Petrophysical Regression regarding Porosity, Permeability, and Water Saturation Driven by Logging-Based Ensemble and Transfer Learnings: A Case Study of Sandy-Mud Reservoirs

Table 1

Classic petrophysical models employed to predict effective porosity, effective permeability, and water saturation of conventional petroleum-bearing reservoirs.

Reservoir parameter	Model	General expression	Variable
Porosity (, %)	Single-log model [4]		: compaction factor; : value from acoustic, density, or neutron log; : logging value of matrix; : logging value of fluid
	Wyllie-Rose model [5]		: acoustic logging value of matrix; : value of acoustic log; : formation factor
	Density-neutron root-mean-square (RMS) model [1]		: compaction factor; : values of density and neutron logs; : density and neutron logging values of matrix; : density and neutron logging values of fluid
Permeability (, mD)	Kozeny-Carman model [6, 7]		: empirical coefficient; : porosity; : specific surface area of rock
	Krumbein-Monk model [8]		: empirical coefficients; : median grain diameter; : standard deviation of
	Timur model [2]		: empirical coefficients; : porosity; : irreducible water saturation
Water saturation (, %)	Archie model [9]		: empirical coefficients; : resistivity of formation water; : resistivity of true formation
	Waxman-Smits-based model [3]		: conductivity of true formation; : coefficient and power of water saturation term; : coefficient and power of shale term
	Poupon-Leveaux-based model [10]		: conductivity of true formation; : coefficient and power of water saturation term; : coefficient and power of cross term; : coefficient and power of shale term