Feed rate changes during straight cuts

I tried cutting 1mm stainless steel on our Wazer recently and I noticed that the speed of cutting is changing a lot, even during straight cut segments. Is this supposed to happen? It repeatedly caused a partial cut as seen in the following picture.

I did not save the gcode but I have example from another WAM generated gcode where it also changes feed midst straight line. Same Y coordinate, increasing feed in quite discrete manner. I am not sure if there is some pressure increase that is supposed to follow this feedrate increase, but it does not seem like that.

Overall, I am having hard time to dial in good feedrates (through adjusting material thickness) that cuts through the material but does not completely obliterate the cutting board underneath. Generally, for a successful cut, I need to go very slow and the waste board gets shreaded very deep. I really hope this can be improved.

Just for reference, I recreated the same cut path (slightly different work origin so the X and Y coordinates don’t match the previous gcode snippet) and used simulation in WAM to visualize where the speed increase happens. It aligns perfectly with where the partial cut happens in the picture above.

That shouldn’t be happening. Are the sections different lines in the original drawing? Are they slightly curved and not straight lines?

No, just straight lines. I had a long back and forth with support on their explanation is that this step-wise speed change is caused by the current implementation of ‘decelerating’ in front of corners or turns and subsequent ‘accelerating’. The feed simply decreases and increases by approximately 10% after 5mm cut travel.