there is a PVT analysis of an oil reservoir that i want to use for EOS regression.
when i split puls fraction into three seudo components the calculated saturation pressure
will be 1622 psi while the observed saturation pressure is 1618 psia.
there are some questions:
how could i know that this is a good splitting?
can i use phase diagram of new mixture(with 3 pseudoComponents) as a valid reference for regression?
or I should compare phase diagram of new mixture with phase diagram of mixture before splitting?

what is more acurate?
phase diagram of mixture before splitting with 1450 psi calculated saturation pressure OR
phase diagram of mixture after splitting with 1622 psi calculated saturation pressure?

thank you