Differential Evolutionary Optimization of an Equivalent Dipole Model for Electromagnetic Emission Analysis

A near-field scanning based equivalent dipole model is introduced to simplify the analysis of the emission from an electronic component at its nearby receivers. The model consists of a set of infinitesimal magnetic dipoles, with the number, position, and moment of the dipoles being unknowns. A nonlinear optimization problem in terms of the unknown parameters is established through minimizing the difference between the magnitude of magnetic fields obtained by near-field scanning and the radiation from the equivalent dipoles. The problem is difficult to solve as the number of dipoles is unknown. A differential evolution method based global optimization algorithm is proposed to solve the problem. A set of subproblems is first derived. It is then solved sequentially one after the other by using the proposed algorithm. A combination of the obtained solutions forms a solution of the original optimization problem. Numerical results are presented to demonstrate the validity and efficiency of the proposed algorithm.