ANN for constraint optimization problem

hello,

How do i modify my ANN algorithm, by incorporating some constraints to perfom my obtained result. the matlab code used is generated from Neural Network Toolbox.

 function [Y,Xf,Af] = ANN_Function(X)
% ===== NEURAL NETWORK CONSTANTS =====

 % Input 1
 x1_step1_xoffset = [-0.964339227389144;-0.906927494859787;-0.9643288955237];
 x1_step1_gain = [1.03826501814344;1.06861700687107;1.27340625419393];
 x1_step1_ymin = -1;

 % Layer 1
 b1 =  [-1.3371193654695128217;-3.7243723447930885406;0.59020209217794505907;-0.26942 281438778381553;-0.084790990077469749475;-0.287833517416886564;-0.6199588700217847359;0.87036361559242081398;1.1136596091577191103;3.5728692098803582766];
 IW1_1 = [0.62172489473929637427 -1.1539538428995010921 0.40749280432490736503;0.33504097052172682192 2.2754223181903578954 -2.0670552187061321803;-0.97421719994837163714 0.16930737512492399777 0.95638889040809083042;3.453458645614659428 2.4120651149524281465 3.9342055592145062093;2.9998512477370034013 2.2723487424133810286 3.3571622229587347874;-1.6838243946405258011 -2.3995748128279066336 0.36244936598086713309;-3.1423700210823852785 1.3845185332230820485 2.4609517018642876884;-0.39703700607817143942 2.4912130193995269956 -0.16941481846512243536;1.406655675671569572 0.92534256006865256428 0.64325984129225455277;1.5085377777493140794 1.4551527878904193525 -1.3090133612083598713];

 % Layer 2
 b2 = [-0.28647967632293369622;-0.77689684809120063136;0.23567045137827014045;-0.50614562961496167848;0.17775471570313430836;0.39409286123122444501;-0.23743319675300361693];
 LW2_1 = [-1.2595809845665446591 0.52253536564831837286 1.6575450213582203496 1.4444644615739332671 -0.79927746466752380705 -0.49440949393277561219 -1.0291039534117272236 -0.10205229528755178914 -1.783179386992490123 0.012007511957539542674;-1.8160587258194498261 0.16525212307258660416 -0.35634974797682900105 0.61911944611294977836 -0.55104180525241264199 0.36270218166368617396 0.70227078264624087645 -0.69192422904692441055 0.63741286998972901401 0.078380036138073788665;-0.39987089599785630156 0.0015263660888982231219 -0.32703999255710186622 -1.0870691355791675115 0.68183572942635206626 0.20086063975571505358 0.53707692508704663048 -0.043958739282595582498 0.30170293659756891591 -0.40387165805595148793;-0.42538078713443683299 -0.41243046298959784579 0.09360967585146644232 -0.54902810742656438237 0.27858672713463300541 0.13466521369669071095 -0.31684402239022979586 -0.016110146899087046668 0.18209067234932024837 -0.38144681930811946691;0.78489623544097297803 0.064197561813166229006 -0.14703146723552590336 0.045763727795912277629 -0.044493781097574965078 0.023102496123388799321 0.19498160438730580135 0.42399577006788119471 0.068627392557574939946 0.2774573470874529546;0.28300904777281810087 -0.040045961378933660202 -0.048322484065649526364 0.23633721304470370339 -0.28761021912207535012 0.22386822762508534757 -0.095795332741767574847 0.31901549272383256106 0.17453188621906071121 -0.40095786179528114523;0.2537157491011907684 -0.037837226082635302959 -0.044856312727524147443 0.24018253583883578117 -0.28974054247799801987 0.23561178051146219881 -0.095989508380685484301 0.30576014862164185848 0.18195538033206490325 0.22157365239611195862];

 % Output 1
 y1_step1_ymin = -1;
 y1_step1_gain = [1.21212121212121;1.53846153846154;0.8;3.33333333333333;1.66666666666667;2;2];
 y1_step1_xoffset = [-0.785398163397448;-0.523598775598299;-0.261799387799149;-0.174532925199433;-0.349065850398866;-0.523598775598299;-0.523598775598299];

 % ===== SIMULATION ========

 % Format Input Arguments
 isCellX = iscell(X);
 if ~isCellX, X = {X}; end;

 % Dimensions
 TS = size(X,2); % timesteps
 if ~isempty(X)
    Q = size(X{1},1); % samples/series
 else
    Q = 0;
 end

 % Allocate Outputs
 Y = cell(1,TS);

 % Time loop
 for ts=1:TS

    % Input 1
    X{1,ts} = X{1,ts}';
    Xp1 =  
 mapminmax_apply(X{1,ts},x1_step1_gain,x1_step1_xoffset,x1_step1_ymin);

    % Layer 1
     a1 = tansig_apply(repmat(b1,1,Q) + IW1_1*Xp1);

    % Layer 2
    a2 = repmat(b2,1,Q) + LW2_1*a1;

    % Output 1
    Y{1,ts} = 
 mapminmax_reverse(a2,y1_step1_gain,y1_step1_xoffset,y1_step1_ymin);
    Y{1,ts} = Y{1,ts}';
end

% Final Delay States Xf = cell(1,0); Af = cell(2,0);

% Format Output Arguments if ~isCellX, Y = cell2mat(Y); end end

% ===== MODULE FUNCTIONS ========

% Map Minimum and Maximum Input Processing Function function y = mapminmax_apply(x,settings_gain,settings_xoffset,settings_ymin) y = bsxfun(@minus,x,settings_xoffset); y = bsxfun(@times,y,settings_gain); y = bsxfun(@plus,y,settings_ymin); end

% Sigmoid Symmetric Transfer Function function a = tansig_apply(n) a = 2 ./ (1 + exp(-2*n)) - 1; end

% Map Minimum and Maximum Output Reverse-Processing Function function x = mapminmax_reverse(y,settings_gain,settings_xoffset,settings_ymin) x = bsxfun(@minus,y,settings_ymin); x = bsxfun(@rdivide,x,settings_gain); x = bsxfun(@plus,x,settings_xoffset); end

[Greg Heath]

Why are you dissatisfied with your current code?

Greg

[studentU]

Thank you for your repply,

The current code is made in general case without take into account my fitness function and constraints to obtain the optimal solution!!!

it's work well, but i want to modify it in order to incorpor the constraints.... how can i do it? is it the recurent neural network the best solution in neural network constraint optimizing? Do you have some idea to this perpose?