The NEOS Server offers
Knitro for the solution of general mixed-integer nonlinearly
constrained optimization problems that can be modeled in
AMPL format. Knitro also can be
used to solve simpler classes of problems, such as unconstrained problems,
bound constrained problems, linear programming (LP) problems and
quadratic programming (QP) problems.
Knitro was developed by Richard Byrd, Mary Beth Hribar, Jorge Nocedal and Richard Waltz with additional help from Guanghui Liu, Marcelo Marazzi, Todd Plantenga and Jose Luis Morales. For a detailed description of Knitro, see R. H. Byrd,
J. Nocedal, and R. A. Waltz. 2006. KNITRO: An Integrated Package for Nonlinear Optimization. In: G. di Pillo and M. Roma, editors. Large-Scale Nonlinear Optimization. Springer-Verlag. p. 35-39.
Knitro is available from Artelys. For more information on Knitro, see the Knitro website.
Using the NEOS Server with Knitro
The user must submit a model in AMPL
format to solve a mixed-integer nonlinearly constrained optimization (MINCO)
section of the AMPL website provides examples of models in AMPL format.
The MINCO problem must be specified by a model file with the options of a data file and a commands file. If the commands file is specified, it must
contain the AMPL solve command. However, the commands file must not
contain the model or data commands. The model and data files
are renamed internally by NEOS.
Knitro offers both interior-point and active-set methods. The user may
specify the algorithm by setting the Knitro/AMPL option "alg" to one of the following values:
0: automatic algorithm selection (default)
1: Interior/Direct algorithm
2: Interior/CG algorithm
3: Active algorithm
The commands file may include AMPL commands and option settings for Knitro.
To specify solver options, add
option knitro_options "OPTIONS";
OPTIONS is a list of one or more of the Knitro options.
See the Knitro/AMPL Reference Manual for a list of the
available Knitro options.