Induction Heating Process: 3D Modeling and Optimisation

International audienceAn increasing number of problems in mechanics and physics involves multiphysics coupled problems. Among these problems, we can often find electromagnetic coupled problems.Electromagnetic couplings may be involved through the use of direct or induced currents for thermal purposes--in order to generate heat inside a work piece in order to get either a prescribed temperature field or some given mechanical or metallurgical properties through an accurate control of temperature evolution with respect to time-, or for solid or fluid mechanics purposes--in order to create magnetic forces such as in fluid mechanics (electromagnetic stirring,...) or solid mechanics (magnetoforming,...).Induction heat treatment processes is therefore quite difficult to control; trying for instance to minimize distortions generated by such a process is not easy. In order to achieve these objectives, we have developed a computational tool which includes an optimsation stage.A 3D finite element modeling tool for local quenching after induction heating processes has already been developed in our laboratory. The modeling of such a multiphysics coupled process needs taking into account electromagnetic, thermal, mechanical and metallurgical phenomenon--as well as their mutual interactions during the whole process: heating and quenching. The model developed is based on Maxwell equations, heat transfer equation, mechanical equilibrium computations, Johnson-Mehl-Avrami and Koistinen-Marburger laws. All these equations and laws may be coupled but some coupling may be neglected.In our study, we will also focus on induction heating process aiming at optimising the Heat Affected Zone (HAZ). Thus problem is formalized as an optimization problem--minimizing a cost function which measures the difference between computed and optimal temperatures--along with some constraints on process parameters.The optimization algorithms may be of two kinds--either zero-order or first-order algorithms. First-order algorithms have proved their efficiency for induction heating processes. However, zero-order algorithms--such as evolution strategy algorithms--are better at reaching global minimal values for cost functions. We use here a method based on Efficient Global Optimization algorithm developed by Jones which is an optimization procedure assisted by a meta model.We will present some results obtained with our numerical tool on the preheating of a part before forging. We will focus on the control of temperature profile at the end of induction heating stag