Grid forming inverter control

Question

I have been working on a project where the three phase voltage source inverter has to be controlled in so called grid forming mode of operation i.e. the inverter creates three phase grid of fixed frequency and voltage. I have studied a literature and I have found that it is pretty common task nowadays. The task is usually solved by following control structure

My problem is that one of the devices connected to the three phase grid formed by the inverter is squirrel cage induction motor based compressor drive. There is no power electronics converter between the compressor drive and the three phase grid i.e. the compressor drive is started via the direct online connection. The compressor drive can be connected to the grid in any time instant and there is no direct information about its connection (no speed sensor of the compressor, no feedback signal from a contactor). The only one indirect information about the compressor drive connection is the current drawn from the inverter which is a sum of the currents consumed by all the devices connected to the grid. In these assumptions the control algorithm in the digital signal processor controlling the inverter has to ensure that the compressor drive successfully starts up. Also in case when the inverter voltage is reduced due to achieving the current limit in the inner control loop in the control structure above.

I have been thinking about solution of that problem and one idea has popped-up in my mind. The idea is based on the equivalent circuit of the induction machine

and the slip formula \$s=\frac{\omega_s - \omega_m}{\omega_s}\$. In other words I can indirectly reduce the inrush current via reducing the slip which can be done via reduction of the grid frequency. The reduction of frequency should be followed in my opinion by the grid voltage magnitude reduction according to the voltage-frequency characteristic so that the induction motor of the compressor drive operates with its nominal flux. It could be said that I would append following control structure above the aforementioned one.

The question is whether that idea could ever work and if so how to realize the modification of the grid frequency (P,PI,PID controller or something else and based on what signal)?

Answer 1

Reducing the frequency will not reduce the motor slip. Reducing the frequency will reduce the reactances in the motor's equivalent motor and increase the current. Reducing the voltage and frequency to maintain constant V/Hz would help, but would not be good for other loads.

One option would be to increase the short-time current capacity of the inverter. It might be possible to do that by adding a lot of capacitance to the DC side of the inverter.

You could also use one of the non-electronic methods of reduced voltage starting: wye-delta, autotransformer, series reactor, or series resistor. An electronic reduced-voltage starter might not be compatible with the inverter. A variable frequency drive (VFD) is another alternative that might not be compatible with the inverter.

A DC powered VFD would work, but the DC source would need to have a voltage range that is compatible with the VFD. Depending on the available DC source, you may be able to find an off-the-shelf VFD that would work.

Answer 2

Grid-forming inverters are an emerging technology that allows solar and other inverter-based energy sources to restart the grid independently.

The inception of electrical inertia (EI) with phase lag of frequency reduction is common. The sensors, controls and use of the Park Transform are used to simply the design with improved performance to various disturbances.

Details to study.