The function lh-zd
Global Mu Sigama % define a global variable
2.5 Mu=[10];
% input averageSigama=0.375 [2]; % input standard deviation
X0=[10; 2.5]; % input initial iteration point
A=[]; B=[]; Aeq=[]; Beq=[];
Lb=[]; Ub=[];
The options=optimset (' Algorithm ', 'active - set');
[X, fval, exitflag, output]=fmincon (@ bata2, X0, A, b, Aeq, beq, lb, ub, @ st, options) % call optimization toolbox
Bata=SQRT (fval)
Pf=CDF (' norm '- bata, 0, 1) % to calculate failure probability
End
The function of CC=bata2 (X) % target functions child
Global Mu Sigama % define a global variable
CC=((X (1) - Mu (1))/Sigama (1)) ^ 2 + ((X (2) - Mu (2))/Sigama (2)) ^ 2; % X are normal distribution
End
Yet function [c, ceq] (X)=st % son constraint function
C=[];
% nonlinear inequality constraintsCeq=stBplm (X);
% nonlinear equality constraintsEnd
The function of Z=stBplm (X)
W1=[0.1242 1.9466; 67.3014 22.5133; 90.9424 to 20.8143];
B1=[1.5036, 4.0199, 3.0905];
W2=[552.2048, 28.3283 to 130.4954];
B2=[10.8555];
XX=[X (1), X (2)];
Z=simuff (XX, w1, b1, 'logsig, w2, b2, and' purelin ');
End
The function (w2 w1, b1, b2)=trainBplm (X);
X1=16 4 10 10 [10];
X2=[2.5 2.5 2.5 3.625 1.375];
P=(X1, X2),
T=[13.6728, 10.8005, 16.5451, 16.8897, 10.3136].
S1=3;
Tp=0.0000001] [1, 1000;
[w2 w1, b1, b2]=initff (p, s1, 'logsig, t and' purelin ');
W2 [w1, b1, b2]=trainlm (w1, b1, 'logsig, w2, b2, and' purelin ', p, t, tp);
End

"" there is an error is as follows:
Ceq=stBplm (X);
% nonlinear equality constraints
Error fmincon (line 622)
[CTMP, ceqtmp]=feval (confcn {3}, X, varargin {that});

Error rgsj_xym_lwltff (line 9)
[X, fval, exitflag, output]=fmincon (@ bata2, X0, A, b, Aeq, beq, lb, ub, @ st, options) % call optimization toolbox

Reason:
Failure in initial nonlinear constraint function evaluation. FMINCON always continue.