[Compositional Simulation]Nothing changes from my simulation than the tutorial one.

[Débora Yohane]

Hi! I'm kinda very desperate because I've tried everything, revised it many times and I continue to be stuck on it for the last month. I kindly ask for your help. 

So, I'm trying to do a Compositional Simulation 1D  in a quadratic domain and solve it with natural variables, it runs normally without errors but the result ploted doesnt change a thing, it's like I've never made any difference for adding an well or even changing pressure. The boundary conditions consider the reservoir being isolated without replacing fluid in it. 

The code is commented as follows:

mrstModule add compositional ad-props ad-core

%===============SET UP PROBLEM====================

%TODO: mudar grid

nx = 50;

ny = 50;

nz = 1;

% Quadratic domain

dims = [nx, ny, nz]; %number of cells in each coordinate direction.

pdims = [1000, 1000, 10]; %physical size in units of meters of the computational domain

G = cartGrid(dims, pdims);

G = computeGeometry(G);

% Generate permeability

K=160*ones(G.cartDims);

K=K(G.cells.indexMap);

rock = makeRock(G, K*milli*darcy, 0.3);  %grid, perm, poro

figure;

plotCellData(G, K);

colorbar;

title('Permeability [mD]');

%===============SET UP COMPOSITIONAL MODEL====================

names = {'Metano'  'Butano'};

% Critical temperatures

Tcrit = [190.6, 408.1];

% Critical pressure

Pcrit = [45.40, 36.00]*101325;

% Critical volumes

Vcrit = [0.0990, 0.2630]*1e-3;

% Acentric factors

acentricFactors = [0.013, 0.1956];

% Mass of components (as MRST uses strict SI, this is in kg/mol)

molarMass = [16.0430, 58.1240]*1e-3;

% Initialize fluid

fluid = CompositionalFluid(names, Tcrit, Pcrit, Vcrit, acentricFactors, molarMass);

%=============SET UP INITIAL STATE============================

Pres = 50*0.0689476*barsa;

totTime = 10*year;

pv = poreVolume(G, rock);

%boundary condition structure as defined by function ‘pside’

%bc = pside([], G, 'West' , Pres); 

%W = verticalWell([], G, rock, nx/2+1, ny/2+1, 1:nz,'Type', 'rate', 'Val', 10000*meter^3/day,'comp_i',[1,0]);

W = verticalWell([], G, rock, nx/2, ny/2, 1:nz, 'sign',-1,'Type', 'rate', 'Val', -10000*meter^3/day,'comp_i',[1,0]);

plotWell(G, W, 'color', 'k');

flowfluid = initSimpleADIFluid('phases','OG','rho', [1000, 500], ...

                       'mu', [1, 1]*centi*poise, ...

                       'n', [2, 2], ...

                       'c', [1e-5, 0]/barsa);

ncomp = fluid.getNumberOfComponents();

s0 = [1, 0];

state0 = initState(G,W, Pres, s0); %Initialise state object for reservoir and wells.

T = 288.71; %K

state0.T = repmat(T, G.cells.num, 1); 

state0.components = repmat([0.85 0.15], G.cells.num, 1);

dt = rampupTimesteps(totTime, 75*day, 12);

schedule = simpleSchedule(dt);

%================SOLVE WITH NATURAL VARIABLES==========

model = NaturalVariablesCompositionalModel(G, rock, flowfluid, fluid, 'water', false);

nls = NonLinearSolver('useRelaxation', true);

[~, states, report] = simulateScheduleAD(state0, model, schedule, 'nonlinearsolver', nls);

%=================LAUNCH PLOT=========================

mrstModule add mrst-gui

figure;

plotToolbar(G, states)

title('Natural variables')

Could you please help me? :( please. 

[Olav Møyner]

Hi,

You need to pass wells/BC to the schedule, otherwise nothing will change with uniform initial conditions. You will then also get errors that tell you to add .components to BC/wells to define injection composition.

Best regards

Olav

---

From: sinte...@googlegroups.com <sinte...@googlegroups.com> on behalf of Débora Yohane <debor...@gmail.com>
Sent: Thursday, November 12, 2020 11:28:30 PM
To: MRST-users: The Matlab Reservoir Simulation Toolbox User Group <sinte...@googlegroups.com>
Subject: [MRST Users] [Compositional Simulation]Nothing changes from my simulation than the tutorial one.

 

--
You received this message because you are subscribed to the Google Groups "MRST-users: The Matlab Reservoir Simulation Toolbox User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sintef-mrst...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sintef-mrst/4194af32-2b26-400c-8b44-a1a89ba93a1dn%40googlegroups.com.