Thank you for your response and for directing me to the other thread.
Apologies if I've misunderstood, but I wasn't considering any losses after the inverter: rather than examining the output at the grid injection point, I was examining the output at the inverter, which is where the limitation is applied. So why are the clipped values of the inverter output, 'EOutInv', dipping below the threshold? For example, I'm imposing a clipping limit of 2.55MW, so I assume that an unclipped 'EOutInv' value of 2.61MW would be clipped to exactly 2.55MW, whereas it is actually clipped to 2.53MW.
I'd appreciate if you can help me to understand the process more