Research of circuit evolution design based on adaptive HereBoy algorithm

Abstract

Aiming at solving the convergence rate problem in the latter stage of circuit evolution design, adaptive HereBoy algorithm together with the population evolution idea of GA is adapted to study the impact of adaptive-adjust factor on the evolution of convergence rate. One circuit model which is based on the similarities between combinatorial circuit and neural network is proposed, on which the matrix encoding scheme of combinatorial circuit is discussed. Besides of this, extrint evolution is adapted to evaluate the circuits fitness value, and adaptive-adjust factor in GA is introduced to improve the population diversity and convergence rate. The implementation of 2-bit binary multiplier circuits evolution shows that the average evolution generations and the evolution time all reduce obviously than conventional HereBoy algorithm, and that the average increase rate raises with the increase of evolution generation in the latter stage of circuit evolution.

Keywords

• electronic circuit
• EHW
• evolutionary of circuits
• neural network
• HereBoy algorithm
• adaptive algorithm