Abstract The paper presents the foundations of an optimization system for medium size computers. The system may be used to solve technical or economic problems with nonlinear objective functions and constraints. The system contains five algorithms of the deterministic and stochastic category. The master algorithm, based on works of Bush and Mosteller, randomly seeks one of these five. It performs computations for a certain time and examines its effectiveness. This experience is then taken as the basis for modification of the probability vector used in the choice of algorithm. The technical examples solved by different algorithms are presented. The results needed for definition of the initial vector illustrate difficulties met by an engineer in the proper choice of algorithm. These difficulties, which occur often in practical cases, prove the demand for further studies on optimization systems.
[1]
Jan Golinski,et al.
Optimal synthesis problems solved by means of nonlinear programming and random methods
,
1970
.
[2]
R. E. Kalman,et al.
Optimum Seeking Methods.
,
1964
.
[3]
Raymond A. Mugele,et al.
A nonlinear digital optimizing program for process control systems
,
1962,
AIEE-IRE '62 (Spring).
[4]
Robert Ridgley Slaymaker.
Mechanical Design and Analysis
,
1959
.
[5]
Frederick Mosteller,et al.
Stochastic Models for Learning
,
1956
.
[6]
H. H. Rosenbrock,et al.
An Automatic Method for Finding the Greatest or Least Value of a Function
,
1960,
Comput. J..
[7]
M. J. Box.
A New Method of Constrained Optimization and a Comparison With Other Methods
,
1965,
Comput. J..