Optimoptions - Invalid solver specified error

Hi, please can someone have a look at mycode to see what's wrong. I keep getting this error message. Thank you

Error using optimoptions

Invalid solver specified. Provide a solver name or handle (such as 'fmincon' or

@fminunc).

Type DOC OPTIMOPTIONS for a list of solvers.

Error in solveModel (line 57)

opt = optimoptions(@fsolve,'Display','iter-detailed','FunValCheck','on','MaxFunctionEvaluations',10000000,'MaxIteration',maxIter,'StepToleance',1e-6,'FunctionTolerance',1e-6);,

%% Solve membrane model
% Description: This script defines feed and membrane parameters, the number
% of collocation points, initial values of the dimensionless model
% variables, and solves the model equations for membrane with
% counter-current flow pattern. 
%
% Variables:
% HFperm :: Permeances for the hollow fiber carbon membrane for [O2 N2]
% param :: struct consisting of the following membrane parameters
%   perm - matrix of the permeances [kmol/kPa*h*m2]
%   area - effective membrane area [m2]
%   R - universal gas constant [m3*kPa/(K*mol)]
% 
% Feed :: strict consisting of the following feed parameters
%   Ns - Shell side flow rate for [O2 N2]
%   Nt - Fiber side folw rate for [O2 N2]
%   Ps - Feed side pressure [kPa]
%   Pt - Permeate side pressure [kPa]
clc
clear all

%% Creating a param struct where membrane parameters are stored.
HFperm = 1e-03*[16.8e-03 3.36e-03];
param.Perm = HFperm;
param.Area = 47.7;
param.Visc = 14.9e-09/3600;
param.R = 8.314;

%% Creating a feed struc where known values for the feed at shell and tube side are stored.
feed.Ns = 0.44*[0.209 0.791];
feed.Nt = zeros(1,2);
feed.Ps = 1034;
feed.Pt = 103.4;
feed.Ts = 273.15 + 20;

%% Orthogonal collocation method and Fsolve
% Discretize and approximate model equation residuals

n = 2;
z = [0 1/2-sqrt(21)/14 1/2 1/2+sqrt(21)/14 1];
A = [0.278 -0.202 0.169 -0.071; ...
    0.398 0.069 0.064 -0.031; ...
    0.387 0.234 0.278 -0.071;
    0.389 0.222 0.389 0.000];

Y0=ones(n+2,4);
%Initial values
Y0(:,1) = 0.4;
Y0(:,2) = 0.1;

Y0(:,3) = 0.03;
Y0(:,4) = 0.97;


%Optimization settings for the solver
maxIter = 2000;
opt = optimoptions(@fsolve,'Display','iter-detailed','FunValCheck','on','MaxFunctionEvaluations',10000000,'MaxIteration',maxIter,'StepTolerance',1e-6,'FunctionTolerance',1e-6);
t0 = tic;
[Y, fval, exitflag, output]=fsolve(@(Y) model_MB(Y, A, n, param, feed),Y0,opt);

