Nature-inspired algorithms for optimal active band pass Filter design

Asmae El Beqal (1) , Loubna Kritele (1) , Bachir Benhala (2) , Izzedine Zorkani (1)

1 USMBA - Université Sidi Mohamed Ben Abdellah
2 UMI - جامعة مولاي إسماعيل = Université Moulay Ismaïl

The basic purpose in active filters design is to calculate the filter transfer function in order to find the adequate discrete components values (Resistors and Capacitors) forming those filters, which providing the best performances. The selection of the components values must be from industrial series so as to reduce costs and make the design more feasible, then, the search for an optimal set of component values from the numerous possible combinations appears to be an exhaustive process, thus diverting the attention towards the application of the metaheuristics which combines high efficiency with minimal computational time. In the literature, various metaheuristics were used for optimal design of analog circuits such as Simulated Annealing (SA) [1], Tabu Search (TS) [2], Particle Swarm Optimization (PSO) [3], Ant Colony Optimization (ACO) [4] and Genetic Algorithms (GA) [5]. In this paper, we propose to apply the Ant Colony Optimization (ACO) [4] and the Genetic Algorithm (GA) [5] techniques, which are increasingly used in the domains of optimization and has been applied successfully in the field of analog circuits, for the optimal sizing of the second order active band pass filter namely Sallen Key (Figure 1), considering two objectives functions, the mid frequency and the selectivity factor.

Domaines: Autre

Asmae El Beqal, Loubna Kritele, Bachir Benhala, Izzedine Zorkani. Nature-inspired algorithms for optimal active band pass Filter design. International Meeting on Advanced Technologies in Energy and Electrical Engineering (IMAT3E18), Nov 2018, Fès, Morocco. ⟨hal-01959009⟩