decline_predict: Arps decline prediction

Description Usage Arguments Value References Examples

View source: R/Rdca_package.R

Description

Create a data frame of decline predictions according to the class of 'decline_lst' and 'time_lst' objects

Usage

1
decline_predict(decline_lst, time_lst)

Arguments

decline_lst

a list object of class 'decline'

time_lst

a list object of class 'time'

Value

a data frame of decline estimates according to the class of 'decline_lst' and 'time_lst' objects

References

\insertRef

Arps1945Rdca

\insertRef

Robertson1988Rdca

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
decline_param_1 <- decline_param(input_unit = "Field", output_unit = "Field",
fluid = "oil",
model = "exponential", qi = 1000, Di = 0.15, b = 0, q_abnd = NULL)
decline_time_1 <- decline_time(c(1:7300), unit = "day")
decline_predict_1 <- decline_predict(decline_param_1, decline_time_1)

head(decline_predict_1)

decline_param_2 <- decline_param(input_unit = "Field", output_unit = "SI",
fluid = "oil",
model = "hyperbolic", qi = 15000, Di = 0.1, b = 0.95, q_abnd = 200)
decline_time_2 <- decline_time(seq(as.Date("2016/04/15"), by = "days",
length.out = 3600), unit = "date")
decline_predict_2 <- decline_predict(decline_param_2, decline_time_2)

head(decline_predict_2)

Rdca documentation built on July 2, 2020, 3:59 a.m.