                                         Norm of      First-order   Trust-region
 Iteration  Func-count     f(x)          step         optimality    radius
     0         17         1.26878                          1.19               1
     1         34        0.519374              1          0.337               1
     2         51        0.428702            2.5          0.569             2.5
     3         68        0.334533            2.5          0.146             2.5
     4         85        0.287467           6.25          0.303            6.25
     5        102         0.25163           6.25           0.12            6.25
     6        119        0.232424         15.625          0.452            15.6
     7        136        0.203663         15.625          0.101            15.6
     8        153        0.171875        39.0625          0.954            39.1
     9        170        0.121609        39.0625          0.169            39.1
    10        187       0.0961458        97.6563          0.571            97.7
    11        204       0.0671728        97.6562          0.118            97.7
    12        205       0.0671728        244.141          0.118             244
    13        222       0.0542653        61.0352          0.438              61
    14        239       0.0281923        152.588           2.83             153
    15        240       0.0281923        380.885           2.83             381
    16        257       0.0157762        95.2212          0.569            95.2
    17        274     0.000715115        238.053          0.106             238
    18        275     0.000715115        48.8918          0.106             595
    19        292     0.000384456        12.2229            0.2            12.2
    20        309      3.6133e-05        30.5574           1.37            30.6
    21        310      3.6133e-05         6.4665           1.37            76.4
    22        327      6.9901e-06        1.61662           0.39            1.62
    23        344     1.25641e-06        4.04156           13.8            4.04
    24        345     1.25641e-06       0.512592           13.8            4.04
    25        362     7.58608e-08       0.128148           4.38           0.128
    26        363     7.58608e-08       0.132648           4.38            0.32
    27        380     6.66623e-09      0.0331621           1.52          0.0332
    28        381     6.66623e-09       0.075564           1.52          0.0829
    29        398     1.17994e-09       0.018891          0.397          0.0189
    30        415     1.00171e-09      0.0472275           26.8          0.0472
    31        416     1.00171e-09      0.0472275           26.8          0.0472
    32        433     8.77872e-10      0.0118069           26.7          0.0118
    33        434     8.77872e-10      0.0118069           26.7          0.0118
    34        451     7.89096e-11     0.00295172           8.05         0.00295
    35        452     7.89096e-11     0.00737929           8.05         0.00738
    36        469     2.51055e-11     0.00184482           4.51         0.00184
    37        486     2.26848e-11     0.00184482           4.31         0.00184
    38        487     2.26848e-11     0.00184482           4.31         0.00184
    39        504     3.02338e-12    0.000461206           1.29        0.000461
    40        505     3.02338e-12     0.00115301           1.29         0.00115
    41        522     1.68554e-12    0.000288254          0.714        0.000288
    42        539     1.62199e-12    0.000288254          0.677        0.000288
    43        540     1.62199e-12    0.000288254          0.677        0.000288
    44        557     1.14061e-12    7.20634e-05          0.202        7.21e-05

fsolve stopped because the sum of squared function values, r, changed by 2.406920e-13
relative to its initial value; this is less than max(options.FunctionTolerance^2,eps) = 1.000000e-12.
r = 1.140608e-12, is less than sqrt(options.FunctionTolerance) = 1.000000e-03.

tf = toc(t0);
disp(tf);

    0.2754

%Plotting the results
Area = param.Area*z;
Res = sum(feed.Ns)*Y(:,1:4);
modelPlot(Area, Res)

Unrecognized function or variable 'modelPlot'.

function F=model_MB(Y,A,n, param, feed)
%Dimensionless variables
Nty = Y(:,1:2);
Nsx = Y(:,3:4);
Pt = feed.Pt/feed.Ps;
Ps = feed.Ps/feed.Ps;
%Dimensionless constants
Kj = param.Area*feed.Ps*param.Perm(1,:)/(sum(feed.Ns));
%Preallocates memory for the set of residuals
F = zeros(n+2,4);
%%Boundary conditions at z=0
dJ1 = (Ps*(Nsx(1,:)/sum(Nsx(1,:)))-Pt*(Nty(1,:)/sum(Nty(1,:))));
F(1,1:2) = Nty(1,:)-feed.Nt/sum(feed.Ns);
F(1,3:4) = A(1,:)*Nsx - Kj.*dJ1;
%%Inner collocation points:
for i=2:n+1
    dJi = (Ps*Nsx(i,:)/sum(Nsx(i,:)) - Pt*Nty(i,:)/sum(Nty(i,:)));
    F(i,1:2) = A(i,:)*Nty - Kj.*dJi;
    F(i,3:4) = A(i,:)*Nsx - Kj.*dJi;
end
%%Boundary conditions at last collocation point, z=n+2
dJn = (Ps*Nsx(n+2,:)/sum(Nsx(n+2,:))-Pt*Nty(n+2,:)/sum(Nty(n+2,:)));
F(n+2,1:2) = A(n+2,:)*Nty - Kj.*dJn;
F(n+2,3:4) = Nsx(n+2,:) - feed.Ns/sum(feed.Ns);
end

Answer 1

[Bruno Luong]

Check if you have license and installed optimization toolbox by typing

ver

Answer 2

[Torsten]

You wrote

'StepToleance'

instead of

'StepTolerance'

...

