论文部分内容阅读
遗传算法可以有效地处理一些常规优化方法不能解决的复杂优化问题。然而,传统遗传算法存在容易陷入局部解、收敛速度慢和处理带约束优化问题效果不佳等缺点。提出了一种采用浮点数编码方法处理不等式约束优化问题的改进遗传算法(Float-encoding Genetic Algorithm,FGA),该算法具有收敛效率高、算法稳定性好和局部搜索能力强等优点。运用该算法对曲柄连杆机构进行优化设计,优化结果表明,改进后的遗传算法比传统遗传算法优化效果更好。
Genetic algorithms can effectively deal with the complex optimization problems that some conventional optimization methods can not solve. However, the traditional genetic algorithms have some disadvantages, such as easy to fall into local solution, slow convergence speed and ineffective processing constraint optimization problems. A Float-encoding Genetic Algorithm (FGA) is proposed to deal with inequality constrained optimization problems by floating-point coding. The proposed algorithm has the advantages of high convergence efficiency, good algorithm stability and strong local search ability. The algorithm is used to optimize the design of the crank and rod mechanism. The optimization results show that the improved genetic algorithm is better than the traditional genetic algorithm.