Parallel evolutionary design of active filter

In order to improve precision of the filter's parameters and performance indexes, an evolutionary design method of active filters was introduced by using improved adaptive parallel genetic algorithms. Selection strategy used an improved expected value model; Crossover strategy selected method which integrated one-point crossover with multiple point crossovers; Mutation strategy selected dual exclusive NOR/exclusive OR logic operators; Adaptive operators realized adaptive adjustment of probabilities of crossover and mutation according to fitness of individuals and evolutionary time; Migration strategy used parallel crossover migration strategy. The improved adaptive parallel genetic algorithm was used in optimization design of parameters and performance indexes for the four-order low-pass Chebyshev active filter. Results of the design had been analyzed and verified. Experimental results are satisfying. The method has a higher convergence speed and can solve the problem of premature effectively. The evolved filter is in line with design requests. Filter quality of the evolvable filter is high, and its ripple is smaller.