Optimization of induction heating using numerical modeling and genetic algorithm

Induction heating is a complex interplay of electromagnetic, thermal and metallurgic phenomena. It is therefore not surprising that only few numerical optimization approaches for solving induction heating problems have been published. The aim of our study was to determine the optimal coil position and the optimal amplitude and frequency of the electric current in order to attain the desired temperature profile in a non-ferromagnetic steel cylinder workpiece. The finite-element numerical modeling in combination with genetic algorithm optimization method was used for this purpose. The heated workpiece was surrounded by a copper coil consisting of four loops. In our calculations the workpiece's temperature-dependent material properties were taken into account. Using a genetic algorithm the following parameters were optimized: position of the single-coil loop, the amplitude and the frequency of the electric current in the coil. Our numerical model was experimentally validated by comparison of measurement and simulation results. The optimized solution and the global optimum had comparable temperature profiles due to a better selection of the electric current parameters compared to the profile obtained by solving the problem intuitively. We demonstrated that the proposed approach can be used for planning of induction heating of steel materials at low-energy consumption and high time-efficiency.