First, is this a four wire or three wire system?
If it’s a four wire system, which I suspect it is, then it appears that Input_2 is probably correct. I don’t know what it is measuring, but it’s hard to get a PF = 1 if the phase reference is wrong. The CT does appear to be reversed, but let’s leave that for now, the IoTaWatt is correcting that.
The other two appear to be assigned to the wrong phase, based on the power-factor. If your description of the system is correct, you should be able to change the phase of Input_1 and Input_3 to another value that produces PF close to 1.
Without seeing how your system is connected, I can’t offer much more. If the power appears to be more than can be reasonably expected from the inverters, then there is something else going on.
I’m not familiar with how Finland works. Seems to me that a total of 2 kWp could be handled by one single-phase inverter. It doesn’t seem worth the expense to have three inverters. Are you sure 1500 Watts isn’t a reasonable output? If my assumptions about phase above are correct, and I don’t have high confidence due to very limited information, the outputs would be about 1500, 1500, and 3000 for inputs 1, 2, and 3 respectively. But that doesn’t seem to be consistent with your stated panel output.