.DriveControl.UsersGuide.CurrentController

Information

Considering the equations shown in User's Guide, Machine and inverter, we can write the transfer function of the drive as:

.

Note: To avoid influence of the current ripple on control, the measured current is filtered. The small time constants of the inverter and the smoothing filter are summed up: T_σ=T_d;+T_sI

Choosing a PI-controller

.

we get the transfer function of the open loop:

.

Compensating the bigger time constant T_a, i.e. T_i=T_a, the transfer function of the closed loop gets:

.

According to the absolute optimum, we check the absolute of this transfer function, trying to keep it = 1 for a wide frequency range:

Comparing the coefficients in numerator and denominator, we find:

• ω⁰: always fulfilled
• ω²: allows to calculate optimal k_p
• ω⁴: not possible to fulfill, results in limited bandwith of the controller

The optimal proportional gain is found as k_p = R_a T_a / (2 T_σ).

The transfer function of the current controlled drive gets:

The numerator's zero can be compensated by a first order filter of the reference, resulting in:

The first order with substitute time constant T_sub = 2 T_σ is an approximation used for the design of the further controllers.

Note: The induced voltage V_i = kφ ω can be calculated from measured speed and acts like a disturbance. Therefore V_i should be used as a feed-forward for the current controller. Furthermore, the voltage that can be applied to the drive is limited. Therefore we use a
limited PI controller with feed-forward and anti-windup.

The example CurrentControlled applies a reference torque step to the torque controlled drive at stand still, the load torque is linearly dependent on speed.

The current controller causes a constant armature current, that causes constant torque delivered by the machine. Speed will rise up to that point of operation, where that torque is in balance with the load torque.