Synthesis of passive filter using object oriented genetic algorithm

Abstract

This paper describes the development of an evolutionary algorithm, the use of genetic algorithm to automatically synthesize analog circuits. The context of the project is the development of passive RLC filters of up to three components, by choosing cutoff frequency and type of filters (low pass, high pass, band pass or notch). The evaluation of each solution was performed by calculating the circuit voltages by nodal analysis for the various possible topologies without the use of simulators or programmable hardware. The proposed genetic algorithm was fully developed on object-oriented language, Java, from a class diagram that shows the relationships between population, individual (candidate circuit), chromosome (genetic representation of the circuit), selection method, crossover, mutation, evaluation of the individual (quality of the circuit), among other classes.