[Trang]

Thank you! But I still get the same error message.

[Torsten]

I don't (see your code above).

Only your model function is missing to test further.

[Trang]

This is my model_MB. I don't understand why I still see the error.

function F=model_MB(Y,A,n, param, feed)

%Dimensionless variables

Nty = Y(:,1:2);
Nsx = Y(:,3:4);
Pt = feed.Pt/feed.Ps;
Ps = feed.Ps/feed.Ps;

%Dimensionless constants
Kj = param.Area*feed.Ps*param.Perm(1,:)/(sum(feed.Ns));

%Preallocates memory for the set of residuals
F = zeros(n+2,4)

%%Boundary conditions at z=0
dJ1 = (Ps*(Nsx(1,:)/sum(Nsx(1,:)))-Pt*(Nty(1,:)/sum(Nty(1,:))));

F(1,1:2) = Nty(1,:)-feed.Nt/sum(feed.Ns);
F(1,3:4) = A(1,:)*Nsx - Kj.*dJ1;

%%Inner collocation points:
for i=2:n+1
    dJi = (Ps*Nsx(i,:)/sum(Nsx(i,:)) - Pt*Nty(i,:)/sum(Nty(i,:)));
    F(i,1:2) = A(i,:)*Nsy - Kj.*dJi;
    F(i,3:4) = A(i,:)*Nsx - Kj.*dJi;
end

%%Boundary conditions at last collocation point, z=n+2
dJn = (Ps*Nsx(n+2,:)/sum(Nsx(n+2,:))-Pt*Nty(n+2,:)/sum(Nty(n+2,:)));
F(n+2,1:2) = A(n+2,:)*Nty - Kj.*dJn;
F(n+2,3:4) = Nsx(n+2,:) - feed.Ns/sum(feed.Ns);
end

[Torsten]

I can't see your complete line of code

opt = optimoptions(...)

from your .png file.

New error (see above).

And "modelPlot" is missing.

[Steven Lord]

function F=model_MB(Y,A,n, param, feed)

%Dimensionless variables

Nty = Y(:,1:2);
Nsx = Y(:,3:4);
Pt = feed.Pt/feed.Ps;
Ps = feed.Ps/feed.Ps;

%Dimensionless constants
Kj = param.Area*feed.Ps*param.Perm(1,:)/(sum(feed.Ns));

%Preallocates memory for the set of residuals
F = zeros(n+2,4)

%%Boundary conditions at z=0
dJ1 = (Ps*(Nsx(1,:)/sum(Nsx(1,:)))-Pt*(Nty(1,:)/sum(Nty(1,:))));

F(1,1:2) = Nty(1,:)-feed.Nt/sum(feed.Ns);
F(1,3:4) = A(1,:)*Nsx - Kj.*dJ1;

%%Inner collocation points:
for i=2:n+1
    dJi = (Ps*Nsx(i,:)/sum(Nsx(i,:)) - Pt*Nty(i,:)/sum(Nty(i,:)));
    F(i,1:2) = A(i,:)*Nsy - Kj.*dJi;

Where in your code have you defined the Nsy variable used on this line? Should this be Nsx or Nty?

[Trang]

Thank you, it should be Nty. Although I'm confused since I don't get the same message as Torsen showed above but keep getting the same error with the optimoptions function. Pls see pics. I used the same code with mistakes pointed out corrected.

[Torsten]

Corrected and code works now.

I still can't read the optimoptions line of your code from the image.

As Bruno Luong noted you should check whether you have licensed and installed the optimization toolbox.

See if this simple example works with your MATLAB:

fun = @(x)x.^2-1;
sol = fsolve(fun,2)

Equation solved.

fsolve completed because the vector of function values is near zero
as measured by the value of the function tolerance, and
the problem appears regular as measured by the gradient.

sol = 1.0000