amplexx2.pro~orig

sqp程序包。用sqp算法实现非线性约束的优化求解

donlp2, v3, 05/29/98, copyright P. Spellucci
Thu Feb 24 16:52:52 2000
amplex_2

     n=         5    nh=         2    ng=         1

  epsx= 1.000e-08 sigsm= 1.490e-08

startvalue
  0.0000000e+00   0.0000000e+00   0.0000000e+00   0.0000000e+00   0.0000000e+00 

  eps=  2.22e-16  tol= 1.98e-323 del0=  1.00e+00 delm=  1.00e-06 tau0=  1.00e+08
  tau=  1.00e-01   sd=  1.00e-01   sw=  3.67e-11  rho=  1.00e-06 rho1=  1.00e-10
 scfm=  1.00e+04  c1d=  1.00e-02 epdi=  1.00e-08
  nre=         5 anal=         0
 vbnd=  1.00e+00 efcn=  1.00e-16 diff=3

 termination reason:
 KT-conditions satisfied, no further correction computed
 evaluations of f                          285
 evaluations of grad f                       0
 evaluations of constraints                135
 evaluations of grads of constraints         0
 final scaling of objective           1.000000e+00
 norm of grad(f)                      8.470486e+00
 lagrangian violation                 6.025075e-08
 feasibility violation                8.881784e-16
 dual feasibility violation           0.000000e+00
 optimizer runtime sec's              1.000000e-02


 optimal value of f =   8.66056499393680e-01

 optimal solution  x =
  6.90794648980667e-01 -3.84144777862419e-01 -1.23560911640061e+00
  7.99409837243688e-01 -7.90870227443600e-01

  multipliers are relativ to scf=1
  nr.    constraint      normgrad (or 1)   multiplier
    1  -4.4408921e-16    3.2958535e+00   -7.2132239e-01
    2   4.4408921e-16    2.3806589e+00    3.0970073e-02
    3   0.0000000e+00    1.5047813e+00    5.3233615e+00

 evaluations of restrictions and their gradients
 (    45,     0) (    45,     0) (    45,     0)
last estimate of condition of active gradients  2.330e+00
last estimate of condition of approx. hessian   6.619e+01
iterative steps total               8
# of restarts                       0
# of full regular updates           8
# of updates                        8
# of full regularized SQP-steps     0