CPA works as other EOS models in Prode (Peng-Robinson, Soave-Redlich-Kwong etc.),
Prode Properties allows to define one BIPs data set per each model (you can define different models for fugacity, enthalpy etc.) and stream,
for a vapor-liquid phase equilibria or a vapor-liquid-liquid phase equilibria normally one gets BIPs from VLE database (which includes a large number of BIPs, I think about 30000 or more),
for liquid-liquid phase equilibria or solid-liquid phase equilibria or hydrates there are specific databases (there are four databases, VLE, LLE, SLE, Hydrates)
to obtain accurate results with liquid-liquid (or solid-liquid or solid-vapor) phase equilibria I calculate the BIPs with the data regression procedure included in Prode, there are examples in the operating manual
actually I load BIPs for CPA from VLE database in Prode,
however, calculated liquid-liquid fractions are different from our reference while vapor-liquid is Ok,
what can I do for calculating the correct vapor-liquid and liquid-liquid fractions ?
can't you include all equilibrium data points (VLE and LLE in your case) when calculating BIPs ?
See the attached image as example,
Prode can calculate optimized values (for both VLE and LLE)