🚀 go-pugleaf

In article <1992Nov6.173939.10662@usl.edu>, tt@swamp.cacs.usl.edu (Thitipong Tanprasert) writes:
|> Hi,
|> 
|>     I'm not a math student and just started to know about numerical
|> optimization a couple of weeks ago. However, I need to optimize an upto
|> 200 variables nonlinear (quadractic and maybe trigonometric) objective
|> functions under many nonlinear (trigonometric) constraints. I have no
|> expectation to find global optimum. Just hope to obtain some good
|> acceptable results. Would anyone please help suggesting a reliable 
|> constrained optimization method that I may program to deal with my
|> problem. I was suggested about Penalty Function and think it's quite
|> interesting. However, I have no idea if some other methods will
|> work better (generally). Also, if I use Penalty Function method, what
|> unconstrained optimization method should I attempt to implement? Is
|> there a ftp-able software about this that I may use directly without
|> having to program myself?
|> 
There may be public domain programs for minimizing subject to
nonlinear constraints in NETLIB:
Otherwise there are excellent routines in NAG and IMSL. 
I have fortran routines for same problem I wrote many years ago
that are not as efficient or reliable you are welcome to
try if you can not get help from elsewhere.


-- 
Murray Dow                         GPO Box 4 Canberra ACT 2601 Australia
Supercomputer Facility             Phone: +61 6 2495028
Australian National University     Fax:   +61 6 2473425
mld900@anusf.anu.edu.au