Decline curve analysis for Rate vs. Cumulative production.

[J Gre]

Hi, I'm trying to manipulate the simplt DCA script to one that does rate vs. cumulative production instead of rate vs. time. I have very little R experience and just tried to remove/replace all the instances of time with cumulative production.

 

This is what I did.

#------------------------------------------------------------------------------

Hyp.q.forward.fun = function(Gp, Hyp.qi, Hyp.b, Hyp.Di ){

# Peter Shaw

# e.g. SPE 98042 (11a)

Hyp.q.theo = Hyp.qi*(1-(Gp/(Hyp.qi*((1-Hyp.b)*Hyp.Di)^(-1))))^(1/(1-Hyp.b))

return(Hyp.q.theo)

}

#------------------------------------------------------------------------------

residfun = function(x,x.cum,y.prod){

Hyp.qi = x[1]

Hyp.b = x[2]

Hyp.Di = x[3]

 

q.theo = Hyp.q.forward.fun(

Gp=x.cum,

Hyp.qi=Hyp.qi,

Hyp.b=Hyp.b,

Hyp.Di=Hyp.Di)

 

#plot(x.cum,y.prod); lines(x.cum,q.theo)

residual = sqrt(sum((q.theo-y.prod)^2))

return(residual)

}

#------------------------------------------------------------------------------

min.data = 3

result = data.frame(

cum.prod = numeric(0),

production = numeric(0),

theo = numeric(0)

)

Hyp.qi =as.numeric(NA)

Hyp.b =as.numeric(NA)

Hyp.Di =as.numeric(NA)

ok = length(col.Production)>0

if(ok) ok = length(col.cumProd) == length(col.Production)

if(ok){

u.order = order(col.cumProd)

 

col.cumProd = col.cumProd[u.order]

col.Production = col.Production[u.order]

 

 

u0 = which.max(col.Production)

u = u0:length(col.Production)

 

x.cum = col.cumProd

y.prod = col.Production

 

x0 = c(y.prod[1])

 

optim.result = optim(

par = x0,

fn=residfun,

x.cum=x.cum, y.prod=y.prod

)

 

Hyp.qi = optim.result$par[1]

Hyp.b = optim.result$par[2]

Hyp.Di = optim.result$par[3]

 

 

 

result = data.frame(

cum.prod = x.cum,

production = y.prod

)

}