UDC 004.896
MATCHING NETWORKS SYNTHESIS BASED ON METAHEURISTIC MIXED-INTEGER MULTIOBJECTIVE ALGORITHM
M. V. Cherkashin, Ph.D. (in technical sciences), associate professor, Department of Computer Control and Design Systems, TSUCSR, Tomsk, Russia; orcid.org/0000-0001-7536-363X, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
K. S. Sarin, Ph.D. (in technical sciences), associate professor, Department of Computer Control and Design Systems, TSUCSR, Tomsk, Russia;
orcid.org/0000-0003-3264-7768, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
I. A. Hodashinsky, Dr. in technical sciences, full professor, Department of Computer Control and Design Systems, TSUCSR, Tomsk, Russia;
orcid.org/0000-0002-9355-7638, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
M. O. Svetlakov, Ph.D. (in technical sciences), associate professor, Department of Computer Control and Design Systems, TSUCSR, Tomsk, Russia; orcid.org/0000-0002-5060-0958, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Synthesis of broadband matching circuits is an important task in the construction of radio technical and telecommunication systems and devices. Usually, the requirements to matching networks are contradictory in terms of quality of the result obtained, which complicates the automation of design process. One of the possible approaches to solving it is to use multiobjective heuristic algorithms that allow combining the pro cedures of structural and parametric synthesis, but the efficiency of such algorithms strongly depends on the method of target function forming and methods of its optimization. The aim of the work is to develop a method for structural-parametric synthesis of broadband matching circuits based on the use of mixed multi objective optimization that searches in space with discrete and real variables. To solve the optimization problem, a cuckoo search metaheuristic algorithm with genetic crossover operator was used. The method allows simultaneously constructing the structure and adjusting the parameters of matching circuit elements. The optimization goals are structure complexity and the deviation of amplitude-frequency characteristic from the required value. As a result of the algorithm, many circuit variants that are not dominated by these goals are formed. Based on the requirements for the device being developed and the compromise between its accuracy and complexity, an expert selects an appropriate option. Several typical examples were solved us ing the method developed: single match problem (Fano problem), double match problem, and matching equalizing problem. The results of the experiment showed high efficiency of constructed circuits in compari son with the circuits of alternative synthesis methods.
Key words: : matching network, broadband matching, synthesis of matching network, metaheuristic algo rithm, multiobjective optimization, mixed-integer optimization, genetic algorithm, cuckoo search metaheuristic.