I think that you would do yourself a favor by learning to use command-line functions instead of relying on the PDE app. Depending on your MATLAB version, you might be able to use the following instead of your current app-based script, at least in part.
model = createpde();
gd = [3,4,0,1,1,0,0,0,1,1]'; % square geometry
dl = decsg(gd);
geometryFromEdges(model,dl); % include geometry
generateMesh(model,'Hmax',0.1);
specifyCoefficients(model,'m',0,'d',1,'c',2,'a',1,'f',0);
setInitialConditions(model,0);
tlist = 0:0.01:100; % do you really need such fine spacing?
for j = 1:5
gfun = @(region,state)sin(2*j*state.time); % one nondefault condition
applyBoundaryCondition(model,'neumann','edge',4,'g',gfun);
result{j}=solvepde(model,tlist); % save all results in a cell array
delete(model.BoundaryConditions); % prepare for next loop
end
To see a plot of the first loop at the 500th time point:
pdeplot(model,'xydata',result{1}.NodalSolution(:,500))
Good luck,
Alan Weiss
MATLAB mathematical toolbox documentation
