Multi-objective optimisation of induction heating processes: methods of the problem solution and examples based on benchmark model

Abstract

The main goal of the researches is the development of new approaches, algorithms and numerical techniques for multi–objective optimisation of design of industrial induction heating installations. A multi–objective optimisation problem is mathematically formulated in terms of the typical optimisation criteria, e.g., maximum heating accuracy and minimum energy consumption. Various mathematical methods and algorithms for multi–objective optimisation, such as Non–dominated Sorting Genetic Algorithm (NSGA–II) and optimal control alternance method, have been implemented and integrated in a user–friendly automated optimal design package. Several optimisation procedures have been tested and investigated for a problem–oriented mathematical model in a number of comparative case studies. A general comparison of the design solutions based on NSGA–II and alternance method leads to their good agreement in all investigated cases. The methodology developed is planned to be applied to more complex real–life problems of the optimal design and control of different induction heating systems.