遗传算法在自编码棒位测量方案寻优中的应用  

Applying genetic algorithms for finding best coding of self coding rod position detection

作　　者：许明阳[1] 王文然[1] 王家英[1] 

机构地区：[1]清华大学核能技术设计研究院

出　　处：《清华大学学报（自然科学版）》1998年第12期4-8,共5页Journal of Tsinghua University(Science and Technology)

基　　金：国家"八五"科技攻关项目

摘　　要：自编码棒位测量方案是针对目前广泛采用的线圈编码电传感器线圈数量多、结构复杂等缺点而提出的新的方案。但自编码方案的寻优，尤其在测量步数较大时，将是十分困难的事；而遗传算法寻优具有效果的鲁棒性和全局优化能力。该文应用遗传算法解决反应堆控制棒位置测量系统中的自编码方案寻优问题，取得了满意的结果，从而为寻找最佳自编码方案提供了有效的方法。针对反应堆控制棒位置的问题，对标准遗传算法ＳＧＡ提出了一些实用性的改进，以提高其寻优成功率，并减少计算量，加快收敛速度。Abstract The self coding rod position detection (SCRPD) is a novel method that is put forward to overcome the shortcomings of the coil coding rod position detection. But it is difficult to find the best self coding. The genetic algorithms will be used to find the satisfactory codings for SCRPD system, and in the last the best coding is found with success. A workable way of finding self coding is put forward. To solve the problem mentioned above, some improvments are also made for the standard genetic algorithms. The efficiency of the algorithms is promoted while the quantity of computing is greatly reduced.

关 键 词：核供热堆 棒位测量 自编码 遗传算法 控制棒 

分 类 号：TL413.207[核科学技术—核技术及应用]