Vector Control

Vector control, also called Field Oriented Control (FOC), is the most advanced technique used to control AC induction and synchronous motors with performance comparable to DC motors. In conventional scalar control, only the magnitude of voltage and frequency are controlled, which limits dynamic response. Vector control solves this by independently controlling the flux-producing and torque-producing components of the stator current, giving precise and fast torque control essential in industrial drives, robotics, and electric vehicles.

Core Concept of Vector Control

In a DC motor, the field winding and armature winding are physically perpendicular, which means flux and torque are naturally decoupled and can be controlled independently. An AC induction motor does not have this physical separation. The same stator current produces both the magnetizing flux and the electromagnetic torque. This coupling is what makes AC motor control difficult at dynamic conditions.

Vector control artificially creates this decoupling in software. The three-phase stator currents ia, ib, ic are transformed into a two-axis rotating reference frame aligned with the rotor flux. This frame has a d-axis (direct axis) aligned with the rotor flux vector and a q-axis (quadrature axis) perpendicular to it. In this frame, the d-axis current component id controls the flux, and the q-axis current component iq directly controls the torque. They can now be controlled with separate PI controllers, exactly like a DC motor.

The transformation from three-phase stationary frame to the rotating dq frame involves two steps. First, the Clarke transform (abc to alpha-beta stationary frame) reduces three phases to two orthogonal components. Then the Park transform (alpha-beta to dq rotating frame) rotates the reference frame to align with the rotor flux angle. The angle of the rotor flux (theta) is either measured using an encoder or estimated using a flux observer, which is called sensorless FOC.

Mathematical Expression

The electromagnetic torque produced by an induction motor in the dq reference frame is expressed as a clean product of flux and torque current. If the rotor flux magnitude is held constant at its reference value, torque becomes directly proportional to the q-axis current iq alone. This is the fundamental equation that makes vector control so powerful.

The torque equation in the rotor-flux-oriented reference frame is given as T = (3/2) * (P/2) * (Lm / Lr) * lambda_r * iq, where P is the number of poles, Lm is the mutual inductance, Lr is the rotor self-inductance, lambda_r is the rotor flux linkage, and iq is the q-axis stator current. By keeping lambda_r constant using the d-axis current controller (id = lambda_r / Lm), the torque expression reduces to T proportional to iq only. Controlling torque then becomes as simple as controlling a single current component.

Practical Understanding

In practice, vector control is implemented digitally in a microcontroller or DSP. The current sensors measure ia and ib, ic is computed from Kirchhoff's law. The Clarke and Park transforms are computed in software. Separate PI controllers regulate id and iq. The inverse Park transform converts the voltage commands back to alpha-beta frame, and then Space Vector PWM (SVPWM) generates the gate pulses for the inverter switches.

The accuracy of the entire control depends heavily on the correctness of the rotor flux angle theta. Even a small error in theta couples the d and q axes and degrades performance. This is why encoder-based FOC is preferred where high accuracy is needed, while sensorless FOC using observers is used where encoder mounting is not possible. The bandwidth of the current controller must be much higher than the speed controller to maintain the inner loop stability.

Given:
Motor: 4-pole induction motor, Lm = 0.15 H, Lr = 0.16 H
Rotor flux reference: lambda_r = 0.9 Wb
Required torque: T = 20 Nm

Why this formula applies:
In FOC with constant rotor flux, torque depends only on iq

Formula:
T = (3/2) * (P/2) * (Lm / Lr) * lambda_r * iq

Substitution:
20 = (3/2) * (4/2) * (0.15 / 0.16) * 0.9 * iq
20 = 1.5 * 2 * 0.9375 * 0.9 * iq
20 = 2.531 * iq

Calculation:
iq = 20 / 2.531

Final Answer:
iq = 7.90 A (q-axis current required to produce 20 Nm torque)

Exam Tip: In GATE questions on FOC, always remember that id controls flux and iq controls torque. If rotor flux is held constant, torque is directly proportional to iq only. The Park transform angle comes from the rotor flux position, not the rotor mechanical angle directly.

Mechanism Explained

• Clarke Transform converts ia, ib, ic into two orthogonal stationary components i_alpha and i_beta, reducing the three-variable problem to two variables.
• Park Transform rotates the stationary alpha-beta frame by the rotor flux angle theta to produce rotating dq components id and iq.
• The d-axis current id is controlled to set rotor flux. Keeping id constant means constant flux, which is the standard operating mode below base speed.
• The q-axis current iq is the torque-producing component. The speed PI controller output directly becomes the iq reference, giving fast torque response.
• Inverse Park and Clarke transforms convert the dq voltage commands back to three-phase references for SVPWM generation.
• The rotor flux angle accuracy is the most critical factor. Encoder-based methods are more accurate; sensorless observers work well above a minimum speed threshold.

Quick Revision

• Vector control decouples flux and torque control in AC motors by transforming currents to a rotating dq reference frame aligned with rotor flux.
• d-axis current id controls rotor flux; q-axis current iq controls electromagnetic torque independently.
• Torque formula: T = (3/2)(P/2)(Lm/Lr) * lambda_r * iq. With constant flux, T is proportional to iq only.
• Clarke transform: abc to alpha-beta (stationary). Park transform: alpha-beta to dq (rotating). Both are linear algebraic transforms.
• Rotor flux angle theta is the critical signal. Error in theta couples d and q axes and degrades dynamic response.
• GATE trap: Do not confuse scalar V/f control (no decoupling) with vector control (full flux-torque decoupling). FOC gives far superior dynamic response.
• Sensorless FOC uses flux observers or back-EMF estimators to compute theta without a physical encoder.