Multiphysics field analysis and multiobjective design optimization: a benchmark problem (original) (raw)
Related papers
COUPLED VI : proceedings of the VI International Conference on Computational Methods for Coupled Problems in Science and Engineering, 2015
In the paper, a bi-objective optimization problem characterized by a multi-physics field analysis is investigated. The optimal design of a pancake inductor, related to the design of industrial devices for the controlled heating of a graphite disk is considered as the benchmark problem. The expected goal of the optimization process is to improve temperature uniformity in the disk as well as electrical efficiency of the inductor. The optimized device is designed using a multi-physics problem: magnetic problem for electrical efficiency computation and thermal problem for temperature uniformity computation. The solution of the relevant bi-objective optimization problem is based on a modified multi-objective genetic algorithm in the class of Non-dominated Sorting Genetic Algorithm. The proposed algorithm exploits the migration concept to vary the population genetic characteristics during optimization process in order to improve the Pareto front approximation.
Sensitivity-based optimal shape design of induction-heating devices
IET Science, Measurement & Technology, 2015
A Design Of Experiment (DOE) strategy applied to multiobjective optimization is proposed in order to evaluate the influence of design variables variations to optimized quantities. A secondary objective function is the sensitivity of a primary objective function to design variable variations evaluated by means of DOE strategy. The optimization problem includes also a third objective function that considers device constraint due to technological limitations on power generator. The proposed case study deals with the design of an electromagnetic device that will be used to carry out laboratory experiments on magneto fluid hyperthermia, that is a clinic treatment for cancer cure. The induction system is designed to apply a controlled time varying magnetic field to biological cells, cultured in Petri dish, mixed with magnetic nanoparticles. The paper presents an original cost-effective method of multi-objective design optimization taking into account design uncertainties.
Automated optimal design techniques for inverse electromagnetic problems
IEEE Transactions on Magnetics, 1992
The solution of Inverse Electromagnetic problems requires to minimize or maximize a Cost Function, related to field quantities. The choice of the best algorithm to be used depends on several factors, e.g. the Cost Function complexity and the number of Design Variables. The paper presents two methods based on different approaches. The first, based on a deterministic approach, considers a quadratic approximation of the Cost Function, whose minimum is faster to calculate. The second, based on a stochastic approach, is derived from the Simulated Annealing algorithm. Both methods, implemented as computer codes, have been applied for the solution of a test synthesis problem where the magnetic field is generated by discrete coils. The results are compared and discussed.
Optimal shape design of devices and systems for induction‐heating: methodologies and applications
COMPEL - The international journal for computation and mathematics in electrical and electronic engineering, 2003
The paper reports recent experiences of the authors in the automated optimal design of devices and systems for induction heating. The results presented have been obtained in the frame of a long‐lasting cooperation between Laboratory of Electroheat, University of Padova and Electromagnetic Devices CAD Laboratory, University of Pavia. In particular, two case studies are discussed; in both cases, the shape design of the inductor is carried out in a systematic way, by minimizing user‐defined objective functions depending on design variables and subject to bounds and constraints. When the design problem is characterized by many objectives which are in mutual conflict, the non‐dominated set of solutions is identified.
Multiobjective optimization in magnetostatics: a proposal for benchmark problems
IEEE Transactions on Magnetics, 1996
Abslruct-A proposal for benchmark problems to test electromagnetic optimization methods, relevant to multiobjective optimization of a Solenoidal Superconducting Magnetic Energy Storage with active and passive shielding is presented. The system has been optimized by means of different optimization procedures based on the Global Search Algorithm, Evolution Strategies, Simulated Annealing and the Conjugate Gradient Method, all coupled to integral or Finite Element codes. A comparison of results is performed and the features of the problem as a test of optimization procedures are discussed.'
Optimal Design of Nonlinear Magnetic Systems Using Finite Elements
2004
An inverse finite element method was developed to find optimal geometric parameters of a magnetic device to approximate a desired magnetic flux density distribution at certain test points and directions selected in the device. The augmented Lagrange multipliers method was utilized to transform the constrained problem consisting of a least-square objective function and a set of constraint equations to the unconstrained problem. A second-order approach based on the Lagrange-Newton method was used to minimize the unconstrained problem to improve the design iteratively. Numerical calculation of derivatives in the second-order design sensitivity analysis becomes a difficult task if saturation in material properties is accounted. A novel approach is developed to minimize the computational effort by directly combining the optimization process with the nonlinear finite element equations. The best capabilities to parametrize the device geometry and to model the nonlinear material characteris...
Constructal thermal optimization of an electromagnet
International Journal of Thermal Sciences, 2004
This paper addresses the fundamental problem of optimizing the geometry of an electromagnet by maximizing at the same time its magnetic performance and thermal performance. The solenoid has a cylindrical shape and a uniform current density. Cooling discs made of high thermal conductivity material are inserted in the coil to collect the heat generated by Joule heating. The collected heat is evacuated to the ambient. For a given magnetic performance and volume, the maximum temperature inside the electromagnet is minimized by selecting the shape of the coil (length and outer radius), the number of cooling discs, and the amount of high thermal conductivity material. The two objectives pursued in this geometric optimization (magnetic and thermal) show that the constructal design method can be extended to the generation of architecture for multi-objective systems.
Optimal design techniques applied to transverse-flux induction heating systems
IEEE Transactions on Magnetics, 1995
In this paper an optimization procedure for the optimal design of a transverse-flux induction heating system is presented. Hooke and Jeeves direct search technique is used for the optimization procedure. The finite element method is used for the electromagnetic field calculation and the integral parameter determination. The results obtained by this procedure are compared with those available in literature.
2008
In this paper we present two applications of the adjoint variable method (AVM). First we consider a design optimization problem in magnetic shielding. The objective is to reduce the magnetic stray field of an axisymmetric induction heating device for the heat treatment of aluminum discs. We involve two types of shielding, the passive and the active shielding. In the former, one needs to optimize the geometry of the passive shield. In the latter, the position of all coils and the real and imaginary components of the currents (when working in the frequency domain) must be determined. Second application involves determination of the dissipation parameter in micromagnetic model of ferromagnetism. The micromagnetic model governed by the Landau-Lifshitz equation includes the dissipation parameter α that in some cases can be a space dependent function. The actual distribution of α however can be unknown and must be determined by measurements of the magnetization in the workpiece. Using AVM method, one obtains the derivative of cost functional in terms of an adjoint variable. The main advantage is that the number of direct problem simulations needed to evaluate the derivative is independent of the number of parameters.