遗传算法在检测耦合腔链各单腔频率中的应用

利用遗传算法来诊断耦合腔链的各单腔频率,不需插入活塞探针进入耦合腔链,由计算机通过微波网络分析仪,从腔链的输入波导获取通带特性数据,便可求解各单腔频率.这种方法可以简化调谐过程,并可以用来诊断密封腔链.此问题可以看作是一个非线性最优化问题,我们选择遗传算法来解决此问题,并且在遗传算法中加入了最小二乘的变异算子,使遗传算法的收敛速度和精度都得到显著改善.数值实验证实了这种算法的优越性,并且与传统的遗传算法进行了比较.

Diagnosis of Individual Cell Frequencies in a Coupled Cavity Chain by Genetic Algorithms

Genetic algorithms (GAs) is used to diagnose individual call frequencies in a coupled cavity chain with no need of any probe insertion. The cell frequencies and couplings between the cells are determined in terms of the measured pass-band performance. It will simplify the tuning processes and make the tuning of the sealed cavity possible. The solution is cast as an optimization problem and GAs is used as a function optimizer to solve this problem. Moreover, in order to expedite the convergence, nonlinear least squares method is added as one kind of mutation into GAs, which makes GAs converge faster and improves the precision of the solution. The effectiveness of the new technique has been demonstrated through some numerical examples, and the results have also been compared with those of the conventional GAs.