论文部分内容阅读
针对解决非线性系统模型变参数估计问题要求运算速度快、效率高的情况 ,提出了实数编码遗传算法的改进算法 ,即分阶段设置收敛判断条件及择优操作等操作步骤。分阶段设置收敛判断条件是指在交叉操作完成之后增加了判断是否收敛的操作步骤 ;择优操作是指用交叉操作后形成的解群中的的适应值最大的优秀个体替代适应值最小的最差个体 ,使参加变异的优秀个体数目增加 ,使得算法在优秀个体局部邻域内的搜索机会增加。通过实例对改进算法与原算法的运算性能进行了比较 ,收敛速度平均提高了 2～ 3倍。实验结果表明 ,改进算法对提高遗传算法的运算速度是可行和有效的。
In order to solve the problem of variable parameter estimation of nonlinear system model, which requires fast computing speed and high efficiency, an improved algorithm of real-coded genetic algorithm is proposed, which includes the steps of setting convergence judgment conditions and preferential operation stage by stage. The step of setting the convergence judgment condition in stages means adding the operation steps of judging whether the convergence is completed after the completion of the crossover operation. The selection operation refers to replacing the best candidate with the smallest fitness value in the solution group formed by the crossover operation with the best fitness value, Individuals make the number of outstanding individuals participating in mutation increase, which makes the search opportunities of the algorithm in local neighborhoods of excellent individuals increase. An example is given to compare the performance of the improved algorithm with that of the original algorithm. The convergence speed is improved by 2 ~ 3 times on average. Experimental results show that the improved algorithm is feasible and effective to improve the computing speed of genetic algorithm.