (3/2) * p * (λm * iq + (Ld - Lq) * id * iq)
Transformations:
- Clarke (abc to αβ)
- Park (αβ to dq)
11. Advantages of Vector Control
- Fast dynamic response
- Accurate torque control
- Higher efficiency
- Wide speed range operation
12. Conclusion
Vector control makes PMSM behave like a separately excited DC motor, enabling
independent torque and flux control. It's the preferred method in applications requiring
high performance and dynamic response.