论文部分内容阅读
为了进一步降低4bit以上中小规模可逆逻辑电路的量子代价,提出一种基于子种群保留的变长染色体编码混合算法.该算法在已有变长染色体编码混合算法的基础上,将子种群保留策略用于变长编码种群的进化,以保持种群多样性,避免陷入局部最优解;定义了变长编码染色体的近似度,以此作为种子提取和子种群划分的基础;提出了子种群重新启动策略和新的启发式子种群更新操作,克服变长编码种群进化过程中的遗传漂移,从而提高可行解率和解的质量.对4bit以上常用标准可逆函数测试结果表明该算法能大大降低合成可逆电路的量子代价.
In order to further reduce the quantum cost of small and medium-sized invertible logic circuits over 4 bits, a variable-length chromosome coding hybrid algorithm based on sub-population reservation is proposed. Based on the hybrid variable-length chromosome coding algorithm, In order to maintain population diversity and avoid falling into the local optimal solution, the genetic similarity of variable length coding chromosomes was defined as the basis of seed extraction and subpopulation classification. A strategy of restarting subpopulation and New heuristic subpopulation update operation to overcome the genetic drift in the evolutionary process of variable length coding population and improve the quality of feasible solution and solution.The test results of the standard reversible function above 4bit showed that this algorithm can greatly reduce the quantum yield of the reversible circuit cost.