SENSORLESS #

Brief #

The sensorless component is responsible for estimating the motor’s velocity and position in real-time using electrical measurements. It applies low-pass filtering, back EMF calculation, velocity compensation, and position integration to achieve this.

Description #

1. Input Reading and Initialization:

• l and r are the inductance and resistance of the motor, respectively. They are ensured to be positive.
• The voltage measurements ud and uq are updated using a delay line (ud2, uq2, ud1, uq1, ud, uq).
• The current measurements id and iq are read from the input pins.

2. Gain Clamping:

• The gains kb, ki, and kl are clamped to ensure they stay within valid ranges.

3. Low-Pass Filtering of Current Changes:

• The changes in currents delta_id and delta_iq are calculated using a low-pass filter. This helps to smooth out noise in the current measurements.
• The previous current values old_id and old_iq are updated to the current measurements.

4. Back EMF Calculation:

• The back EMF (ed and eq) is estimated using the motor’s electrical equations. The back EMF is the voltage generated by the motor due to its rotation.
• The equations used are:

ed = ud - r * id - PIN(delta_id) * l / period + vel * l * iq * kb;
eq = uq - r * iq - PIN(delta_iq) * l / period - vel * l * id * kb;

5. Velocity Compensation:

• The estimated velocity vel is adjusted to compensate for the back EMF. The goal is to make the d-axis back EMF (ed) approach zero, which helps in stabilizing the control.
• The compensation is done using the following equation:

vel -= SIGN2(eq, 1.0) * ed * ki * period;

6. Startup Boost:

• If the absolute value of the estimated velocity is below a minimum threshold (min_vel), a boost is added to help the motor start up more reliably.
• The boost is applied using the following equation:

vel += SIGN2(id * iq, 0.1) * PIN(vel_boost);

7. Velocity Limiting:

• The estimated velocity is limited to ensure it does not exceed the maximum allowed value (max_vel).

8. Position Update:

• The estimated position pos is updated by integrating the estimated velocity over time.
• The position is wrapped using the mod function to ensure it stays within a valid range (e.g., 0 to 2π for angular position).

9. Output Updates:

• The estimated back EMF (ed and eq), velocity (vel), and position (pos) are updated on the output pins.

Pins #

References #

Source: .c File