Modified Bayesian optimization algorithm for planar array design

Design of electrical and electronic systems with complex EMC constrains requires often to exploit the peculiarities of some population based global optimizers. One of the main drawbacks of the adoption of these optimizers for system design is represented by the difficulty of introducing in the algorithm all the heuristic knowledge already available in the field. In order to overcome this problem, Bayesian optimization algorithms (BOAs), classified as estimation of distribution algorithm, can be very effective since they are based on the definition of distributions of promising solutions using the information extracted from the entire set of good solutions. Unfortunately, their straightforward implementations usually lack of exploration feature and they are easily trapped in local maxima. In order to overcome this drawback and to develop a Bayesian optimization algorithm with both exploitation and exploration mechanisms, in this paper a modified BOA is proposed by adding a suitable mutation scheme to the traditional one in order to ensure the effectiveness of the algorithm. The here proposed new algorithm has been tested on different mathematical test functions and on a typical EM design problem, a planar array synthesis to show its performance.