The purpose of this paper is to present two optimization methodologies based on interval branch‐and‐bound algorithm.
These techniques decrease the total time of computation, even in spite of discrete nature of some of the design variables. Computational experiments performed on multivariable optimization problem reveal great accuracy and technical validity of developed approaches. As an example, the optimal design of the induction machine (IM) was treated, where the aim was to find the set of the most efficient and, at the same time, cheapest in the manufacturing configurations.
In this paper, two approaches were developed for resolving the problem of optimal design of IM with discrete variables. The strategy of constructing the meta‐models is utilized and put in practice. The methods show relatively high efficiency and robustness of obtained results.
These approaches are the core technics of the developed industrial application, which help identify the set of optimal configurations of IM with the criteria of optimality such as total cost of manufacturing and the efficiency of IM.
