openinverter forum

cloudy wrote: ↑Sun Mar 05, 2023 6:51 pm A few questions come to mind:

• Do you correct at a flat rate per cycle, linearly against error size or even apply some exponent?

• Do you pull / return throttle at the same rate?

• Do you have hysteresis or just set a target well outside of "straight and steady" slipAngle noise?

Firstly, I should clarify that my algorithm is limiting current, not wheelspin, and it runs in the main inverter loop at 8.8kHz, so things are going to be a bit different compared with running the algorithm at 100Hz, however I hope the same principles apply.

1) I apply a correction that is linear multiple of the error error.
2) My algorithm *can* increase and decrease at different rates, but using the same rate is fine for most circumstances.
3) There's no hysteresis, but because correction is a multiple of error, when everything is okay, the corrections are negligible.

For example, in your case, if the slip is 20% then the throtmax can be assumed to be correct, so no change is required. If it was 40% then you would decrease throtmax twice as fast as if it was 30%.

I hope this makes sense, and I'll be interested to know if this helps. You'll still be contending with throttle ramps of course, but hopefully this will smooth things out.

Not able to get as close with an integral approach, tried a few things. Probably need to look towards PID - the derivative should prevent the overshoot both of these approaches give.

Back to testing this. This time implemented a full PID controller.

Parameters need a fair bit of tuning, but best result yet. Still oscillating but slip angle was kept within 40% - current is being pulled back within 50ms of the slip starting, but still takes 200ms to control.