In my project I have a lean mixture in the second bank at WOT (about 10% compared with the first bank). For example Bank1= 10.6 AFR, Bank2 = 11.3 AFR. These conclusions on indications of wideband lambdas and on voltage narrowband probes. And indirectly on the larger retard values in the second bank. Swapped wb probes - the same effect.
  I have 1.9 boost peak at WOT, and 740cc(4 bar base pressure) injectors with 2 walbro inline. Fuel pressure is normal - stable 5.9 at 1.9 bar boost. In my opinion there is high frequency pressure pulsation in the rail at high flow. And the flow area of ​​the serial connected 6AN-fuel rails is not enough for equally supply both banks. This is confirmed by the fact that with a smaller boost delta AFR is less or null.
  Injection times TI_W and TI2_W are always equal by logging. I decided to consider the option of adjusting the injection time of only one of banks. According to RKTI it can only be corrected by changing relative fuel mass(RK_W and RK2_W from GK modul) which depend on "Lambda controller output"(FR_W,FR2_W in LR), Lambda limitation and other functions. If the injection is separate by banks it means that we can influence it. But I searched and could not find the other bank divisions except catalysts heating , sensor diagnosis and other useless functions. Has anyone studied this issue?

Of course I can change the fuel supply to the rail, but this issue still interesting.

yes, I looked it. As I understand it is a factor of balance between banks. Setting there for example 0.1 will multiply relative fuel mass by 0.9 bank2 and simultaneously multiply by 1.1 Bank1. I'll try it. thanks