Equivalent function NCONF/NLQPL

[CHARAF-EDDINE_A_]

hello folks!

I am looking for an equivalent function to NCONF/DNCONF which has been defined in the IMSL library. Here is the description of this function:

Solve a general nonlinear programming problem using the successive quadratic programming algorithm and a finite difference gradient.

Call NCONF (FCN, M, ME, N, XGUESS, IBTYPE, XLB, XUB, XSCALE, IPRINT, MAXITN, X, FVALUE)

In nutshell, I have to minimize a multi-variable function knowing that one (or more) condition have to be accomplished.

Thank you in advance.

[Steven_L_Intel1]

A comment that NCONF/DNCONF are not in the IMSL we offer due to a third-party licensing restriction. They have been labeled as "deprecated" since IMSL 2.0.

[mecej4 · Accepted solution]

The currently available subroutine in IMSL with the same functionality is NNLPF. The argument list and the interface for the user-supplied function subroutines are, of course, slightly different.

The current (updated) versions of NLPQLP may be obtained from the author, Professor Schittkowski, at http://www.ai7.uni-bayreuth.de/nlpqlp.htm.

If your problem involves mininizing the sum of squares of a set of functions with, at most, linear constraints, look at TRNLSPBC in MKL.

[Steven_L_Intel1]

Do you mean the TRNLSPBC routines? In the MKL index under TR Routines > Nonlinear Least Squares Problem with Linear (Bound) Constraints

[CHARAF-EDDINE_A_]

YES, I found it. Unfortunately, I have to deal with non linear constraints...

[Steven_L_Intel1]

I'm moving this to the MKL forum.