Development of Decline Curve Analysis Parameters for Tight Oil Wells Using a Machine Learning Algorithm

Table 2

Parameters and associated distribution to generate the input dataset.

Parameters	Minimum value	Maximum value	Distribution function
Grid size, direction(DI) (ft)	75	125	Uniform[75,125]
Grid size, direction(DJ) (ft)	30	80	Uniform[30,80
Grid size, direction(DK) (ft)	1	5	Uniform[1,5]
Matrix permeability (mD)	0.0001	1	Power[10,Random[-4,0]]
Porosity	0.05	0.15	Uniform[0.05,0.15]
Horizontal well length (ft)	1800	6000	Triangle[1800,3500,6000]
Fracture half-length (ft)	100	850	Triangle[100,380,850]
Fracture spacing (ft)	75	500
Effective fracture permeability (mD)	1	100	Uniform[1,100]
Layer-up	1	3	Uniform[1,3]
Layer-down	1	3	Uniform[1,3]
Monitored oil rate (bbl/day)	1.5	2.5	Triangle[1.5, 2, 2.5]
Initial reservoir pressure (psi)	2000	6000	Uniform[2000,6000]
Bubble point pressure, psi	400	6000	Initial reservoir pressure/random[1,15]
Operating BHP (psi)	200	3000	Initial reservoir pressure/random[10,30]

Note: the range of each parameter was set based on Chinese oilfield practice, such as Ordos basin [31–33].